Small Capacity UPS
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request. I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer. I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario. Oh, and of course, being able to centrally manage them from my own iron would be great too. :-) ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com
You can buy a rectifier and batteries so it doesn't have to be a science project. Back in the day, all our large PBX installations had batteries and a rectifier. Sometimes isolated battery rooms adjacent to the switch room. There must be smaller, less expensive rectifiers. The catch is all the gear needs to support the DC power source. LaMarche has been around and was a common brand. I guess those portable, solar power banks are basically rectifiers too as long as they have some DC outputs to use. *Brandon Svec* On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG < nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks. We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Brandon Svec via NANOG" <nanog@lists.nanog.org> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Brandon Svec" <bsvec@teamonesolutions.com> Sent: Sunday, April 6, 2025 2:18:54 PM Subject: [NANOG] Re: Small Capacity UPS You can buy a rectifier and batteries so it doesn't have to be a science project. Back in the day, all our large PBX installations had batteries and a rectifier. Sometimes isolated battery rooms adjacent to the switch room. There must be smaller, less expensive rectifiers. The catch is all the gear needs to support the DC power source. LaMarche has been around and was a common brand. I guess those portable, solar power banks are basically rectifiers too as long as they have some DC outputs to use. *Brandon Svec* On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG < nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HDTWBYM4...
If you can keep everything at 48v PoE or 12v barrel plugs, the el-cheapo Amazon / eBay vendors have some mini UPS options that are in the 10-20whr range. Ubiquiti has also made some smaller UPS options for direct DC output. You also might be able to utilize the ont's integrated battery and run a parasitic load off of the battery directly. A benefit there could be dying gasp support. Eric Sent via the Samsung Galaxy S21+ 5G, an AT&T 5G smartphone Get Outlook for Android<https://aka.ms/AAb9ysg> ________________________________ From: Mike Hammett via NANOG <nanog@lists.nanog.org> Sent: Sunday, April 6, 2025 3:29:03 PM To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Mike Hammett <nanog@ics-il.net> Subject: [NANOG] Re: Small Capacity UPS Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks. We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Brandon Svec via NANOG" <nanog@lists.nanog.org> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Brandon Svec" <bsvec@teamonesolutions.com> Sent: Sunday, April 6, 2025 2:18:54 PM Subject: [NANOG] Re: Small Capacity UPS You can buy a rectifier and batteries so it doesn't have to be a science project. Back in the day, all our large PBX installations had batteries and a rectifier. Sometimes isolated battery rooms adjacent to the switch room. There must be smaller, less expensive rectifiers. The catch is all the gear needs to support the DC power source. LaMarche has been around and was a common brand. I guess those portable, solar power banks are basically rectifiers too as long as they have some DC outputs to use. *Brandon Svec* On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG < nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HDTWBYM4... _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/WCDPWL6O...
Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks.
We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing.
You should be able to put the electronics (rectifier, charge controller, RPi/NUC for mgmt ) in a fairly small project box, along with an appropriately sized battery, and have a nice, clean looking solution. If you have requirements that there aren't good existing commercial solutions for, make it yourself. How do you think all this was done in the beginning? On Sun, Apr 6, 2025 at 3:41 PM Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks.
We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing.
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
----- Original Message ----- From: "Brandon Svec via NANOG" <nanog@lists.nanog.org> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Brandon Svec" <bsvec@teamonesolutions.com> Sent: Sunday, April 6, 2025 2:18:54 PM Subject: [NANOG] Re: Small Capacity UPS
You can buy a rectifier and batteries so it doesn't have to be a science project. Back in the day, all our large PBX installations had batteries and a rectifier. Sometimes isolated battery rooms adjacent to the switch room. There must be smaller, less expensive rectifiers. The catch is all the gear needs to support the DC power source. LaMarche has been around and was a common brand. I guess those portable, solar power banks are basically rectifiers too as long as they have some DC outputs to use. *Brandon Svec*
On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG < nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HDTWBYM4...
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/WCDPWL6O...
"If you have requirements that there aren't good existing commercial solutions for, make it yourself. How do you think all this was done in the beginning? " *nods* That's how I started my WISP >20 years ago. I mean, I didn't build anything from scratch, but I was selecting mini PCI cards, adapters, antennas, etc. "in a fairly small project box" Ya know, that might really be the solution. No one (well, few) is opening their electric breakers and criticizing the electrician's work or their alarm panels or... Put the mess behind a box. Same mess, but inside a pretty box. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Tom Beecher" <beecher@beecher.cc> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Mike Hammett" <nanog@ics-il.net> Sent: Sunday, April 6, 2025 4:05:51 PM Subject: Re: [NANOG] Re: Small Capacity UPS Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks. We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing. You should be able to put the electronics (rectifier, charge controller, RPi/NUC for mgmt ) in a fairly small project box, along with an appropriately sized battery, and have a nice, clean looking solution. If you have requirements that there aren't good existing commercial solutions for, make it yourself. How do you think all this was done in the beginning? On Sun, Apr 6, 2025 at 3:41 PM Mike Hammett via NANOG < nanog@lists.nanog.org > wrote: Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks. We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Brandon Svec via NANOG" < nanog@lists.nanog.org > To: "North American Network Operators Group" < nanog@lists.nanog.org > Cc: "Brandon Svec" < bsvec@teamonesolutions.com > Sent: Sunday, April 6, 2025 2:18:54 PM Subject: [NANOG] Re: Small Capacity UPS You can buy a rectifier and batteries so it doesn't have to be a science project. Back in the day, all our large PBX installations had batteries and a rectifier. Sometimes isolated battery rooms adjacent to the switch room. There must be smaller, less expensive rectifiers. The catch is all the gear needs to support the DC power source. LaMarche has been around and was a common brand. I guess those portable, solar power banks are basically rectifiers too as long as they have some DC outputs to use. *Brandon Svec* On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG < nanog@lists.nanog.org > wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HDTWBYM4... _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/WCDPWL6O...
Also I think James' comment about efficiency is important to consider. Said you tossed a commercial UPS in there with 0% efficiency, and the draw was 100W instead of 50W. Assuming the average US electricity pricing of $0.16, that waste is costing around $70 a year.. Assuming this isn't an off-grid scenario where every W matters, is it really worth jumping through all these hoops for that? On Sun, Apr 6, 2025 at 5:12 PM Mike Hammett <nanog@ics-il.net> wrote:
"If you have requirements that there aren't good existing commercial solutions for, make it yourself. How do you think all this was done in the beginning? "
*nods* That's how I started my WISP >20 years ago. I mean, I didn't build anything from scratch, but I was selecting mini PCI cards, adapters, antennas, etc.
"in a fairly small project box"
Ya know, that might really be the solution. No one (well, few) is opening their electric breakers and criticizing the electrician's work or their alarm panels or... Put the mess behind a box. Same mess, but inside a pretty box.
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
----- Original Message ----- From: "Tom Beecher" <beecher@beecher.cc> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Mike Hammett" <nanog@ics-il.net> Sent: Sunday, April 6, 2025 4:05:51 PM Subject: Re: [NANOG] Re: Small Capacity UPS
Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks.
We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing.
You should be able to put the electronics (rectifier, charge controller, RPi/NUC for mgmt ) in a fairly small project box, along with an appropriately sized battery, and have a nice, clean looking solution.
If you have requirements that there aren't good existing commercial solutions for, make it yourself. How do you think all this was done in the beginning?
On Sun, Apr 6, 2025 at 3:41 PM Mike Hammett via NANOG < nanog@lists.nanog.org > wrote:
Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks.
We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing.
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
----- Original Message ----- From: "Brandon Svec via NANOG" < nanog@lists.nanog.org > To: "North American Network Operators Group" < nanog@lists.nanog.org > Cc: "Brandon Svec" < bsvec@teamonesolutions.com > Sent: Sunday, April 6, 2025 2:18:54 PM Subject: [NANOG] Re: Small Capacity UPS
You can buy a rectifier and batteries so it doesn't have to be a science project. Back in the day, all our large PBX installations had batteries and a rectifier. Sometimes isolated battery rooms adjacent to the switch room. There must be smaller, less expensive rectifiers. The catch is all the gear needs to support the DC power source. LaMarche has been around and was a common brand. I guess those portable, solar power banks are basically rectifiers too as long as they have some DC outputs to use. *Brandon Svec*
On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG < nanog@lists.nanog.org > wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HDTWBYM4...
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/WCDPWL6O...
Very true. I could be letting the nerd in me get in the way of just getting the job done. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Tom Beecher" <beecher@beecher.cc> To: "Mike Hammett" <nanog@ics-il.net> Cc: "North American Network Operators Group" <nanog@lists.nanog.org> Sent: Sunday, April 6, 2025 5:16:54 PM Subject: Re: [NANOG] Re: Small Capacity UPS Also I think James' comment about efficiency is important to consider. Said you tossed a commercial UPS in there with 0% efficiency, and the draw was 100W instead of 50W. Assuming the average US electricity pricing of $0.16, that waste is costing around $70 a year.. Assuming this isn't an off-grid scenario where every W matters, is it really worth jumping through all these hoops for that? On Sun, Apr 6, 2025 at 5:12 PM Mike Hammett < nanog@ics-il.net > wrote: "If you have requirements that there aren't good existing commercial solutions for, make it yourself. How do you think all this was done in the beginning? " *nods* That's how I started my WISP >20 years ago. I mean, I didn't build anything from scratch, but I was selecting mini PCI cards, adapters, antennas, etc. "in a fairly small project box" Ya know, that might really be the solution. No one (well, few) is opening their electric breakers and criticizing the electrician's work or their alarm panels or... Put the mess behind a box. Same mess, but inside a pretty box. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Tom Beecher" < beecher@beecher.cc > To: "North American Network Operators Group" < nanog@lists.nanog.org > Cc: "Mike Hammett" < nanog@ics-il.net > Sent: Sunday, April 6, 2025 4:05:51 PM Subject: Re: [NANOG] Re: Small Capacity UPS Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks. We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing. You should be able to put the electronics (rectifier, charge controller, RPi/NUC for mgmt ) in a fairly small project box, along with an appropriately sized battery, and have a nice, clean looking solution. If you have requirements that there aren't good existing commercial solutions for, make it yourself. How do you think all this was done in the beginning? On Sun, Apr 6, 2025 at 3:41 PM Mike Hammett via NANOG < nanog@lists.nanog.org > wrote: Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks. We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Brandon Svec via NANOG" < nanog@lists.nanog.org > To: "North American Network Operators Group" < nanog@lists.nanog.org > Cc: "Brandon Svec" < bsvec@teamonesolutions.com > Sent: Sunday, April 6, 2025 2:18:54 PM Subject: [NANOG] Re: Small Capacity UPS You can buy a rectifier and batteries so it doesn't have to be a science project. Back in the day, all our large PBX installations had batteries and a rectifier. Sometimes isolated battery rooms adjacent to the switch room. There must be smaller, less expensive rectifiers. The catch is all the gear needs to support the DC power source. LaMarche has been around and was a common brand. I guess those portable, solar power banks are basically rectifiers too as long as they have some DC outputs to use. *Brandon Svec* On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG < nanog@lists.nanog.org > wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HDTWBYM4... _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/WCDPWL6O...
Based on that customer description what is wrong with a small and inexpensive UPS besides the efficiency? I would (and do) offer customers like this the option for one. And if possible, sell or build into the price yearly PM visits and a three year replacement schedule. I don’t think anyone mentioned a generator yet. If the pizza/car wash/car lot would lose significant money in an outage it should be simple to explain why you suggest they spend $X on a generator and/or a UPS. Cheap insurance, basically. I doubt all their gear can be powered with DC so those options are gone. If they decline, then fine. You explained and offered a solution. Some of those businesses may actually function perfectly fine with cell service in an outage and not need anything else. This was something that happened with PBX customers when smartphones became ubiquitous. Business can continue as usual (or with small, acceptable deficits) in many cases with just a cell phone.
On Apr 6, 2025, at 12:29 PM, Mike Hammett <nanog@ics-il.net> wrote:
Sure, but I can't just drop that into a car wash, a pizza joint, a used car lot, etc. These aren't places that have battery rooms or even equipment racks.
We may look at it and think it's cool and geek out on how the different (still COTS) components were assembled, wired, etc. A layperson will just call that a mess and question if I know what I'm doing.
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
----- Original Message ----- From: "Brandon Svec via NANOG" <nanog@lists.nanog.org> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Brandon Svec" <bsvec@teamonesolutions.com> Sent: Sunday, April 6, 2025 2:18:54 PM Subject: [NANOG] Re: Small Capacity UPS
You can buy a rectifier and batteries so it doesn't have to be a science project. Back in the day, all our large PBX installations had batteries and a rectifier. Sometimes isolated battery rooms adjacent to the switch room. There must be smaller, less expensive rectifiers. The catch is all the gear needs to support the DC power source. LaMarche has been around and was a common brand. I guess those portable, solar power banks are basically rectifiers too as long as they have some DC outputs to use. *Brandon Svec*
On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG < nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HDTWBYM4...
On Sun, Apr 6, 2025 at 11:55 AM Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
If you're willing to consider a *small* science project: Buy a quality UPS in the 300VA range Make a hole in the battery compartment and extend the leads outside (minor soldering work and you can make it look professional with some heat shrink tubing). Connect an external battery that's 2-3 times the capacity. It'll charge real slow, but the charger likely has the capacity to charge the larger battery and that'll give you the desired run time. And if you make the effort to be neat about it, it should be customer deployable. Otherwise, buy a 1000VA UPS that supports external battery modules, buy as many battery modules as it takes for the runtime you want and don't sweat the efficiency. Or if you're dealing with wall wart equipment only, you could try some of these: https://www.amazon.com/s?k=12v+ups Regards, Bill Herrin -- William Herrin bill@herrin.us https://bill.herrin.us/
On 06/04/2025 19:55:20, "Mike Hammett via NANOG" <nanog@lists.nanog.org> wrote:
Most normal UPSes get terribly inefficient at lower load ratings.
You don't say how long you want to run for. Assuming the mention of extendable batteries means you are looking for several hours or more we've not found any small ones. Generally extendable batteries start with 1kVA or greater units. Due to the lack of suitable devices some UK providers have resorted to - https://www.ispreview.co.uk/index.php/2025/03/broadband-isps-vodafone-bt-kco... but it relies on feeding the devices DC. Our FWA sites are around 50W and are in cabinets so large batteries and feeding things DC is easy. We tend to power everything with POE so all that is needed is a switch with DC in (we've used Netonix but there are others such as Mikrotik). For sites where we want a packaged system delivering AC we've used lightly loaded Eaton 5PX range with extra batteries and just accepted the efficiency, which has not been terrible. There are plenty of consumer portable battery units that offer a lot more battery capacity than UPS generally have. Some operate as UPS and provide AC or DC out. If you feed your devices DC then there is no inverter low load inefficiency. For example you could run all day on just a battery expansion pack (there are larger ones) https://bluettipower.co.uk/products/b80-expansion-battery which can be charged off an AC adaptor and not need the main unit. brandon
I'd like to have a couple of options. Stuff exists for the say 1 hour of up time area, though may not have the remote management I'm looking for. It starts to get really week above that, at say 4 or 8 hours. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Brandon Butterworth" <brandon@bogons.net> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Brandon Butterworth" <brandon@bogons.net>, "Mike Hammett" <nanog@ics-il.net> Sent: Sunday, April 6, 2025 3:49:06 PM Subject: Re: [NANOG] Small Capacity UPS On 06/04/2025 19:55:20, "Mike Hammett via NANOG" <nanog@lists.nanog.org> wrote:
Most normal UPSes get terribly inefficient at lower load ratings.
You don't say how long you want to run for. Assuming the mention of extendable batteries means you are looking for several hours or more we've not found any small ones. Generally extendable batteries start with 1kVA or greater units. Due to the lack of suitable devices some UK providers have resorted to - https://www.ispreview.co.uk/index.php/2025/03/broadband-isps-vodafone-bt-kco... but it relies on feeding the devices DC. Our FWA sites are around 50W and are in cabinets so large batteries and feeding things DC is easy. We tend to power everything with POE so all that is needed is a switch with DC in (we've used Netonix but there are others such as Mikrotik). For sites where we want a packaged system delivering AC we've used lightly loaded Eaton 5PX range with extra batteries and just accepted the efficiency, which has not been terrible. There are plenty of consumer portable battery units that offer a lot more battery capacity than UPS generally have. Some operate as UPS and provide AC or DC out. If you feed your devices DC then there is no inverter low load inefficiency. For example you could run all day on just a battery expansion pack (there are larger ones) https://bluettipower.co.uk/products/b80-expansion-battery which can be charged off an AC adaptor and not need the main unit. brandon
weak* Ugh.... ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Mike Hammett via NANOG" <nanog@lists.nanog.org> To: "Brandon Butterworth" <brandon@bogons.net> Cc: "North American Network Operators Group" <nanog@lists.nanog.org>, "Mike Hammett" <nanog@ics-il.net> Sent: Sunday, April 6, 2025 3:59:06 PM Subject: [NANOG] Re: Small Capacity UPS I'd like to have a couple of options. Stuff exists for the say 1 hour of up time area, though may not have the remote management I'm looking for. It starts to get really week above that, at say 4 or 8 hours. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Brandon Butterworth" <brandon@bogons.net> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Brandon Butterworth" <brandon@bogons.net>, "Mike Hammett" <nanog@ics-il.net> Sent: Sunday, April 6, 2025 3:49:06 PM Subject: Re: [NANOG] Small Capacity UPS On 06/04/2025 19:55:20, "Mike Hammett via NANOG" <nanog@lists.nanog.org> wrote:
Most normal UPSes get terribly inefficient at lower load ratings.
You don't say how long you want to run for. Assuming the mention of extendable batteries means you are looking for several hours or more we've not found any small ones. Generally extendable batteries start with 1kVA or greater units. Due to the lack of suitable devices some UK providers have resorted to - https://www.ispreview.co.uk/index.php/2025/03/broadband-isps-vodafone-bt-kco... but it relies on feeding the devices DC. Our FWA sites are around 50W and are in cabinets so large batteries and feeding things DC is easy. We tend to power everything with POE so all that is needed is a switch with DC in (we've used Netonix but there are others such as Mikrotik). For sites where we want a packaged system delivering AC we've used lightly loaded Eaton 5PX range with extra batteries and just accepted the efficiency, which has not been terrible. There are plenty of consumer portable battery units that offer a lot more battery capacity than UPS generally have. Some operate as UPS and provide AC or DC out. If you feed your devices DC then there is no inverter low load inefficiency. For example you could run all day on just a battery expansion pack (there are larger ones) https://bluettipower.co.uk/products/b80-expansion-battery which can be charged off an AC adaptor and not need the main unit. brandon _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/GWNJPF5D...
Mike, Good question and I had the same one for a local radio station (KRYZ LPFM). They have a radio transmitter, PC, network gear and a raspberry pi to power if the grid goes down and it does as it is in California and in the Sierra's. This also had to stay up for at least 5 hours for emergency comms and not cost a crazy amount as it is a community station. What I went with was a solar charger with no solar cells and a couple of high quality LiFEPO4 batteries. Everything was off the shelf and the solar charger has an app and with a bit of wifi I can “see” the status of everything. We pull about 200 watts max and double that when charging the batteries after a power outage, the solar charger is about 90% efficient. We recently had a snow storm power went out and we were up on the air for 5 1/2 hours with no problems, even though the batteries are 6 years old! Of course the hardware we purchased is way out of date now but new ones offer the same options, we used PowMr:- https://powmr.com/collections/all-in-one-inverter-charger All in One Inverters powmr.com For batteries which seemed to be the critical high quality thing to focus on we used https://battlebornbatteries.com/ Reliable Lithium-Ion Batteries battlebornbatteries.com The net result is neat remote and supportable, both companies where super helpful.. The only thing to be super careful of when building was the size of the cables to the batteries as I had to purchase equipment to put of the cable clamps and make sure everything has the right (big) fuses. Hope that helps.. Pic below of the solution in the transmitter shed. Colin Constable 
On Apr 6, 2025, at 13:49, Brandon Butterworth via NANOG <nanog@lists.nanog.org> wrote:
On 06/04/2025 19:55:20, "Mike Hammett via NANOG" <nanog@lists.nanog.org> wrote:
Most normal UPSes get terribly inefficient at lower load ratings.
You don't say how long you want to run for. Assuming the mention of extendable batteries means you are looking for several hours or more we've not found any small ones. Generally extendable batteries start with 1kVA or greater units.
Due to the lack of suitable devices some UK providers have resorted to -
https://www.ispreview.co.uk/index.php/2025/03/broadband-isps-vodafone-bt-kco...
but it relies on feeding the devices DC.
Our FWA sites are around 50W and are in cabinets so large batteries and feeding things DC is easy. We tend to power everything with POE so all that is needed is a switch with DC in (we've used Netonix but there are others such as Mikrotik).
For sites where we want a packaged system delivering AC we've used lightly loaded Eaton 5PX range with extra batteries and just accepted the efficiency, which has not been terrible.
There are plenty of consumer portable battery units that offer a lot more battery capacity than UPS generally have. Some operate as UPS and provide AC or DC out. If you feed your devices DC then there is no inverter low load inefficiency.
For example you could run all day on just a battery expansion pack (there are larger ones) https://bluettipower.co.uk/products/b80-expansion-battery which can be charged off an AC adaptor and not need the main unit.
brandon
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/4GKCI3PQ...
Mike, I suggest that you may be using the wrong efficiency measurement. I prefer to measure Customer Service and Satisfaction for effort and dollars expended. Customers with small to medium needs as you describe could use (customer serviceable) 600VA to 1500VA units to power the loads you described. Calendar entries for battery replacement and training customer to understand alert sounds are appropriate for this customer description. Norbert Weiner recommends optimizing humans over hardware. - James R. Cutler james.cutler@consultant.com (Top poster because that is Apple Mail default, Real engineers use defaults where possible to leave time for real work.)
On Apr 6, 2025, at 2:55 PM, Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
5/12/24v lithium DC ups that powers all the devices On Sun, Apr 6, 2025, 11:56 AM Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market. Many propositions are with a special controller that emulates lead batteries for the respective voltage (it permits connecting it to the old staff). LiFePO is much-much better on a), b), c). But has bigger weight. Probably, UPS in the future will be based only on LiFePO. Then additional controller to emulate lead batteries would not be needed. It is potentially possible to upgrade any UPS (makes sense for old lead batteries): 1) cut lead batteries (sell them to any battery store, it is not about money - such garbage must be recycled), 2)connect the LiFePO with a controller that emulates lead batteries for the respective voltage. Eduard -----Original Message----- From: TJ Trout via NANOG <nanog@lists.nanog.org> Sent: Monday, April 7, 2025 00:27 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: TJ Trout <tj@pcguys.us> Subject: [NANOG] Re: Small Capacity UPS 5/12/24v lithium DC ups that powers all the devices On Sun, Apr 6, 2025, 11:56 AM Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P 2J2YK3S6257XKFJ54NWABALF62DACL/
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5IDSJPIL...
Agree, LFP is the way to go. I did that in many types of UPS, different models & capacity, such as EATON and APC, but many others as well. There are many suppliers, even in Amazon, Alibaba, Aliexpress, etc. that already have LFP cells that internally have the BMS and work in a transparent way to replace lead-acid, using same connectors and dimensions (but much lower weight). They are more expensive (3-4 times the price of the lead-acid ones), but they will last for 15-20 year, and they can be discharged up to 90% instead of 50%. They often provide a bit “extra" real power/capacity as well. Regards, Jordi @jordipalet
El 7 abr 2025, a las 8:30, Vasilenko Eduard via NANOG <nanog@lists.nanog.org> escribió:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market. Many propositions are with a special controller that emulates lead batteries for the respective voltage (it permits connecting it to the old staff). LiFePO is much-much better on a), b), c). But has bigger weight.
Probably, UPS in the future will be based only on LiFePO. Then additional controller to emulate lead batteries would not be needed. It is potentially possible to upgrade any UPS (makes sense for old lead batteries): 1) cut lead batteries (sell them to any battery store, it is not about money - such garbage must be recycled), 2)connect the LiFePO with a controller that emulates lead batteries for the respective voltage. Eduard -----Original Message----- From: TJ Trout via NANOG <nanog@lists.nanog.org> Sent: Monday, April 7, 2025 00:27 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: TJ Trout <tj@pcguys.us> Subject: [NANOG] Re: Small Capacity UPS
5/12/24v lithium DC ups that powers all the devices
On Sun, Apr 6, 2025, 11:56 AM Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P 2J2YK3S6257XKFJ54NWABALF62DACL/
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5IDSJPIL... _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/2WV7TORN...
********************************************** IPv4 is over Are you ready for the new Internet ? http://www.theipv6company.com The IPv6 Company This electronic message contains information which may be privileged or confidential. The information is intended to be for the exclusive use of the individual(s) named above and further non-explicilty authorized disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited and will be considered a criminal offense. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited, will be considered a criminal offense, so you must reply to the original sender to inform about this communication and delete it.
On 4/7/25 09:02, jordi.palet--- via NANOG wrote:
Agree, LFP is the way to go. I did that in many types of UPS, different models & capacity, such as EATON and APC, but many others as well.
There are many suppliers, even in Amazon, Alibaba, Aliexpress, etc. that already have LFP cells that internally have the BMS and work in a transparent way to replace lead-acid, using same connectors and dimensions (but much lower weight). They are more expensive (3-4 times the price of the lead-acid ones), but they will last for 15-20 year, and they can be discharged up to 90% instead of 50%. They often provide a bit “extra" real power/capacity as well.
If you are going to build your own Li-Ion battery, a BMS is a must. But if you buy a pre-built Li-Ion battery, it will always ship with a BMS. In fact, many of the 12V Li-Ion batteries may be purchased with a Bluetooth controller option, which gives you real-time data of the BMS's performance via a phone app. I wouldn't say Li-Ion batteries are 3 - 4 times the cost of LA. At least in our market, you're looking at 2X, for just a pre-built battery with a BMS. If you want a larger system (24V or 48V), then yes, the extra controls, protections, build structure, e.t.c. can contribute to a 4X price difference. The only issue with using 12V Li-Ion batteries in a UPS is that most UPS's are 24VDC systems. That means you'll need 2x 12V Li-Ion batteries... the problem being that you'd have to balance them, and external balancers don't fit well inside small UPS chassis. Mark.
Hi Mark, Yes, in fact I did that. I used a BMS to build my “home” battery with 90 LFP cells for a total of 80Kwh. I didn’t meant for a UPS replacement for lead-acid you should add a BMS, because they already have an small internal BMS, as they have inside multiple LFP cells, and yes, they also have choices with bluetooth. In fact, this is the way I replaced my car lead-acid battery, which previously was dying every 2-3 years. It is true that if your UPS has multiple lead-acid cells, because they most of the time are in series, they will not be automatically balanced. What I do in those cases is fully charge each cell before using them. If they are new and grade “A”, normally, unless there is faulty cell, they will remain balanced across multiple years. In fact, I tested that with an APC UPS with hast 4 20Ah lead-acid cells, and 4 external ones. Replaced with LFP ones, and after a couple of years, they were still balanced. Checked again after 2 more years, and still are balanced. Of course, if they have also bluetooth, you will be able to check that without removing them, as it allows to see the SoC of each individual cell (voltage is not good enough with LFP). Regarding price it may depend on each market, of course. Here a 8Ah 12V small cell (the one mostly used for UPSs up to 1600VA or so) cost 11-12 Euros, while the LFP one is around 38 (without bluetooth). Those prices aren’t including VAT (21% extra). Regards, Jordi @jordipalet
El 7 abr 2025, a las 11:38, Mark Tinka <mark@tinka.africa> escribió:
On 4/7/25 09:02, jordi.palet--- via NANOG wrote:
Agree, LFP is the way to go. I did that in many types of UPS, different models & capacity, such as EATON and APC, but many others as well.
There are many suppliers, even in Amazon, Alibaba, Aliexpress, etc. that already have LFP cells that internally have the BMS and work in a transparent way to replace lead-acid, using same connectors and dimensions (but much lower weight). They are more expensive (3-4 times the price of the lead-acid ones), but they will last for 15-20 year, and they can be discharged up to 90% instead of 50%. They often provide a bit “extra" real power/capacity as well.
If you are going to build your own Li-Ion battery, a BMS is a must.
But if you buy a pre-built Li-Ion battery, it will always ship with a BMS.
In fact, many of the 12V Li-Ion batteries may be purchased with a Bluetooth controller option, which gives you real-time data of the BMS's performance via a phone app.
I wouldn't say Li-Ion batteries are 3 - 4 times the cost of LA. At least in our market, you're looking at 2X, for just a pre-built battery with a BMS. If you want a larger system (24V or 48V), then yes, the extra controls, protections, build structure, e.t.c. can contribute to a 4X price difference.
The only issue with using 12V Li-Ion batteries in a UPS is that most UPS's are 24VDC systems. That means you'll need 2x 12V Li-Ion batteries... the problem being that you'd have to balance them, and external balancers don't fit well inside small UPS chassis.
Mark.
********************************************** IPv4 is over Are you ready for the new Internet ? http://www.theipv6company.com The IPv6 Company This electronic message contains information which may be privileged or confidential. The information is intended to be for the exclusive use of the individual(s) named above and further non-explicilty authorized disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited and will be considered a criminal offense. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited, will be considered a criminal offense, so you must reply to the original sender to inform about this communication and delete it.
On 4/7/25 11:55, jordi.palet--- via NANOG wrote:
I didn’t meant for a UPS replacement for lead-acid you should add a BMS, because they already have an small internal BMS, as they have inside multiple LFP cells, and yes, they also have choices with bluetooth. In fact, this is the way I replaced my car lead-acid battery, which previously was dying every 2-3 years.
While you can buy UPS's with Li-Ion cells, most UPS's are sold with LA batteries. Of course, a Li-Ion-based UPS will come with a BMS and balancer included. Not such a big issue for a car or any other application that requires only one 12V battery (like a gate or garage motor), since no balancing is required. You just need to buy the right 12V battery that has sufficient capacity to handle a motor engine crank. Personally, even though LFP is safe with a relatively high thermal runaway temperature threshold, I continue to use LA for my cars, even in cars where the battery is not located in the engine bay.
It is true that if your UPS has multiple lead-acid cells, because they most of the time are in series, they will not be automatically balanced. What I do in those cases is fully charge each cell before using them. If they are new and grade “A”, normally, unless there is faulty cell, they will remain balanced across multiple years. In fact, I tested that with an APC UPS with hast 4 20Ah lead-acid cells, and 4 external ones. Replaced with LFP ones, and after a couple of years, they were still balanced. Checked again after 2 more years, and still are balanced. Of course, if they have also bluetooth, you will be able to check that without removing them, as it allows to see the SoC of each individual cell (voltage is not good enough with LFP).
Voltage drift is much less of an issue with LA than with Li-Ion. True, including a balancer for your LA batteries is nice, but not terribly critical as compared to Li-Ion. Li-Ion voltage drift can be a real problem if you have a large pack built in parallel to maximize capacity. If one of the batteries is running at a lower voltage, it will shut your entire system off even if the rest of the batteries still have enough charge in them. For large systems, you'd be running some kind of CAN bus between each of the batteries in the pack to communicate voltage and SoC info, so that the entire pack can balance itself. Another issue that can develop with poor balancing is chemical and digital drift, where the calculated SoC is different from what the actual chemical voltage is. The impact here is a potentially incorrect SoH (State of Health), which may lead you to thinking your battery is aging faster than it really is. Mark.
On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium. Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate). NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties. LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging. Mark.
Thanks to many people for good references! I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice. Why has the industry stopped innovating? I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves? I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP. I did assume that any vendor not supporting LFP our days should go out of the market very fast. It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way. It may be some patent, but I doubt that a patent could completely block innovations. Eduard -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Monday, April 7, 2025 12:32 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium. Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate). NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties. LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging. Mark.
a quick search I found this. never heard of this company but looks decent. https://goldenmateenergy.com/products/goldenmate-8-outlets-ups-1000va-600w-ups-lifepo4-battery-backup-and-surge-protector?variant=51304496103734&country=US¤cy=USD&utm_medium=product_sync&utm_source=google&utm_content=sag_organic&utm_campaign=sag_organic&https%3A%2F%2Fgoldenmateenergy.com%2F%3Futm_source=google&utm_medium=cpc&utm_campaign=pla1209&utm_id=22004982719&gad_source=1&gclid=CjwKCAjwtdi_BhACEiwA97y8BNMHwkx_E-TokD2ZGSNg_oMHW_y56tNF6xRFfRCmeF8cv1YjOS5F-BoCLbgQAvD_BwE On Wed, Apr 9, 2025, 2:32 AM Vasilenko Eduard via NANOG < nanog@lists.nanog.org> wrote:
Thanks to many people for good references!
I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice.
Why has the industry stopped innovating? I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves? I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP. I did assume that any vendor not supporting LFP our days should go out of the market very fast. It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way.
It may be some patent, but I doubt that a patent could completely block innovations.
Eduard -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Monday, April 7, 2025 12:32 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: Re: [NANOG] Re: Small Capacity UPS
On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium.
Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate).
NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties.
LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5AHCONMM...
Thanks. I am more interested in the bigger one that has a “pre-order” warning: 2000VA/1600W LiFePO4 UPS Power Battery Backup & Surge Protector<https://goldenmateenergy.com/products/goldenmate-back-ups-8-outlets-2000va-1600w-lifepo4-battery-backup-and-surge-protector> Ed/ From: Javier J <javier@advancedmachines.us> Sent: Wednesday, April 9, 2025 10:50 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Mark Tinka <mark@tinka.africa>; Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: Re: [NANOG] Re: Small Capacity UPS a quick search I found this. never heard of this company but looks decent. https://goldenmateenergy.com/products/goldenmate-8-outlets-ups-1000va-600w-ups-lifepo4-battery-backup-and-surge-protector?variant=51304496103734&country=US¤cy=USD&utm_medium=product_sync&utm_source=google&utm_content=sag_organic&utm_campaign=sag_organic&https%3A%2F%2Fgoldenmateenergy.com%2F%3Futm_source=google&utm_medium=cpc&utm_campaign=pla1209&utm_id=22004982719&gad_source=1&gclid=CjwKCAjwtdi_BhACEiwA97y8BNMHwkx_E-TokD2ZGSNg_oMHW_y56tNF6xRFfRCmeF8cv1YjOS5F-BoCLbgQAvD_BwE On Wed, Apr 9, 2025, 2:32 AM Vasilenko Eduard via NANOG <nanog@lists.nanog.org<mailto:nanog@lists.nanog.org>> wrote: Thanks to many people for good references! I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice. Why has the industry stopped innovating? I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves? I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP. I did assume that any vendor not supporting LFP our days should go out of the market very fast. It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way. It may be some patent, but I doubt that a patent could completely block innovations. Eduard -----Original Message----- From: Mark Tinka <mark@tinka.africa<mailto:mark@tinka.africa>> Sent: Monday, April 7, 2025 12:32 To: North American Network Operators Group <nanog@lists.nanog.org<mailto:nanog@lists.nanog.org>> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com<mailto:vasilenko.eduard@huawei.com>> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium. Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate). NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties. LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging. Mark. _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5AHCONMM...
I had seen those. I got excited (for other projects) until I saw that it had no communications capability. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest-IX http://www.midwest-ix.com ----- Original Message ----- From: "Javier J via NANOG" <nanog@lists.nanog.org> To: "North American Network Operators Group" <nanog@lists.nanog.org> Cc: "Javier J" <javier@advancedmachines.us> Sent: Wednesday, April 9, 2025 2:49:58 AM Subject: [NANOG] Re: Small Capacity UPS a quick search I found this. never heard of this company but looks decent. https://goldenmateenergy.com/products/goldenmate-8-outlets-ups-1000va-600w-ups-lifepo4-battery-backup-and-surge-protector?variant=51304496103734&country=US¤cy=USD&utm_medium=product_sync&utm_source=google&utm_content=sag_organic&utm_campaign=sag_organic&https%3A%2F%2Fgoldenmateenergy.com%2F%3Futm_source=google&utm_medium=cpc&utm_campaign=pla1209&utm_id=22004982719&gad_source=1&gclid=CjwKCAjwtdi_BhACEiwA97y8BNMHwkx_E-TokD2ZGSNg_oMHW_y56tNF6xRFfRCmeF8cv1YjOS5F-BoCLbgQAvD_BwE On Wed, Apr 9, 2025, 2:32 AM Vasilenko Eduard via NANOG < nanog@lists.nanog.org> wrote:
Thanks to many people for good references!
I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice.
Why has the industry stopped innovating? I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves? I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP. I did assume that any vendor not supporting LFP our days should go out of the market very fast. It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way.
It may be some patent, but I doubt that a patent could completely block innovations.
Eduard -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Monday, April 7, 2025 12:32 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: Re: [NANOG] Re: Small Capacity UPS
On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium.
Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate).
NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties.
LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5AHCONMM...
NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/QJX3N3VT...
There are some of the big boys who are doing LFP - Eaton does so for example - a whole line of them.. https://www.eaton.com/us/en-us/products/backup-power-ups-surge-it-power-dist... Vertiv/Leibert - https://www.vertiv.com/en-us/solutions/learn-about/vertiv-lithium-ion-ups/ APC - https://www.apc.com/us/en/campaign/lithium-ion-ups.jsp Even cyberpower - https://www.cyberpowersystems.com/promotions/smart-app-sinewave-lithium-ups/ I recall shopping lithium UPSes over many, many years - some that weren't LFP at all, but even then, much higher density than LFP can provide. Also big name manufacturers at the times. I'm thinking back to say, 2015 maybe? NMC as mentioned, and yes, even in lipo types as well. The innovation is there, if you can afford it - it's still more expensive than SLA just on virtue of the cell costs, alone. Space savings may or may not be worth it to you, longevity argument may win the budget fight, it may not - sometimes spending $X every 2-3 years is a better option for cashflow reasons than spending $2*X for 5-8 years. Then again.... LFP support is a given in almost any UPS since you can have drop-in replacement cells. If it's a lead-acid UPS, it's a LFP UPS. No need to hack on a BMS unless you're building the entire UPS yourself - the drop-in replacement cells have them built in. LFP UPSes from the mainstreams like above may have the BMS system integrated into something other than the cells, but likely it's part of the entire battery module instead and swapped when you swap packs - so the fundamental UPS itself remains the same barring metering/reporting circuitry to talk to the BMS. Just swapping out my rackmount tripplites was 2x the cost for LFP cells to replace the SLA cells, and I only went from 9Ah per cell to 10Ah. Though, justifiably, I did not have to swap UPSes and should not have to swap batteries for another 8-10 years, plus a slightly extended runtime. So my ROI is approximately 4-5 years. -----Original Message----- From: Vasilenko Eduard via NANOG <nanog@lists.nanog.org> Sent: Wednesday, April 9, 2025 2:31 AM To: Mark Tinka <mark@tinka.africa>; North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: [NANOG] Re: Small Capacity UPS Thanks to many people for good references! I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice. Why has the industry stopped innovating? I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves? I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP. I did assume that any vendor not supporting LFP our days should go out of the market very fast. It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way. It may be some patent, but I doubt that a patent could completely block innovations. Eduard -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Monday, April 7, 2025 12:32 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium. Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate). NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties. LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging. Mark. _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5AHCONMM...
Your first 3 UPSes are lithium-ion. I am scared of lithium-ion in the apartment. Your last URL is for LFP. Thanks! But it is for 110V (I need 220V), and 3kVA is too small for my needs. Looks like I need to search again. I would not agree that dependency on the quality of an additional BMS is a good thing. And additional BMS should cost additional money that could be big enough for 5kVA. It is for sure not an optimal solution. Eduard -----Original Message----- From: Gary Sparkes <gary@kisaracorporation.com> Sent: Wednesday, April 9, 2025 11:27 To: North American Network Operators Group <nanog@lists.nanog.org>; Mark Tinka <mark@tinka.africa> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: RE: [NANOG] Re: Small Capacity UPS There are some of the big boys who are doing LFP - Eaton does so for example - a whole line of them.. https://www.eaton.com/us/en-us/products/backup-power-ups-surge-it-power-dist... Vertiv/Leibert - https://www.vertiv.com/en-us/solutions/learn-about/vertiv-lithium-ion-ups/ APC - https://www.apc.com/us/en/campaign/lithium-ion-ups.jsp Even cyberpower - https://www.cyberpowersystems.com/promotions/smart-app-sinewave-lithium-ups/ I recall shopping lithium UPSes over many, many years - some that weren't LFP at all, but even then, much higher density than LFP can provide. Also big name manufacturers at the times. I'm thinking back to say, 2015 maybe? NMC as mentioned, and yes, even in lipo types as well. The innovation is there, if you can afford it - it's still more expensive than SLA just on virtue of the cell costs, alone. Space savings may or may not be worth it to you, longevity argument may win the budget fight, it may not - sometimes spending $X every 2-3 years is a better option for cashflow reasons than spending $2*X for 5-8 years. Then again.... LFP support is a given in almost any UPS since you can have drop-in replacement cells. If it's a lead-acid UPS, it's a LFP UPS. No need to hack on a BMS unless you're building the entire UPS yourself - the drop-in replacement cells have them built in. LFP UPSes from the mainstreams like above may have the BMS system integrated into something other than the cells, but likely it's part of the entire battery module instead and swapped when you swap packs - so the fundamental UPS itself remains the same barring metering/reporting circuitry to talk to the BMS. Just swapping out my rackmount tripplites was 2x the cost for LFP cells to replace the SLA cells, and I only went from 9Ah per cell to 10Ah. Though, justifiably, I did not have to swap UPSes and should not have to swap batteries for another 8-10 years, plus a slightly extended runtime. So my ROI is approximately 4-5 years. -----Original Message----- From: Vasilenko Eduard via NANOG <nanog@lists.nanog.org> Sent: Wednesday, April 9, 2025 2:31 AM To: Mark Tinka <mark@tinka.africa>; North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: [NANOG] Re: Small Capacity UPS Thanks to many people for good references! I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice. Why has the industry stopped innovating? I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves? I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP. I did assume that any vendor not supporting LFP our days should go out of the market very fast. It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way. It may be some patent, but I doubt that a patent could completely block innovations. Eduard -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Monday, April 7, 2025 12:32 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium. Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate). NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties. LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging. Mark. _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5AHCONMM...
So, they're all selling LFP units. LFP is a type of lithium-ion battery. I could (and do) call a LiFePo4 cell a lithium-ion battery and be 100% correct - because it is. LiPo's, NMC/PO4 (be it in 18650 or other formfactor, etc), LFP, you name it, they're all lithium-ion batteries because that's the mechanism they operate by. They're just different chemistries and formfactors of lithium-ion batteries. https://www.eaton.com/us/en-us/products/backup-power-ups-surge-it-power-dist... This liebert information sheet (second link of products after eaton in my list) - https://www.vertiv.com/491005/globalassets/products/critical-power/uninterru... - from this information sheet - "Safe and Reliable: Both VRLA and Li-ion ensure safety and reliability. Lithium-Ion (LiFePO4) battery chemistry is safest and integrated battery management system (BMS) also ensures safe and reliable operation." The few APC MSDS (Materials Safety Data Sheet) sheets I glanced at indicate Lithium Iron Phosphate as the cell ingredient as well - which means LFP chemistry. So they're shipping LiFePo4 cells as well. Weird they didn't make it (easier) to find that, though. https://www.apc.com/us/en/faqs/FA158828/ - I looked at XBP48RM2U-LI_NAM_EN_SDSV1.0.pdf and SMTL750_1000_1500_LI_EU_EN_SDS_V2.0 (1).pdf That second sheet shows that https://www.apc.com/us/en/product/SMTL750RM2UC/apc-smartups-line-interactive-750va-lithiumion-rack-tower-2u-120v-6x-nema-515r-outlets-smartconnect-port+smartslot-short-depth-avr-lcd/?range=61915-smartups&parent-subcategory-id=88976&selectedNodeId=23679172486 is a LFP based UPS system. Though not all models - you can see from APCRBC173-LI_NAM_EN_SDS_V1.3.pdf that specific pack is NMC. So you'd just have to investigate each model to see what it is. So yes, all four are selling LFP units, though APC seems to be a bit harder to pick out which is which, not that it really matters much anymore from a safety perspective, but if it concerns you, the information IS available. I'd feel perfectly comfortable with any of them technology wise these days (and in the past, too) though. -----Original Message----- From: Vasilenko Eduard <vasilenko.eduard@huawei.com> Sent: Wednesday, April 9, 2025 4:57 AM To: Gary Sparkes <gary@kisaracorporation.com>; North American Network Operators Group <nanog@lists.nanog.org>; Mark Tinka <mark@tinka.africa> Subject: RE: [NANOG] Re: Small Capacity UPS Your first 3 UPSes are lithium-ion. I am scared of lithium-ion in the apartment. Your last URL is for LFP. Thanks! But it is for 110V (I need 220V), and 3kVA is too small for my needs. Looks like I need to search again. I would not agree that dependency on the quality of an additional BMS is a good thing. And additional BMS should cost additional money that could be big enough for 5kVA. It is for sure not an optimal solution. Eduard -----Original Message----- From: Gary Sparkes <gary@kisaracorporation.com> Sent: Wednesday, April 9, 2025 11:27 To: North American Network Operators Group <nanog@lists.nanog.org>; Mark Tinka <mark@tinka.africa> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: RE: [NANOG] Re: Small Capacity UPS There are some of the big boys who are doing LFP - Eaton does so for example - a whole line of them.. https://www.eaton.com/us/en-us/products/backup-power-ups-surge-it-power-dist... Vertiv/Leibert - https://www.vertiv.com/en-us/solutions/learn-about/vertiv-lithium-ion-ups/ APC - https://www.apc.com/us/en/campaign/lithium-ion-ups.jsp Even cyberpower - https://www.cyberpowersystems.com/promotions/smart-app-sinewave-lithium-ups/ I recall shopping lithium UPSes over many, many years - some that weren't LFP at all, but even then, much higher density than LFP can provide. Also big name manufacturers at the times. I'm thinking back to say, 2015 maybe? NMC as mentioned, and yes, even in lipo types as well. The innovation is there, if you can afford it - it's still more expensive than SLA just on virtue of the cell costs, alone. Space savings may or may not be worth it to you, longevity argument may win the budget fight, it may not - sometimes spending $X every 2-3 years is a better option for cashflow reasons than spending $2*X for 5-8 years. Then again.... LFP support is a given in almost any UPS since you can have drop-in replacement cells. If it's a lead-acid UPS, it's a LFP UPS. No need to hack on a BMS unless you're building the entire UPS yourself - the drop-in replacement cells have them built in. LFP UPSes from the mainstreams like above may have the BMS system integrated into something other than the cells, but likely it's part of the entire battery module instead and swapped when you swap packs - so the fundamental UPS itself remains the same barring metering/reporting circuitry to talk to the BMS. Just swapping out my rackmount tripplites was 2x the cost for LFP cells to replace the SLA cells, and I only went from 9Ah per cell to 10Ah. Though, justifiably, I did not have to swap UPSes and should not have to swap batteries for another 8-10 years, plus a slightly extended runtime. So my ROI is approximately 4-5 years. -----Original Message----- From: Vasilenko Eduard via NANOG <nanog@lists.nanog.org> Sent: Wednesday, April 9, 2025 2:31 AM To: Mark Tinka <mark@tinka.africa>; North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: [NANOG] Re: Small Capacity UPS Thanks to many people for good references! I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice. Why has the industry stopped innovating? I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves? I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP. I did assume that any vendor not supporting LFP our days should go out of the market very fast. It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way. It may be some patent, but I doubt that a patent could completely block innovations. Eduard -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Monday, April 7, 2025 12:32 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium. Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate). NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties. LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging. Mark. _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5AHCONMM...
Just reread your note about 5KVA and 220 again, Eaton LFP 1.5 to 6KVA available in one model line here with capabilities you could need - https://www.eaton.com/content/dam/eaton/products/backup-power-ups-surge-it-p... - though i'm sure they cost a lot..... but that's just one example of many. As to "additional" BMS, drop-in replacement cells have a built in BMS - you just must make sure the drop-in replacement is rated for the load it may have to supply. You just don't get as much 'intelligent' data out of it, since the BMS is inside the case with the cells, so all you have "access" to is voltage at the terminals, usually, just like a standard SLA. I speak of units like these: https://dakotalithium.com/product/dakota-lithium-12v-10ah-battery/ where you would use them in most any capacity you would a standard SLA battery without changing chargers, equipment, or anything else external, since the BMS is integral to the battery. The important thing to note about those, though - is the ratings. "10A max, 14.4V recommended" for charging and "20A max continuous, 50A max 300 mS pulse." for discharge, in this example. So when shopping for that kind of cell, those are the maximum charge/discharge rates of the integral BMS. When doing swap in replacements, ensuring that you are within those specs is important. Bringing your own BMS comes into play when you start buying individual cells, such as these and assembling your own battery banks as I tend to do - https://imrbatteries.com/products/eve-lf105-105ah-3-2v-lifepo4-prismatic-bat... - because that is just a bare 3.2v LFP cell, no protections of any kind, for integrating into your own product/system. Four of those and a good BMS would give you a 105Ah replacement for a 12V battery in approximately half the space. I assemble them into multi-power-output systems for camping and solar equipment. -----Original Message----- From: Gary Sparkes via NANOG <nanog@lists.nanog.org> Sent: Wednesday, April 9, 2025 5:21 AM To: Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org>; Mark Tinka <mark@tinka.africa> Cc: Gary Sparkes <gary@kisaracorporation.com> Subject: [NANOG] Re: Small Capacity UPS So, they're all selling LFP units. LFP is a type of lithium-ion battery. I could (and do) call a LiFePo4 cell a lithium-ion battery and be 100% correct - because it is. LiPo's, NMC/PO4 (be it in 18650 or other formfactor, etc), LFP, you name it, they're all lithium-ion batteries because that's the mechanism they operate by. They're just different chemistries and formfactors of lithium-ion batteries. https://www.eaton.com/us/en-us/products/backup-power-ups-surge-it-power-dist... This liebert information sheet (second link of products after eaton in my list) - https://www.vertiv.com/491005/globalassets/products/critical-power/uninterru... - from this information sheet - "Safe and Reliable: Both VRLA and Li-ion ensure safety and reliability. Lithium-Ion (LiFePO4) battery chemistry is safest and integrated battery management system (BMS) also ensures safe and reliable operation." The few APC MSDS (Materials Safety Data Sheet) sheets I glanced at indicate Lithium Iron Phosphate as the cell ingredient as well - which means LFP chemistry. So they're shipping LiFePo4 cells as well. Weird they didn't make it (easier) to find that, though. https://www.apc.com/us/en/faqs/FA158828/ - I looked at XBP48RM2U-LI_NAM_EN_SDSV1.0.pdf and SMTL750_1000_1500_LI_EU_EN_SDS_V2.0 (1).pdf That second sheet shows that https://www.apc.com/us/en/product/SMTL750RM2UC/apc-smartups-line-interactive-750va-lithiumion-rack-tower-2u-120v-6x-nema-515r-outlets-smartconnect-port+smartslot-short-depth-avr-lcd/?range=61915-smartups&parent-subcategory-id=88976&selectedNodeId=23679172486 is a LFP based UPS system. Though not all models - you can see from APCRBC173-LI_NAM_EN_SDS_V1.3.pdf that specific pack is NMC. So you'd just have to investigate each model to see what it is. So yes, all four are selling LFP units, though APC seems to be a bit harder to pick out which is which, not that it really matters much anymore from a safety perspective, but if it concerns you, the information IS available. I'd feel perfectly comfortable with any of them technology wise these days (and in the past, too) though. -----Original Message----- From: Vasilenko Eduard <vasilenko.eduard@huawei.com> Sent: Wednesday, April 9, 2025 4:57 AM To: Gary Sparkes <gary@kisaracorporation.com>; North American Network Operators Group <nanog@lists.nanog.org>; Mark Tinka <mark@tinka.africa> Subject: RE: [NANOG] Re: Small Capacity UPS Your first 3 UPSes are lithium-ion. I am scared of lithium-ion in the apartment. Your last URL is for LFP. Thanks! But it is for 110V (I need 220V), and 3kVA is too small for my needs. Looks like I need to search again. I would not agree that dependency on the quality of an additional BMS is a good thing. And additional BMS should cost additional money that could be big enough for 5kVA. It is for sure not an optimal solution. Eduard -----Original Message----- From: Gary Sparkes <gary@kisaracorporation.com> Sent: Wednesday, April 9, 2025 11:27 To: North American Network Operators Group <nanog@lists.nanog.org>; Mark Tinka <mark@tinka.africa> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: RE: [NANOG] Re: Small Capacity UPS There are some of the big boys who are doing LFP - Eaton does so for example - a whole line of them.. https://www.eaton.com/us/en-us/products/backup-power-ups-surge-it-power-dist... Vertiv/Leibert - https://www.vertiv.com/en-us/solutions/learn-about/vertiv-lithium-ion-ups/ APC - https://www.apc.com/us/en/campaign/lithium-ion-ups.jsp Even cyberpower - https://www.cyberpowersystems.com/promotions/smart-app-sinewave-lithium-ups/ I recall shopping lithium UPSes over many, many years - some that weren't LFP at all, but even then, much higher density than LFP can provide. Also big name manufacturers at the times. I'm thinking back to say, 2015 maybe? NMC as mentioned, and yes, even in lipo types as well. The innovation is there, if you can afford it - it's still more expensive than SLA just on virtue of the cell costs, alone. Space savings may or may not be worth it to you, longevity argument may win the budget fight, it may not - sometimes spending $X every 2-3 years is a better option for cashflow reasons than spending $2*X for 5-8 years. Then again.... LFP support is a given in almost any UPS since you can have drop-in replacement cells. If it's a lead-acid UPS, it's a LFP UPS. No need to hack on a BMS unless you're building the entire UPS yourself - the drop-in replacement cells have them built in. LFP UPSes from the mainstreams like above may have the BMS system integrated into something other than the cells, but likely it's part of the entire battery module instead and swapped when you swap packs - so the fundamental UPS itself remains the same barring metering/reporting circuitry to talk to the BMS. Just swapping out my rackmount tripplites was 2x the cost for LFP cells to replace the SLA cells, and I only went from 9Ah per cell to 10Ah. Though, justifiably, I did not have to swap UPSes and should not have to swap batteries for another 8-10 years, plus a slightly extended runtime. So my ROI is approximately 4-5 years. -----Original Message----- From: Vasilenko Eduard via NANOG <nanog@lists.nanog.org> Sent: Wednesday, April 9, 2025 2:31 AM To: Mark Tinka <mark@tinka.africa>; North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: [NANOG] Re: Small Capacity UPS Thanks to many people for good references! I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice. Why has the industry stopped innovating? I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves? I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP. I did assume that any vendor not supporting LFP our days should go out of the market very fast. It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way. It may be some patent, but I doubt that a patent could completely block innovations. Eduard -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Monday, April 7, 2025 12:32 To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Vasilenko Eduard <vasilenko.eduard@huawei.com> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/7/25 08:30, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market.
Ummh, the Li in LiFePO4 (a.k.a LFP) is Lithium. Li-Ion batteries come in all manner of chemistries, where different materials may be chosen for the cathode. While there are a number of them, the most common ones are NMC (Nickel Manganese Cobalt) and PO4 (Phosphate). NMC has generally been used in EV's due to its high energy density, but OEM's have now started moving to LFP for this, where LFP was traditionally used for stationery applications (home backup, for example) due its higher safety properties. LFP is cheaper than NMC. And while NMC will provide fewer cycles than LFP, the time period will vary based on usage, even with factoring in calendar aging. Mark. _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5AHCONMM... _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/VU42GFWI...
On 4/9/25 11:21, Gary Sparkes wrote:
So, they're all selling LFP units.
LFP is a type of lithium-ion battery. I could (and do) call a LiFePo4 cell a lithium-ion battery and be 100% correct - because it is. LiPo's, NMC/PO4 (be it in 18650 or other formfactor, etc), LFP, you name it, they're all lithium-ion batteries because that's the mechanism they operate by. They're just different chemistries and formfactors of lithium-ion batteries.
https://www.eaton.com/us/en-us/products/backup-power-ups-surge-it-power-dist...
This liebert information sheet (second link of products after eaton in my list) - https://www.vertiv.com/491005/globalassets/products/critical-power/uninterru... - from this information sheet - "Safe and Reliable: Both VRLA and Li-ion ensure safety and reliability. Lithium-Ion (LiFePO4) battery chemistry is safest and integrated battery management system (BMS) also ensures safe and reliable operation."
The few APC MSDS (Materials Safety Data Sheet) sheets I glanced at indicate Lithium Iron Phosphate as the cell ingredient as well - which means LFP chemistry. So they're shipping LiFePo4 cells as well. Weird they didn't make it (easier) to find that, though. https://www.apc.com/us/en/faqs/FA158828/ - I looked at XBP48RM2U-LI_NAM_EN_SDSV1.0.pdf and SMTL750_1000_1500_LI_EU_EN_SDS_V2.0 (1).pdf
That second sheet shows that https://www.apc.com/us/en/product/SMTL750RM2UC/apc-smartups-line-interactive-750va-lithiumion-rack-tower-2u-120v-6x-nema-515r-outlets-smartconnect-port+smartslot-short-depth-avr-lcd/?range=61915-smartups&parent-subcategory-id=88976&selectedNodeId=23679172486 is a LFP based UPS system.
Though not all models - you can see from APCRBC173-LI_NAM_EN_SDS_V1.3.pdf that specific pack is NMC. So you'd just have to investigate each model to see what it is.
So yes, all four are selling LFP units, though APC seems to be a bit harder to pick out which is which, not that it really matters much anymore from a safety perspective, but if it concerns you, the information IS available.
I'd feel perfectly comfortable with any of them technology wise these days (and in the past, too) though.
Most Li-Ion batteries sold for stationery applications will be LFP, just because they are cheaper and more thermally stable. NMC is not far off, but it doesn't make sense for stationery applications. You'd just be wasting energy for no gain. Mark.
Datacenter scale NMC UPSes have been a thing for a rather long time. As well as rack sized, and many others. Long before LFP units were even thought about being commercialized. https://www.renewableenergymagazine.com/energy_saving/kokam-launches-kups-li... - There was another line too of datacenter-scale that I can't remember the name of right now in 2017 as well. And many before then. And after, too, from the usual big names as well. As noted below, one of the APC units is NMC instead of LiFePo4. They sell a mix of the technologies. LiFePo4 came into more prominence later on. For most high-density installations in the somewhat distant past though, more 'traditional' higher density technologies were used - not LFP. I've been party to many large installations where even at that size, LFP would not have had the required density. LFP is a compromise - less capacity than other chemistries, but kinda-near-LA stability/safety. NMC is "old tech" in many ways, but on the higher end of density. I've got thousands of NMC cells here that get utilized/built into various applications, some, even, for preventing power interruption. Most of these, are all stationary applications - some house sized, some larger, many smaller. -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Wednesday, April 9, 2025 7:58 AM To: Gary Sparkes <gary@kisaracorporation.com>; Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/9/25 11:21, Gary Sparkes wrote:
So, they're all selling LFP units.
LFP is a type of lithium-ion battery. I could (and do) call a LiFePo4 cell a lithium-ion battery and be 100% correct - because it is. LiPo's, NMC/PO4 (be it in 18650 or other formfactor, etc), LFP, you name it, they're all lithium-ion batteries because that's the mechanism they operate by. They're just different chemistries and formfactors of lithium-ion batteries.
https://www.eaton.com/us/en-us/products/backup-power-ups-surge-it-powe r-distribution/backup-power-ups/lithium-ion-batteries-/eaton-9-series- lithium-ion-batteries.html
This liebert information sheet (second link of products after eaton in my list) - https://www.vertiv.com/491005/globalassets/products/critical-power/uninterru... - from this information sheet - "Safe and Reliable: Both VRLA and Li-ion ensure safety and reliability. Lithium-Ion (LiFePO4) battery chemistry is safest and integrated battery management system (BMS) also ensures safe and reliable operation."
The few APC MSDS (Materials Safety Data Sheet) sheets I glanced at indicate Lithium Iron Phosphate as the cell ingredient as well - which means LFP chemistry. So they're shipping LiFePo4 cells as well. Weird they didn't make it (easier) to find that, though. https://www.apc.com/us/en/faqs/FA158828/ - I looked at XBP48RM2U-LI_NAM_EN_SDSV1.0.pdf and SMTL750_1000_1500_LI_EU_EN_SDS_V2.0 (1).pdf
That second sheet shows that https://www.apc.com/us/en/product/SMTL750RM2UC/apc-smartups-line-interactive-750va-lithiumion-rack-tower-2u-120v-6x-nema-515r-outlets-smartconnect-port+smartslot-short-depth-avr-lcd/?range=61915-smartups&parent-subcategory-id=88976&selectedNodeId=23679172486 is a LFP based UPS system.
Though not all models - you can see from APCRBC173-LI_NAM_EN_SDS_V1.3.pdf that specific pack is NMC. So you'd just have to investigate each model to see what it is.
So yes, all four are selling LFP units, though APC seems to be a bit harder to pick out which is which, not that it really matters much anymore from a safety perspective, but if it concerns you, the information IS available.
I'd feel perfectly comfortable with any of them technology wise these days (and in the past, too) though.
Most Li-Ion batteries sold for stationery applications will be LFP, just because they are cheaper and more thermally stable. NMC is not far off, but it doesn't make sense for stationery applications. You'd just be wasting energy for no gain. Mark.
On 4/9/25 14:12, Gary Sparkes wrote:
Datacenter scale NMC UPSes have been a thing for a rather long time. As well as rack sized, and many others. Long before LFP units were even thought about being commercialized.
NMC is older than LFP because EV's required NMC chemistries during a time when stationery storage was not a thing. And yes, NMC makes sense for data centre applications because of the high energy density in such an environment. But AFAIK, most data centres still rely on more on LA than Li-Ion. Mark.
I would say that NMC is older than LFP because it was invented far beforehand. It's what enabled the transition off of NiMH and NiCAD - invented sometime around the 1980s. I've got many examples from the late 90s and early 2000s, one of my favorite being the iPaq 3850 sitting on my desk (and of course, tons and tons of cellular phones). But that's obviously not stationary.... LiFePo4 is far newer - late 90s/early 2000s - and not commercially viable until later. LFP makes more sense in a lot of applications because of voltages - it can be a "drop in" replacement for 12V systems due to the 3.2V nominal - and lifespan/cycle count. But NMC was commercialized far earlier in fixed storage, mostly apparent to me in the availability of house-scale lithium battery systems/solar, lithium UPS market at all scales, from 1U / switch rack units to full scall datacenter deployments. And Including according to some cursory research, grid-scale power systems as well - as LiFePo4 tech is still relatively new and developing. EV usage was more a selection because of density, but not a driver of that density development. Even EVs are moving in the LFP direction for safety reasons as that technology evolves. EVs using non-NMC existed before as well, due to technology limitations and evolution. Schneider's forecast in 2017 or 2018 for 40% of the datacenter market to be lithium based by around 2025. -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Wednesday, April 9, 2025 2:21 PM To: Gary Sparkes <gary@kisaracorporation.com>; Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/9/25 14:12, Gary Sparkes wrote:
Datacenter scale NMC UPSes have been a thing for a rather long time. As well as rack sized, and many others. Long before LFP units were even thought about being commercialized.
NMC is older than LFP because EV's required NMC chemistries during a time when stationery storage was not a thing. And yes, NMC makes sense for data centre applications because of the high energy density in such an environment. But AFAIK, most data centres still rely on more on LA than Li-Ion. Mark.
Correction on that, the lithium polymers such as in modern cellphones are a fair bit different. NMC was used in non-LiPo cells for early mobile devices, however, was a thing in many variants. I was thinking more along the lines of things like early laptops that switched off nicad/nimh while looking at devices on my desk -----Original Message----- From: Gary Sparkes via NANOG <nanog@lists.nanog.org> Sent: Wednesday, April 9, 2025 3:14 PM To: Mark Tinka <mark@tinka.africa>; Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org> Cc: Gary Sparkes <gary@kisaracorporation.com> Subject: [NANOG] Re: Small Capacity UPS I would say that NMC is older than LFP because it was invented far beforehand. It's what enabled the transition off of NiMH and NiCAD - invented sometime around the 1980s. I've got many examples from the late 90s and early 2000s, one of my favorite being the iPaq 3850 sitting on my desk (and of course, tons and tons of cellular phones). But that's obviously not stationary.... LiFePo4 is far newer - late 90s/early 2000s - and not commercially viable until later. LFP makes more sense in a lot of applications because of voltages - it can be a "drop in" replacement for 12V systems due to the 3.2V nominal - and lifespan/cycle count. But NMC was commercialized far earlier in fixed storage, mostly apparent to me in the availability of house-scale lithium battery systems/solar, lithium UPS market at all scales, from 1U / switch rack units to full scall datacenter deployments. And Including according to some cursory research, grid-scale power systems as well - as LiFePo4 tech is still relatively new and developing. EV usage was more a selection because of density, but not a driver of that density development. Even EVs are moving in the LFP direction for safety reasons as that technology evolves. EVs using non-NMC existed before as well, due to technology limitations and evolution. Schneider's forecast in 2017 or 2018 for 40% of the datacenter market to be lithium based by around 2025. -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Wednesday, April 9, 2025 2:21 PM To: Gary Sparkes <gary@kisaracorporation.com>; Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/9/25 14:12, Gary Sparkes wrote:
Datacenter scale NMC UPSes have been a thing for a rather long time. As well as rack sized, and many others. Long before LFP units were even thought about being commercialized.
NMC is older than LFP because EV's required NMC chemistries during a time when stationery storage was not a thing. And yes, NMC makes sense for data centre applications because of the high energy density in such an environment. But AFAIK, most data centres still rely on more on LA than Li-Ion. Mark. _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/JDWUYN7O...
On 4/9/25 21:17, Gary Sparkes wrote:
Correction on that, the lithium polymers such as in modern cellphones are a fair bit different. NMC was used in non-LiPo cells for early mobile devices, however, was a thing in many variants.
NMC can still be found in things like handheld power tools. Today, Li-Po (Lithium Polymer) and LCO (Lithium Cobalt Oxide) are typical options for mobile devices.
I was thinking more along the lines of things like early laptops that switched off nicad/nimh while looking at devices on my desk
Back when we spent quite a bit of cash on camcorders, we experienced plenty of NiMH batteries. It's quite an inferior battery, but it served its purpose well, at the time. I still find NiMH batteries shipping in modern gear today, e.g., low-end electric toothbrushes. I generally avoid these where possible, and buy Lithium-based toothbrushes :-). Mark.
Yeah, it's a shame how many electric toothbrushes I've had to throw out when the batteries go bad. Good brands also, but they don't make them serviceable. Since I'm in a hurricane zone, I have flashlights with 18650 cells. more cells than flashlights. Worst case scenario is that I can't get gas for the generator and the power is out for a week or whatever (and if you read my previous posts I'll solve that problem soon) I can charge batteries from my off grid backyard solar. My current problem is that comcast/xfinity goes out more than the power flickers. I can't wait for buried fiber on my street to come one day. On Sat, Apr 12, 2025 at 2:51 AM Mark Tinka via NANOG <nanog@lists.nanog.org> wrote:
On 4/9/25 21:17, Gary Sparkes wrote:
Correction on that, the lithium polymers such as in modern cellphones are a fair bit different. NMC was used in non-LiPo cells for early mobile devices, however, was a thing in many variants.
NMC can still be found in things like handheld power tools.
Today, Li-Po (Lithium Polymer) and LCO (Lithium Cobalt Oxide) are typical options for mobile devices.
I was thinking more along the lines of things like early laptops that switched off nicad/nimh while looking at devices on my desk
Back when we spent quite a bit of cash on camcorders, we experienced plenty of NiMH batteries. It's quite an inferior battery, but it served its purpose well, at the time.
I still find NiMH batteries shipping in modern gear today, e.g., low-end electric toothbrushes. I generally avoid these where possible, and buy Lithium-based toothbrushes :-).
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/O6SPTMAZ...
On 4/12/25 19:15, Javier J wrote:
Yeah, it's a shame how many electric toothbrushes I've had to throw out when the batteries go bad. Good brands also, but they don't make them serviceable.
Just always check to see what battery chemistry it comes with. They will usually label that on that back, inside the charging slot or on the charger itself. Avoid NiMH, stick to Li-Ion.
Since I'm in a hurricane zone, I have flashlights with 18650 cells. more cells than flashlights. Worst case scenario is that I can't get gas for the generator and the power is out for a week or whatever (and if you read my previous posts I'll solve that problem soon) I can charge batteries from my off grid backyard solar.
Sounds like you are sorted. Mark.
The sad thing is, buried fiber is not always an end-all, be-all answer. I’ve seen a significant number of “operators” that bury the fiber, but don’t put their OLT cabinets/huts on gensets… or if they do have generators, they are not maintained. This was proven to be a huge issue during Hurricane Beryl last year, where every FTTx operator in Houston (except for one) had large quantities of outages throughout the entire metro.
On Apr 12, 2025, at 12:16, Javier J via NANOG <nanog@lists.nanog.org> wrote:
My current problem is that comcast/xfinity goes out more than the power flickers. I can't wait for buried fiber on my street to come one day.
On 4/9/25 21:13, Gary Sparkes wrote:
I would say that NMC is older than LFP because it was invented far beforehand. It's what enabled the transition off of NiMH and NiCAD - invented sometime around the 1980s. I've got many examples from the late 90s and early 2000s, one of my favorite being the iPaq 3850 sitting on my desk (and of course, tons and tons of cellular phones). But that's obviously not stationary....
LiFePo4 is far newer - late 90s/early 2000s - and not commercially viable until later.
LFP makes more sense in a lot of applications because of voltages - it can be a "drop in" replacement for 12V systems due to the 3.2V nominal - and lifespan/cycle count.
But NMC was commercialized far earlier in fixed storage, mostly apparent to me in the availability of house-scale lithium battery systems/solar, lithium UPS market at all scales, from 1U / switch rack units to full scall datacenter deployments. And Including according to some cursory research, grid-scale power systems as well - as LiFePo4 tech is still relatively new and developing.
EV usage was more a selection because of density, but not a driver of that density development. Even EVs are moving in the LFP direction for safety reasons as that technology evolves. EVs using non-NMC existed before as well, due to technology limitations and evolution.
Largely agree, yes.
Schneider's forecast in 2017 or 2018 for 40% of the datacenter market to be lithium based by around 2025.
It's going to be a mind shift... like moving network operators across from 100G-LR4 plugs to 100G-FR1. Seems obvious, but is unnecessarily complicated :-). Mark.
On 4/9/25 10:56, Vasilenko Eduard wrote:
Your first 3 UPSes are lithium-ion. I am scared of lithium-ion in the apartment.
In a UPS, even I would be concerned. It's such a cramped space, and Li-Ion are not as resilient as LA when exposed to high ambient temperatures. That does not mean they will catch fire, but it surely means they will age significantly faster. But for a standard home stationery backup, Li-Ion just fine. I'm coming up on 6 years for my deployment, which is housed in my garage. 700Ah battery pack spread across 7x 100Ah 48V LFP batteries. Mark.
An interesting thing to be concerned about, given average temperature exposures compared to consumer electronics devices with usually inadequate cooling, compared to LiFePo4 in a UPS, which has been my baseline for years now as SLA's age out. -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Wednesday, April 9, 2025 7:47 AM To: Vasilenko Eduard <vasilenko.eduard@huawei.com>; Gary Sparkes <gary@kisaracorporation.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/9/25 10:56, Vasilenko Eduard wrote:
Your first 3 UPSes are lithium-ion. I am scared of lithium-ion in the apartment.
In a UPS, even I would be concerned. It's such a cramped space, and Li-Ion are not as resilient as LA when exposed to high ambient temperatures. That does not mean they will catch fire, but it surely means they will age significantly faster. But for a standard home stationery backup, Li-Ion just fine. I'm coming up on 6 years for my deployment, which is housed in my garage. 700Ah battery pack spread across 7x 100Ah 48V LFP batteries. Mark.
On 4/9/25 13:49, Gary Sparkes wrote:
An interesting thing to be concerned about, given average temperature exposures compared to consumer electronics devices with usually inadequate cooling, compared to LiFePo4 in a UPS, which has been my baseline for years now as SLA's age out.
It's often hard to keep Li-Ion batteries at the recommended 25°C. Some environments are worse than others, and while Li-Ion batteries can tolerate +/- 15°C, it's often not recommended. Some active cooling may be required for summer, or for perennially warmer regions. Of course, you also have to think about extremely cool environments. Li-Ion batteries do not like to be charged below 0°C. In such cases, better to place a heater next to the battery so it can warm up before you start charging it. Charging it when it's that cold will lead to plating, which causes short circuits and capacity loss. Mark.
Interesting thoughts - in most consumer devices, the temperature regulation is true. Heat's one of the worst enemies of laptop and cellphone batteries, after all. In house and datacenter environments for larger units, not really an issue. For operational spec for a lot of cells I'm used to, I'm used to 0 to 50c for design charge range for cells, and -20 to 75 for optimal discharge range. Obviously, energy output (and energy to charge) differ across the range, but that's well within design specification for the rated lifespans of the cells. Long term storage, of course, is different, but that's a much easier solved problem, and design specs are often in the -30c to 25c range for optimal capacity retention over long term storage. But operationally, 25-50c is perfectly fine. Past ... above 5c, I think, you're looking at a 10% capacity fluctuation in that 5c to 50c range. Of course, there's a ton of other factors that go into such things as well, such as actual recovered (discharged) capacity such as current draw, and whatnot, other than temperature. While true, heat is the enemy of lithium batteries of any type, deep discharging is the second worst enemy (BMS prevents actual cell damaging deep discharge, but you really should never go below 30% of rated capacity if you can ever help it - this is the most stressful zone), then charging too rapidly - too high rate of charge. As always though, it's best to look at the datasheets before making any decisions overall, of course. They'll cover all these operational specs and more. -----Original Message----- From: Mark Tinka <mark@tinka.africa> Sent: Wednesday, April 9, 2025 8:05 AM To: Gary Sparkes <gary@kisaracorporation.com>; Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/9/25 13:49, Gary Sparkes wrote:
An interesting thing to be concerned about, given average temperature exposures compared to consumer electronics devices with usually inadequate cooling, compared to LiFePo4 in a UPS, which has been my baseline for years now as SLA's age out.
It's often hard to keep Li-Ion batteries at the recommended 25°C. Some environments are worse than others, and while Li-Ion batteries can tolerate +/- 15°C, it's often not recommended. Some active cooling may be required for summer, or for perennially warmer regions. Of course, you also have to think about extremely cool environments. Li-Ion batteries do not like to be charged below 0°C. In such cases, better to place a heater next to the battery so it can warm up before you start charging it. Charging it when it's that cold will lead to plating, which causes short circuits and capacity loss. Mark.
On 4/9/25 14:27, Gary Sparkes wrote:
Long term storage, of course, is different, but that's a much easier solved problem, and design specs are often in the -30c to 25c range for optimal capacity retention over long term storage. But operationally, 25-50c is perfectly fine. Past ... above 5c, I think, you're looking at a 10% capacity fluctuation in that 5c to 50c range.
Yes, most LFP batteries will work even at 60°C, but that does not mean you should do it if you want to get the most out of your investment. The biggest issue is battery aging, and keeping it within recommended temperature values will ensure they last longer than if you don't.
While true, heat is the enemy of lithium batteries of any type, deep discharging is the second worst enemy (BMS prevents actual cell damaging deep discharge, but you really should never go below 30% of rated capacity if you can ever help it - this is the most stressful zone), then charging too rapidly - too high rate of charge.
The point of the BMS is to prevent you from operating your battery in dangerous territory. 30% is not a reasonable floor to stop discharging a Li-Ion battery. All you are doing there is limiting your available capacity. The recommended floor to stick to is 20%. However, it is perfectly fine to go as low as 10% without any appreciable loss of overall capacity. And in an emergency, you can even go to 5% and still maintain good long term capacity. This is where Li-Ion outshines LA by a mile. Limiting your capacity by not discharging to, at least, 20% SoC, is quite pointless based on all the information we have so far, because calendar aging is still going to get you either way. Mark.
I agree. The cheap solar batteries in my shed have gone down to the point the BMS turned them off. So I added a second solar panel, then I added 3 more batteries in parallel. I just recently (with the help of my landscapers awesome shovel) moved the panels to a more optimal location and buried the DC cable to the shed. Now I never get lower than 70% SOC. I'm at 100% by 1pm on sunny days. It's a test off-grid system to power poe cams, smart bulbs, watch what wild animals try to eat my chickens etc. But really it is to see what location is best because I won't be cutting down trees I don't own, or don't want to. Going strong since 2013. Some of these batteries now have bluetooth, I added a bluetooth module on mine. BMS protects the cells, keeps them balanced etc, just to add to that. I couldn't find the video but Will Prowse on youtube has a solar powered gate opener operating for years, battery exposed to the elements in las vegas, going strong after years. I don't think I will ever buy a lead acid based battery again. On Thu, Apr 10, 2025 at 1:13 AM Mark Tinka via NANOG <nanog@lists.nanog.org> wrote:
On 4/9/25 14:27, Gary Sparkes wrote:
Long term storage, of course, is different, but that's a much easier solved problem, and design specs are often in the -30c to 25c range for optimal capacity retention over long term storage. But operationally, 25-50c is perfectly fine. Past ... above 5c, I think, you're looking at a 10% capacity fluctuation in that 5c to 50c range.
Yes, most LFP batteries will work even at 60°C, but that does not mean you should do it if you want to get the most out of your investment.
The biggest issue is battery aging, and keeping it within recommended temperature values will ensure they last longer than if you don't.
While true, heat is the enemy of lithium batteries of any type, deep discharging is the second worst enemy (BMS prevents actual cell damaging deep discharge, but you really should never go below 30% of rated capacity if you can ever help it - this is the most stressful zone), then charging too rapidly - too high rate of charge.
The point of the BMS is to prevent you from operating your battery in dangerous territory.
30% is not a reasonable floor to stop discharging a Li-Ion battery. All you are doing there is limiting your available capacity. The recommended floor to stick to is 20%. However, it is perfectly fine to go as low as 10% without any appreciable loss of overall capacity. And in an emergency, you can even go to 5% and still maintain good long term capacity.
This is where Li-Ion outshines LA by a mile.
Limiting your capacity by not discharging to, at least, 20% SoC, is quite pointless based on all the information we have so far, because calendar aging is still going to get you either way.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/5UTUKUX4...
“30% is not a reasonable floor to stop discharging a Li-Ion battery. All you are doing there is limiting your available capacity. The recommended floor to stick to is 20%. However, it is perfectly fine to go as low as 10% without any appreciable loss of overall capacity. And in an emergency, you can even go to 5% and still maintain good long term capacity. This is where Li-Ion outshines LA by a mile. Limiting your capacity by not discharging to, at least, 20% SoC, is quite pointless based on all the information we have so far, because calendar aging is still going to get you either way.” This is… not quite correct. Age can be mostly mitigated by proper storage – at around 40% SoC for storage charge. I routinely pull out 15-20 (and older!) packs that were manufacturer properly storage charged and once jumped up a bit return with 80-90% capacity! But 30% is a reasonable floor for normal usage, going below that IS a stress zone for the battery – you sharply reduce the number of possible full cycles if you delve into this range regularly. Obviously, when you actually need to use the battery, use it up entirely, that’s what it’s there for. This guideline’s more in general in terms of if you can stop at that point, you really, really should do so. For what it’s worth, I was misremembering the floors slightly. It was 40% and 20% I remember, where stopping discharge at 40% gave you nearly *double* the number of full cycles compared to a 20% discharge. IE say you get 600 if you routinely go to 20%, routinely going to 40% could net you 1500 instead. From: Mark Tinka <mark@tinka.africa> Sent: Thursday, April 10, 2025 1:12 AM To: Gary Sparkes <gary@kisaracorporation.com>; Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/9/25 14:27, Gary Sparkes wrote: Long term storage, of course, is different, but that's a much easier solved problem, and design specs are often in the -30c to 25c range for optimal capacity retention over long term storage. But operationally, 25-50c is perfectly fine. Past ... above 5c, I think, you're looking at a 10% capacity fluctuation in that 5c to 50c range. Yes, most LFP batteries will work even at 60°C, but that does not mean you should do it if you want to get the most out of your investment. The biggest issue is battery aging, and keeping it within recommended temperature values will ensure they last longer than if you don't. While true, heat is the enemy of lithium batteries of any type, deep discharging is the second worst enemy (BMS prevents actual cell damaging deep discharge, but you really should never go below 30% of rated capacity if you can ever help it - this is the most stressful zone), then charging too rapidly - too high rate of charge. The point of the BMS is to prevent you from operating your battery in dangerous territory. Mark.
On 4/10/25 22:44, Gary Sparkes wrote:
This is… not quite correct.
Age can be mostly mitigated by proper storage – at around 40% SoC for storage charge. I routinely pull out 15-20 (and older!) packs that were manufacturer properly storage charged and once jumped up a bit return with 80-90% capacity!
I'm not talking about batteries that are in storage. Batteries in storage are not providing value to anyone :-). Yes, it is true that Li-Ion batteries store better than LA, because Li-Ion self-discharge can be 1.5% - 2% per month. LA batteries self-discharge at a rate of 4% - 6% per month. But that is not what we are talking about here.
But 30% is a reasonable floor for normal usage, going below that IS a stress zone for the battery – you sharply reduce the number of possible full cycles if you delve into this range regularly.
No, not true at all. While you do reduce SoH when you discharge below 20% or if you hold SoC above 80% for too long, how far you go in either direction and for what period does matter. The lower you discharge below 20%, the more internal resistance you introduce into the battery. The higher you charge above 80%, the faster you cause capacity fade due to a high battery voltage. Battery manufacturers all generally use a DoD (Depth of Discharge) of 80% to rate a battery, i.e., discharging from 100% down to 20%. EV's typically charge to 80% and discharged to 30% to increase mileage. But as the battery ages, this gap widens. This range does not really apply to stationery applications, because EV's put more stress on their batteries than a home backup solution, for example.
Obviously, when you actually need to use the battery, use it up entirely, that’s what it’s there for. This guideline’s more in general in terms of if you can stop at that point, you really, really should do so.
For what it’s worth, I was misremembering the floors slightly. It was 40% and 20% I remember, where stopping discharge at 40% gave you nearly **double** the number of full cycles compared to a 20% discharge. IE say you get 600 if you routinely go to 20%, routinely going to 40% could net you 1500 instead.
The less you give the battery a full charge (so keeping it below 100%), the longer it will last. The less you discharge a battery, the longer it will last. But the largest capacity loss actually occurs when the battery remains at a higher voltage for too long. However, cells can only balance when they have a full charge, although once balanced, they will remain balanced even at lower voltages for several months before they begin to drift and need to be re-balanced. In practical terms, the battery becomes useless to you with a partial discharge because you are limiting the amount of capacity you get out of it. And while you will get more cycles from a partial discharge, you are still going to have to deal with calendar aging, which is inevitable. Current research and testing data indicates that the biggest reason for capacity loss is calendar aging and maintaining a high voltage for too long (100% SoC), rather than cycling. Mark.
The less you give the battery a full charge (so keeping it below 100%), the longer it will last. The less you discharge a battery, the longer it will last.
This is why lifepo4 is a perfect swap in for old UPS devices like im doing. usually full charge is 14.6v is what the chargers my talentcell and solar charge controller cut off at, the batteries rest at about 13.5 full charge. Currently one of my UPS is floating at 13.7 each battery, another older UPS at 13.8. I've easily gotten more runtime out of these than what I have experienced with SLA or AGM batteries. my use case (home & home-office), keeping access points (4) internet, router, switches etc, even home theater (with the exception of the sub amps) up and running gives me like 2+ hours. Plenty of time to get the generator up and running. Last time we had an outage for an hour plus I wasn't even home. Didn't even notice it. - J On Fri, Apr 11, 2025 at 12:40 AM Mark Tinka via NANOG <nanog@lists.nanog.org> wrote:
On 4/10/25 22:44, Gary Sparkes wrote:
This is… not quite correct.
Age can be mostly mitigated by proper storage – at around 40% SoC for storage charge. I routinely pull out 15-20 (and older!) packs that were manufacturer properly storage charged and once jumped up a bit return with 80-90% capacity!
I'm not talking about batteries that are in storage. Batteries in storage are not providing value to anyone :-).
Yes, it is true that Li-Ion batteries store better than LA, because Li-Ion self-discharge can be 1.5% - 2% per month. LA batteries self-discharge at a rate of 4% - 6% per month. But that is not what we are talking about here.
But 30% is a reasonable floor for normal usage, going below that IS a stress zone for the battery – you sharply reduce the number of possible full cycles if you delve into this range regularly.
No, not true at all. While you do reduce SoH when you discharge below 20% or if you hold SoC above 80% for too long, how far you go in either direction and for what period does matter. The lower you discharge below 20%, the more internal resistance you introduce into the battery. The higher you charge above 80%, the faster you cause capacity fade due to a high battery voltage.
Battery manufacturers all generally use a DoD (Depth of Discharge) of 80% to rate a battery, i.e., discharging from 100% down to 20%.
EV's typically charge to 80% and discharged to 30% to increase mileage. But as the battery ages, this gap widens. This range does not really apply to stationery applications, because EV's put more stress on their batteries than a home backup solution, for example.
Obviously, when you actually need to use the battery, use it up entirely, that’s what it’s there for. This guideline’s more in general in terms of if you can stop at that point, you really, really should do so.
For what it’s worth, I was misremembering the floors slightly. It was 40% and 20% I remember, where stopping discharge at 40% gave you nearly **double** the number of full cycles compared to a 20% discharge. IE say you get 600 if you routinely go to 20%, routinely going to 40% could net you 1500 instead.
The less you give the battery a full charge (so keeping it below 100%), the longer it will last. The less you discharge a battery, the longer it will last. But the largest capacity loss actually occurs when the battery remains at a higher voltage for too long. However, cells can only balance when they have a full charge, although once balanced, they will remain balanced even at lower voltages for several months before they begin to drift and need to be re-balanced.
In practical terms, the battery becomes useless to you with a partial discharge because you are limiting the amount of capacity you get out of it. And while you will get more cycles from a partial discharge, you are still going to have to deal with calendar aging, which is inevitable.
Current research and testing data indicates that the biggest reason for capacity loss is calendar aging and maintaining a high voltage for too long (100% SoC), rather than cycling.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/VQI2SBNI...
On 4/11/25 14:45, Javier J via NANOG wrote:
This is why lifepo4 is a perfect swap in for old UPS devices like im doing. usually full charge is 14.6v is what the chargers my talentcell and solar charge controller cut off at, the batteries rest at about 13.5 full charge.
Currently one of my UPS is floating at 13.7 each battery, another older UPS at 13.8.
I've easily gotten more runtime out of these than what I have experienced with SLA or AGM batteries. my use case (home & home-office), keeping access points (4) internet, router, switches etc, even home theater (with the exception of the sub amps) up and running gives me like 2+ hours. Plenty of time to get the generator up and running. Last time we had an outage for an hour plus I wasn't even home. Didn't even notice it.
Even if an LA and Li-Ion battery may have the same capacity, the operating assumption for an LA battery is that it will not be discharged below 50%. While you can discharge an LA battery below 50%, it will significantly reduce its lifespan, which is why they tend to be replaced every so often. In other words, Li-Ion batteries allow you to access nearly all of the battery's capacity without negatively impacting lifespan. This is the primary advantage Li-Ion has over LA. To get the same usable capacity from LA as you would Li-Ion, you'd have to double/triple/quadruple your battery size to gain usable capacity at a maximum DoD of 50%. Mark.
There is also the discharge curve, I think this is why my UPS upgrades have been lasting way longer on battery: [image: image.png] And the batteries are way lighter. Here is output from my bluetooth solar project. you can see it maxes out and then just maintains at somewhere in the 13.5 volts. 100% SOC area. [image: solar-april-11.jpg] As you can see, the voltage is stable and then jumps to 100% when fully charged. I find that interesting. On Fri, Apr 11, 2025 at 9:29 AM Mark Tinka <mark@tinka.africa> wrote:
On 4/11/25 14:45, Javier J via NANOG wrote:
This is why lifepo4 is a perfect swap in for old UPS devices like im doing. usually full charge is 14.6v is what the chargers my talentcell and solar charge controller cut off at, the batteries rest at about 13.5 full charge.
Currently one of my UPS is floating at 13.7 each battery, another older UPS at 13.8.
I've easily gotten more runtime out of these than what I have experienced with SLA or AGM batteries. my use case (home & home-office), keeping access points (4) internet, router, switches etc, even home theater (with the exception of the sub amps) up and running gives me like 2+ hours. Plenty of time to get the generator up and running. Last time we had an outage for an hour plus I wasn't even home. Didn't even notice it.
Even if an LA and Li-Ion battery may have the same capacity, the operating assumption for an LA battery is that it will not be discharged below 50%. While you can discharge an LA battery below 50%, it will significantly reduce its lifespan, which is why they tend to be replaced every so often.
In other words, Li-Ion batteries allow you to access nearly all of the battery's capacity without negatively impacting lifespan. This is the primary advantage Li-Ion has over LA.
To get the same usable capacity from LA as you would Li-Ion, you'd have to double/triple/quadruple your battery size to gain usable capacity at a maximum DoD of 50%.
Mark.
On 4/11/25 17:02, Javier J wrote:
There is also the discharge curve, I think this is why my UPS upgrades have been lasting way longer on battery:
Yes, Li-Ion's discharge profile is rather flat compared to LA which is more linear. Li-Ion will hold a steady voltage throughout discharge, and sharply drop off as cell voltage gets into the region of 2.9V - 3.0V (LFP). The BMS will kick in somewhere between 2.5V - 2.8V to protect the battery from further discharge.
And the batteries are way lighter.
Much higher energy density for Li-Ion than LA, which translates into a lot less floor space and weight for the same capacity.
Here is output from my bluetooth solar project. you can see it maxes out and then just maintains at somewhere in the 13.5 volts. 100% SOC area.
These are good values. For 12V LFP batteries, I'll charge them to about 14.2V, and then let them float at 13.7V. But 13.5V to float is great to increase battery life. Mark.
30% being a reasonable floor absolutely is true. Far less stress to go 100 to 30 and back to 100, then 90 to 20 and back to 90, etc. Keeping 30 as an operational floor lets you use full capacity as needed and remain at full functional charge with maximum lifespan retention/recovery. EV’s charge/operate that way to extend pack lifespan primarily. 4.2v float is fine if you aren’t routinely low end stressing it. Yes, the internal resistance bit is true, but that really does start to kick in around 25-30. A lot of datasheets I’ve worked with talk about 30% and stress zones. Far better to remain at 100% float for battery lifespan than to routinely dip below 30%. Calendar aging is not as cut and dried as it may seem. This is blatantly obvious with cellphones, of course, but holds true for lots of other implementations too. I expect 90% capacity on 10-year-old batteries stored properly almost always, usually – at a minimum. For capacity ratings, I’ve always done 100% DoD discharge tests to get around the manufacturer ratings for capacity. Cycle durability of course, you are generally right. But cycle durability is what truly matters in the long run for lifespan, not calendar aging, for batteries that often see use. And 30% is a sweet spot between usable capacity and lifespan extension to often double the manufacturer’s rated cycle count. Remember, cycle count means you’re actually using it – so I’m not charging or discharging any less, only doing so to specific levels. I’d also point out a lot of research is also indicative of low discharge levels being the leading factor to degradation, not capacity float charge status. Heat is the the number one factor (outside of or at high end of design spec heat, as often seen in consumer devices). Low discharge is factor #2. High / full charge stress comes in around #3. Better to engage #3 than #2 or #1 for lifespan retention. From: Mark Tinka <mark@tinka.africa> Sent: Friday, April 11, 2025 12:39 AM To: Gary Sparkes <gary@kisaracorporation.com>; Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/10/25 22:44, Gary Sparkes wrote: This is… not quite correct. Age can be mostly mitigated by proper storage – at around 40% SoC for storage charge. I routinely pull out 15-20 (and older!) packs that were manufacturer properly storage charged and once jumped up a bit return with 80-90% capacity! I'm not talking about batteries that are in storage. Batteries in storage are not providing value to anyone :-). Yes, it is true that Li-Ion batteries store better than LA, because Li-Ion self-discharge can be 1.5% - 2% per month. LA batteries self-discharge at a rate of 4% - 6% per month. But that is not what we are talking about here. But 30% is a reasonable floor for normal usage, going below that IS a stress zone for the battery – you sharply reduce the number of possible full cycles if you delve into this range regularly. No, not true at all. While you do reduce SoH when you discharge below 20% or if you hold SoC above 80% for too long, how far you go in either direction and for what period does matter. The lower you discharge below 20%, the more internal resistance you introduce into the battery. The higher you charge above 80%, the faster you cause capacity fade due to a high battery voltage. Battery manufacturers all generally use a DoD (Depth of Discharge) of 80% to rate a battery, i.e., discharging from 100% down to 20%. EV's typically charge to 80% and discharged to 30% to increase mileage. But as the battery ages, this gap widens. This range does not really apply to stationery applications, because EV's put more stress on their batteries than a home backup solution, for example. Obviously, when you actually need to use the battery, use it up entirely, that’s what it’s there for. This guideline’s more in general in terms of if you can stop at that point, you really, really should do so. For what it’s worth, I was misremembering the floors slightly. It was 40% and 20% I remember, where stopping discharge at 40% gave you nearly *double* the number of full cycles compared to a 20% discharge. IE say you get 600 if you routinely go to 20%, routinely going to 40% could net you 1500 instead. The less you give the battery a full charge (so keeping it below 100%), the longer it will last. The less you discharge a battery, the longer it will last. But the largest capacity loss actually occurs when the battery remains at a higher voltage for too long. However, cells can only balance when they have a full charge, although once balanced, they will remain balanced even at lower voltages for several months before they begin to drift and need to be re-balanced. In practical terms, the battery becomes useless to you with a partial discharge because you are limiting the amount of capacity you get out of it. And while you will get more cycles from a partial discharge, you are still going to have to deal with calendar aging, which is inevitable. Current research and testing data indicates that the biggest reason for capacity loss is calendar aging and maintaining a high voltage for too long (100% SoC), rather than cycling. Mark.
On 4/11/25 17:40, Gary Sparkes wrote:
30% being a reasonable floor absolutely is true.
I didn't say 30% DoD was unreasonable. I said that claiming that going below it to 20% is dangerous is not true. But, happy to agree to disagree.
Far less stress to go 100 to 30 and back to 100, then 90 to 20 and back to 90, etc. Keeping 30 as an operational floor lets you use full capacity as needed and remain at full functional charge with maximum lifespan retention/recovery.
That can be said of any charge/discharge window if you are not consistently discharging below a given threshold, whatever that is to you.
EV’s charge/operate that way to extend pack lifespan primarily.
Right, but the key motivation for that is for the car manufacturers to meet warranty claims, normally at least 10 years. And they will remotely manage those charge/discharge profiles to put warranty objectives over range maximization.
4.2v float is fine if you aren’t routinely low end stressing it.
4.2V float is for NMC. LFP is usually around 3.5V float.
Yes, the internal resistance bit is true, but that really does start to kick in around 25-30. A lot of datasheets I’ve worked with talk about 30% and stress zones.
Internal resistance in Li-Ion cells is highest as the battery approaches a fully-charged or fully-discharged state. In other words, internal resistance is highest at 100% SoC and below 20% SoC. But since we know that Li-Ion batteries have a non-linear voltage curve until about 10% SoC, internal resistance is most dangerous below this SoC value.
Far better to remain at 100% float for battery lifespan than to routinely dip below 30%.
In my experience, not going below 20% SoC will be better than holding a 100% charge for an extended period of time, especially if you are not actively controlling ambient temperature. Li-Ion batteries really do not like holding a full voltage for too long, although, for me, that would not be as bad as routinely running an SoC below 20%.
Calendar aging is not as cut and dried as it may seem. This is blatantly obvious with cellphones, of course, but holds true for lots of other implementations too.
Li-Ion batteries have a completely different use-case for cellphones than for home backup, because we prioritize capacity and peak performance for cellphones vs. home backup. This is why it is quite normal for people to expect their phone battery to be pretty "useless" after an average of 3 years.
I expect 90% capacity on 10-year-old batteries stored properly almost always, usually – at a minimum.
Again, stored batteries have no value to anyone :-). The better metric is how to maintain working batteries for 10 years and see how much capacity you've retained by that time.
But cycle durability is what truly matters in the long run for lifespan, not calendar aging, for batteries that often see use. And 30% is a sweet spot between usable capacity and lifespan extension to often double the manufacturer’s rated cycle count.
Yes, we all want cycle durability, but calendar aging is unavoidable. And since the biggest contributor to calendar aging is ambient temperature, most owners will lose capacity due to that because they do not have active cooling for their batteries. I disagree with 30% being a recommended DoD floor (most OEM's do not suggest that), but that's okay :-).
Remember, cycle count means you’re actually using it – so I’m not charging or discharging any less, only doing so to specific levels.
Li-Ion batteries are not cheap. People will always prioritize capacity, and the money to buy a larger pack just to save 10% from your discharge cycle does not justify the extra longevity in environments where most people will let batteries overheat.
I’d also point out a lot of research is also indicative of low discharge levels being the leading factor to degradation, not capacity float charge status. Heat is the the number one factor (outside of or at high end of design spec heat, as often seen in consumer devices). Low discharge is factor #2. High / full charge stress comes in around #3. Better to engage #3 than #2 or #1 for lifespan retention.
Maintaining a long term 100% SoC state is not problematic if you can actively cool your battery. Most owners will not, which is why this can be more damaging to your battery than a low SoC. Remember, most batteries will spend the majority of their life closer to 100% SoC than at 1% SoC. Mark.
where do you source those batteries?
The cheapest ones that will fit in the ups from amazon with a decent rating. I literally use a tape measure to measure the old dead lead acid batteries to see what will fit. I have found that I can get batteries with more amp hours than the lead acid replacements. I figure even if the manufacturer is crap will outlast LA battery technology. - J On Fri, Apr 11, 2025 at 4:52 PM Mark Tinka via NANOG <nanog@lists.nanog.org> wrote:
On 4/11/25 17:40, Gary Sparkes wrote:
30% being a reasonable floor absolutely is true.
I didn't say 30% DoD was unreasonable. I said that claiming that going below it to 20% is dangerous is not true.
But, happy to agree to disagree.
Far less stress to go 100 to 30 and back to 100, then 90 to 20 and back to 90, etc. Keeping 30 as an operational floor lets you use full capacity as needed and remain at full functional charge with maximum lifespan retention/recovery.
That can be said of any charge/discharge window if you are not consistently discharging below a given threshold, whatever that is to you.
EV’s charge/operate that way to extend pack lifespan primarily.
Right, but the key motivation for that is for the car manufacturers to meet warranty claims, normally at least 10 years. And they will remotely manage those charge/discharge profiles to put warranty objectives over range maximization.
4.2v float is fine if you aren’t routinely low end stressing it.
4.2V float is for NMC. LFP is usually around 3.5V float.
Yes, the internal resistance bit is true, but that really does start to kick in around 25-30. A lot of datasheets I’ve worked with talk about 30% and stress zones.
Internal resistance in Li-Ion cells is highest as the battery approaches a fully-charged or fully-discharged state. In other words, internal resistance is highest at 100% SoC and below 20% SoC. But since we know that Li-Ion batteries have a non-linear voltage curve until about 10% SoC, internal resistance is most dangerous below this SoC value.
Far better to remain at 100% float for battery lifespan than to routinely dip below 30%.
In my experience, not going below 20% SoC will be better than holding a 100% charge for an extended period of time, especially if you are not actively controlling ambient temperature.
Li-Ion batteries really do not like holding a full voltage for too long, although, for me, that would not be as bad as routinely running an SoC below 20%.
Calendar aging is not as cut and dried as it may seem. This is blatantly obvious with cellphones, of course, but holds true for lots of other implementations too.
Li-Ion batteries have a completely different use-case for cellphones than for home backup, because we prioritize capacity and peak performance for cellphones vs. home backup. This is why it is quite normal for people to expect their phone battery to be pretty "useless" after an average of 3 years.
I expect 90% capacity on 10-year-old batteries stored properly almost always, usually – at a minimum.
Again, stored batteries have no value to anyone :-).
The better metric is how to maintain working batteries for 10 years and see how much capacity you've retained by that time.
But cycle durability is what truly matters in the long run for lifespan, not calendar aging, for batteries that often see use. And 30% is a sweet spot between usable capacity and lifespan extension to often double the manufacturer’s rated cycle count.
Yes, we all want cycle durability, but calendar aging is unavoidable. And since the biggest contributor to calendar aging is ambient temperature, most owners will lose capacity due to that because they do not have active cooling for their batteries.
I disagree with 30% being a recommended DoD floor (most OEM's do not suggest that), but that's okay :-).
Remember, cycle count means you’re actually using it – so I’m not charging or discharging any less, only doing so to specific levels.
Li-Ion batteries are not cheap. People will always prioritize capacity, and the money to buy a larger pack just to save 10% from your discharge cycle does not justify the extra longevity in environments where most people will let batteries overheat.
I’d also point out a lot of research is also indicative of low discharge levels being the leading factor to degradation, not capacity float charge status. Heat is the the number one factor (outside of or at high end of design spec heat, as often seen in consumer devices). Low discharge is factor #2. High / full charge stress comes in around #3. Better to engage #3 than #2 or #1 for lifespan retention.
Maintaining a long term 100% SoC state is not problematic if you can actively cool your battery. Most owners will not, which is why this can be more damaging to your battery than a low SoC.
Remember, most batteries will spend the majority of their life closer to 100% SoC than at 1% SoC.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/S4CZUXPX...
This is probably not what you want but for a truly tiny DIY "UPS" you can use a USB-C PD compatible power pack / bank with an external mini PCB trigger board that lets you manually select output voltage. Normally with PD (Power Delivery) spec power packs the device and power pack "negotiate" proper voltage and current for fast charging depending on multiple factors. These small PCB trigger boards instead let you manually select output voltage (you probably are gonna want 12V I'm guessing for most devices). You can then wire this to a barrel connector or whatever an individual device uses (probably would want one battery pack and board per device). This is, of course, assuming you can power your devices directly off direct current (DC) power. If you need AC I have also had luck experimenting with the same setup outputting 12v DC to small tiny fanless AC inverters. By using an existing USB-C PD power bank you can size it appropriately for the device and time needed and also be reasonably confident that charging and discharging of the internal cells is done safely (assuming you choose a reputable power bank brand). To properly do this you would probably want a power bank with two ports so you could use one for input (battery charging) and the other for output (to the control board and then the final device). Usually these are called "USB-C PD Trigger Boards". I couldn't find the exact one I have but it has a button to select output voltage (5v, 12v, 18v, etc). No idea how efficient these are - some may use buck converters or cheaper parts or vary in quality. Others can be permanently set to a specific voltage via jumper or soldering and appear to be simpler/ smaller in design. I also found this project (I have never used it) which looks to be of possibly of higher quality (and significantly higher price): https://lectronz.com/products/pocketpd This is something similar to what I have used (I have not used this specific board - so buyer beware of course): https://a.co/d/i1mtXJw AITRIP Type-C USB-C Fast Charging Trigger Module (ZY12PDN with Screw Terminal) The people I know doing the vanlife thing also have amazing large and portable battery setups - would be worth looking at forum posts on their setups. For portable prebuilt setups (not custom LiFePO4 installs integrated into the van) I see the "EcoFlow" brand recommended a lot. I also don't know if this was specifically discussed in this thread but have you looked at the smaller Cyberpower "pure sine wave" UPS models ? They are not designed to be rack mounted but are quite compact. I have an older model and just running some network equipment at home it seems to last a few hours. I have not measured the efficiency under low loads like this though - it could be similarly terrible. Ideal world you would avoid the use of an AC inverter entirely (by using network equipment that can run off DC power directly). Does anyone know if the efficiency drops off at lower draws is because the AC inverter being less efficient at low power draw? Or is it somewhere else in the power chain ? On Fri, 11 Apr 2025 at 15:28, Javier J via NANOG <nanog@lists.nanog.org> wrote:
where do you source those batteries?
The cheapest ones that will fit in the ups from amazon with a decent rating.
I literally use a tape measure to measure the old dead lead acid batteries to see what will fit. I have found that I can get batteries with more amp hours than the lead acid replacements.
I figure even if the manufacturer is crap will outlast LA battery technology.
- J
On Fri, Apr 11, 2025 at 4:52 PM Mark Tinka via NANOG < nanog@lists.nanog.org> wrote:
On 4/11/25 17:40, Gary Sparkes wrote:
30% being a reasonable floor absolutely is true.
I didn't say 30% DoD was unreasonable. I said that claiming that going below it to 20% is dangerous is not true.
But, happy to agree to disagree.
Far less stress to go 100 to 30 and back to 100, then 90 to 20 and back to 90, etc. Keeping 30 as an operational floor lets you use full capacity as needed and remain at full functional charge with maximum lifespan retention/recovery.
That can be said of any charge/discharge window if you are not consistently discharging below a given threshold, whatever that is to
you.
EV’s charge/operate that way to extend pack lifespan primarily.
Right, but the key motivation for that is for the car manufacturers to meet warranty claims, normally at least 10 years. And they will remotely manage those charge/discharge profiles to put warranty objectives over range maximization.
4.2v float is fine if you aren’t routinely low end stressing it.
4.2V float is for NMC. LFP is usually around 3.5V float.
Yes, the internal resistance bit is true, but that really does start to kick in around 25-30. A lot of datasheets I’ve worked with talk about 30% and stress zones.
Internal resistance in Li-Ion cells is highest as the battery approaches a fully-charged or fully-discharged state. In other words, internal resistance is highest at 100% SoC and below 20% SoC. But since we know that Li-Ion batteries have a non-linear voltage curve until about 10% SoC, internal resistance is most dangerous below this SoC value.
Far better to remain at 100% float for battery lifespan than to routinely dip below 30%.
In my experience, not going below 20% SoC will be better than holding a 100% charge for an extended period of time, especially if you are not actively controlling ambient temperature.
Li-Ion batteries really do not like holding a full voltage for too long, although, for me, that would not be as bad as routinely running an SoC below 20%.
Calendar aging is not as cut and dried as it may seem. This is blatantly obvious with cellphones, of course, but holds true for lots of other implementations too.
Li-Ion batteries have a completely different use-case for cellphones than for home backup, because we prioritize capacity and peak performance for cellphones vs. home backup. This is why it is quite normal for people to expect their phone battery to be pretty "useless" after an average of 3 years.
I expect 90% capacity on 10-year-old batteries stored properly almost always, usually – at a minimum.
Again, stored batteries have no value to anyone :-).
The better metric is how to maintain working batteries for 10 years and see how much capacity you've retained by that time.
But cycle durability is what truly matters in the long run for lifespan, not calendar aging, for batteries that often see use. And 30% is a sweet spot between usable capacity and lifespan extension to often double the manufacturer’s rated cycle count.
Yes, we all want cycle durability, but calendar aging is unavoidable. And since the biggest contributor to calendar aging is ambient temperature, most owners will lose capacity due to that because they do not have active cooling for their batteries.
I disagree with 30% being a recommended DoD floor (most OEM's do not suggest that), but that's okay :-).
Remember, cycle count means you’re actually using it – so I’m not charging or discharging any less, only doing so to specific levels.
Li-Ion batteries are not cheap. People will always prioritize capacity, and the money to buy a larger pack just to save 10% from your discharge cycle does not justify the extra longevity in environments where most people will let batteries overheat.
I’d also point out a lot of research is also indicative of low discharge levels being the leading factor to degradation, not capacity float charge status. Heat is the the number one factor (outside of or at high end of design spec heat, as often seen in consumer devices). Low discharge is factor #2. High / full charge stress comes in around #3. Better to engage #3 than #2 or #1 for lifespan retention.
Maintaining a long term 100% SoC state is not problematic if you can actively cool your battery. Most owners will not, which is why this can be more damaging to your battery than a low SoC.
Remember, most batteries will spend the majority of their life closer to 100% SoC than at 1% SoC.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/S4CZUXPX... _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/AY2K4WBT...
For a small micro UPS, why not something like this? https://goprecisiongroup.com/product/fttx-desktop-ups-12v-36w-indoor-pr1236d... For Lithium Ion https://goprecisiongroup.com/product/24-hour-kit-li-36-micro-ups-12v-36w-ind... We have seen these in the $60-80 range Thank you Travis Garrison -----Original Message----- From: Christopher I. Sinclair via NANOG <nanog@lists.nanog.org> Sent: Friday, April 11, 2025 6:27 PM To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Gary Sparkes <gary@kisaracorporation.com>; Christopher I. Sinclair <chrisisinclair@gmail.com> Subject: [NANOG] Re: Small Capacity UPS This is probably not what you want but for a truly tiny DIY "UPS" you can use a USB-C PD compatible power pack / bank with an external mini PCB trigger board that lets you manually select output voltage. Normally with PD (Power Delivery) spec power packs the device and power pack "negotiate" proper voltage and current for fast charging depending on multiple factors. These small PCB trigger boards instead let you manually select output voltage (you probably are gonna want 12V I'm guessing for most devices). You can then wire this to a barrel connector or whatever an individual device uses (probably would want one battery pack and board per device). This is, of course, assuming you can power your devices directly off direct current (DC) power. If you need AC I have also had luck experimenting with the same setup outputting 12v DC to small tiny fanless AC inverters. By using an existing USB-C PD power bank you can size it appropriately for the device and time needed and also be reasonably confident that charging and discharging of the internal cells is done safely (assuming you choose a reputable power bank brand). To properly do this you would probably want a power bank with two ports so you could use one for input (battery charging) and the other for output (to the control board and then the final device). Usually these are called "USB-C PD Trigger Boards". I couldn't find the exact one I have but it has a button to select output voltage (5v, 12v, 18v, etc). No idea how efficient these are - some may use buck converters or cheaper parts or vary in quality. Others can be permanently set to a specific voltage via jumper or soldering and appear to be simpler/ smaller in design. I also found this project (I have never used it) which looks to be of possibly of higher quality (and significantly higher price): https://lectronz.com/products/pocketpd This is something similar to what I have used (I have not used this specific board - so buyer beware of course): https://a.co/d/i1mtXJw AITRIP Type-C USB-C Fast Charging Trigger Module (ZY12PDN with Screw Terminal) The people I know doing the vanlife thing also have amazing large and portable battery setups - would be worth looking at forum posts on their setups. For portable prebuilt setups (not custom LiFePO4 installs integrated into the van) I see the "EcoFlow" brand recommended a lot. I also don't know if this was specifically discussed in this thread but have you looked at the smaller Cyberpower "pure sine wave" UPS models ? They are not designed to be rack mounted but are quite compact. I have an older model and just running some network equipment at home it seems to last a few hours. I have not measured the efficiency under low loads like this though - it could be similarly terrible. Ideal world you would avoid the use of an AC inverter entirely (by using network equipment that can run off DC power directly). Does anyone know if the efficiency drops off at lower draws is because the AC inverter being less efficient at low power draw? Or is it somewhere else in the power chain ? On Fri, 11 Apr 2025 at 15:28, Javier J via NANOG <nanog@lists.nanog.org> wrote:
where do you source those batteries?
The cheapest ones that will fit in the ups from amazon with a decent rating.
I literally use a tape measure to measure the old dead lead acid batteries to see what will fit. I have found that I can get batteries with more amp hours than the lead acid replacements.
I figure even if the manufacturer is crap will outlast LA battery technology.
- J
On Fri, Apr 11, 2025 at 4:52 PM Mark Tinka via NANOG < nanog@lists.nanog.org> wrote:
On 4/11/25 17:40, Gary Sparkes wrote:
30% being a reasonable floor absolutely is true.
I didn't say 30% DoD was unreasonable. I said that claiming that going below it to 20% is dangerous is not true.
But, happy to agree to disagree.
Far less stress to go 100 to 30 and back to 100, then 90 to 20 and back to 90, etc. Keeping 30 as an operational floor lets you use full capacity as needed and remain at full functional charge with maximum lifespan retention/recovery.
That can be said of any charge/discharge window if you are not consistently discharging below a given threshold, whatever that is to
you.
EV’s charge/operate that way to extend pack lifespan primarily.
Right, but the key motivation for that is for the car manufacturers to meet warranty claims, normally at least 10 years. And they will remotely manage those charge/discharge profiles to put warranty objectives over range maximization.
4.2v float is fine if you aren’t routinely low end stressing it.
4.2V float is for NMC. LFP is usually around 3.5V float.
Yes, the internal resistance bit is true, but that really does start to kick in around 25-30. A lot of datasheets I’ve worked with talk about 30% and stress zones.
Internal resistance in Li-Ion cells is highest as the battery approaches a fully-charged or fully-discharged state. In other words, internal resistance is highest at 100% SoC and below 20% SoC. But since we know that Li-Ion batteries have a non-linear voltage curve until about 10% SoC, internal resistance is most dangerous below this SoC value.
Far better to remain at 100% float for battery lifespan than to routinely dip below 30%.
In my experience, not going below 20% SoC will be better than holding a 100% charge for an extended period of time, especially if you are not actively controlling ambient temperature.
Li-Ion batteries really do not like holding a full voltage for too long, although, for me, that would not be as bad as routinely running an SoC below 20%.
Calendar aging is not as cut and dried as it may seem. This is blatantly obvious with cellphones, of course, but holds true for lots of other implementations too.
Li-Ion batteries have a completely different use-case for cellphones than for home backup, because we prioritize capacity and peak performance for cellphones vs. home backup. This is why it is quite normal for people to expect their phone battery to be pretty "useless" after an average of 3 years.
I expect 90% capacity on 10-year-old batteries stored properly almost always, usually – at a minimum.
Again, stored batteries have no value to anyone :-).
The better metric is how to maintain working batteries for 10 years and see how much capacity you've retained by that time.
But cycle durability is what truly matters in the long run for lifespan, not calendar aging, for batteries that often see use. And 30% is a sweet spot between usable capacity and lifespan extension to often double the manufacturer’s rated cycle count.
Yes, we all want cycle durability, but calendar aging is unavoidable. And since the biggest contributor to calendar aging is ambient temperature, most owners will lose capacity due to that because they do not have active cooling for their batteries.
I disagree with 30% being a recommended DoD floor (most OEM's do not suggest that), but that's okay :-).
Remember, cycle count means you’re actually using it – so I’m not charging or discharging any less, only doing so to specific levels.
Li-Ion batteries are not cheap. People will always prioritize capacity, and the money to buy a larger pack just to save 10% from your discharge cycle does not justify the extra longevity in environments where most people will let batteries overheat.
I’d also point out a lot of research is also indicative of low discharge levels being the leading factor to degradation, not capacity float charge status. Heat is the the number one factor (outside of or at high end of design spec heat, as often seen in consumer devices). Low discharge is factor #2. High / full charge stress comes in around #3. Better to engage #3 than #2 or #1 for lifespan retention.
Maintaining a long term 100% SoC state is not problematic if you can actively cool your battery. Most owners will not, which is why this can be more damaging to your battery than a low SoC.
Remember, most batteries will spend the majority of their life closer to 100% SoC than at 1% SoC.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/S4 CZUXPXIQDP3EQSK4CQUBSDH4AMLMNP/ _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/AY 2K4WBTINZOAXS27WKXUW7LFNCC7MNT/
NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/WX5YGMYZ...
Hi Mark, You are talking a lot about how damaging temperature is to the battery. I do not remember you or Gary saying that 50 degrees Celsius is already not good. But how could it happen in the home environment?!? People like 22+-2 – they would start to do something urgently if the temperature reached 28. Home has no source for 100kWt to charge really fast, hence, the current is relatively small. Eduard From: Mark Tinka <mark@tinka.africa> Sent: Friday, April 11, 2025 23:51 To: Gary Sparkes <gary@kisaracorporation.com>; Vasilenko Eduard <vasilenko.eduard@huawei.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/11/25 17:40, Gary Sparkes wrote: 30% being a reasonable floor absolutely is true. I didn't say 30% DoD was unreasonable. I said that claiming that going below it to 20% is dangerous is not true. But, happy to agree to disagree. Far less stress to go 100 to 30 and back to 100, then 90 to 20 and back to 90, etc. Keeping 30 as an operational floor lets you use full capacity as needed and remain at full functional charge with maximum lifespan retention/recovery. That can be said of any charge/discharge window if you are not consistently discharging below a given threshold, whatever that is to you. EV’s charge/operate that way to extend pack lifespan primarily. Right, but the key motivation for that is for the car manufacturers to meet warranty claims, normally at least 10 years. And they will remotely manage those charge/discharge profiles to put warranty objectives over range maximization. 4.2v float is fine if you aren’t routinely low end stressing it. 4.2V float is for NMC. LFP is usually around 3.5V float. Yes, the internal resistance bit is true, but that really does start to kick in around 25-30. A lot of datasheets I’ve worked with talk about 30% and stress zones. Internal resistance in Li-Ion cells is highest as the battery approaches a fully-charged or fully-discharged state. In other words, internal resistance is highest at 100% SoC and below 20% SoC. But since we know that Li-Ion batteries have a non-linear voltage curve until about 10% SoC, internal resistance is most dangerous below this SoC value. Far better to remain at 100% float for battery lifespan than to routinely dip below 30%. In my experience, not going below 20% SoC will be better than holding a 100% charge for an extended period of time, especially if you are not actively controlling ambient temperature. Li-Ion batteries really do not like holding a full voltage for too long, although, for me, that would not be as bad as routinely running an SoC below 20%. Calendar aging is not as cut and dried as it may seem. This is blatantly obvious with cellphones, of course, but holds true for lots of other implementations too. Li-Ion batteries have a completely different use-case for cellphones than for home backup, because we prioritize capacity and peak performance for cellphones vs. home backup. This is why it is quite normal for people to expect their phone battery to be pretty "useless" after an average of 3 years. I expect 90% capacity on 10-year-old batteries stored properly almost always, usually – at a minimum. Again, stored batteries have no value to anyone :-). The better metric is how to maintain working batteries for 10 years and see how much capacity you've retained by that time. But cycle durability is what truly matters in the long run for lifespan, not calendar aging, for batteries that often see use. And 30% is a sweet spot between usable capacity and lifespan extension to often double the manufacturer’s rated cycle count. Yes, we all want cycle durability, but calendar aging is unavoidable. And since the biggest contributor to calendar aging is ambient temperature, most owners will lose capacity due to that because they do not have active cooling for their batteries. I disagree with 30% being a recommended DoD floor (most OEM's do not suggest that), but that's okay :-). Remember, cycle count means you’re actually using it – so I’m not charging or discharging any less, only doing so to specific levels. Li-Ion batteries are not cheap. People will always prioritize capacity, and the money to buy a larger pack just to save 10% from your discharge cycle does not justify the extra longevity in environments where most people will let batteries overheat. I’d also point out a lot of research is also indicative of low discharge levels being the leading factor to degradation, not capacity float charge status. Heat is the the number one factor (outside of or at high end of design spec heat, as often seen in consumer devices). Low discharge is factor #2. High / full charge stress comes in around #3. Better to engage #3 than #2 or #1 for lifespan retention. Maintaining a long term 100% SoC state is not problematic if you can actively cool your battery. Most owners will not, which is why this can be more damaging to your battery than a low SoC. Remember, most batteries will spend the majority of their life closer to 100% SoC than at 1% SoC. Mark.
On 4/14/25 08:45, Vasilenko Eduard wrote:
Hi Mark,
You are talking a lot about how damaging temperature is to the battery. I do not remember you or Gary saying that 50 degrees Celsius is already not good.
But how could it happen in the home environment?!?
People like 22+-2 – they would start to do something urgently if the temperature reached 28.
Home has no source for 100kWt to charge really fast, hence, the current is relatively small.
In most homes with a battery backup, it is typically installed in the garage, shed or basement if one exists. Climate control in these places is not often a priority, e.g., parking your car in your garage after being out on the road comes both with heat off the body of the car, as well as heat being dissipated by the engine, brakes, discs, exhaust pipes, e.t.c. That can easily send a closed garage's temperature up several degrees if the garage is not running any form of active cooling. Some installations in a shed will have, at least, an extractor to suck warm air out of the structure. But while that may lower the temperature, it's often not enough to keep it at 25°C, which is the most optimal for Li-Ion batteries. You don't need to charge a Li-Ion battery fast for the temperature to rise appreciably. Just charging an LFP battery at 0.25C (so 25A) can raise cell temperature by 1 - 3 degrees for every 10% of SoC gained assuming room ambient room temperature. For my installation, my battery is in my garage. In addition to fans and an extractor, I also have active cooling with an air conditioner. Going into our winter now, the a/c will no longer be needed since ambient temperatures will peak at about 10°C - 15°C, keeping the battery cells between 18°C - 25°C even during a 0.25C charge rate. But during our summer, without active cooling, the cells will live between 32°C - 35°C, which is too high. Mark.
32-35c is perfectly acceptable, not too high at all. From: Mark Tinka <mark@tinka.africa> Sent: Monday, April 14, 2025 5:53 AM To: Vasilenko Eduard <vasilenko.eduard@huawei.com>; Gary Sparkes <gary@kisaracorporation.com>; North American Network Operators Group <nanog@lists.nanog.org> Subject: Re: [NANOG] Re: Small Capacity UPS On 4/14/25 08:45, Vasilenko Eduard wrote: Hi Mark, You are talking a lot about how damaging temperature is to the battery. I do not remember you or Gary saying that 50 degrees Celsius is already not good. But how could it happen in the home environment?!? People like 22+-2 – they would start to do something urgently if the temperature reached 28. Home has no source for 100kWt to charge really fast, hence, the current is relatively small. In most homes with a battery backup, it is typically installed in the garage, shed or basement if one exists. Climate control in these places is not often a priority, e.g., parking your car in your garage after being out on the road comes both with heat off the body of the car, as well as heat being dissipated by the engine, brakes, discs, exhaust pipes, e.t.c. That can easily send a closed garage's temperature up several degrees if the garage is not running any form of active cooling. Some installations in a shed will have, at least, an extractor to suck warm air out of the structure. But while that may lower the temperature, it's often not enough to keep it at 25°C, which is the most optimal for Li-Ion batteries. You don't need to charge a Li-Ion battery fast for the temperature to rise appreciably. Just charging an LFP battery at 0.25C (so 25A) can raise cell temperature by 1 - 3 degrees for every 10% of SoC gained assuming room ambient room temperature. For my installation, my battery is in my garage. In addition to fans and an extractor, I also have active cooling with an air conditioner. Going into our winter now, the a/c will no longer be needed since ambient temperatures will peak at about 10°C - 15°C, keeping the battery cells between 18°C - 25°C even during a 0.25C charge rate. But during our summer, without active cooling, the cells will live between 32°C - 35°C, which is too high. Mark.
On 4/9/25 10:27, Gary Sparkes wrote:
The innovation is there, if you can afford it - it's still more expensive than SLA just on virtue of the cell costs, alone. Space savings may or may not be worth it to you, longevity argument may win the budget fight, it may not - sometimes spending $X every 2-3 years is a better option for cashflow reasons than spending $2*X for 5-8 years.
If your commercial power is very stable, LA will often times be cheaper than Li-Ion. But if your commercial goes out with any reasonable degree of frequency, and for extended periods, LA batteries will eventually run out of their capacity due to the frequent deep discharge. In such a case, Li-Ion will be a better investment. The longevity difference for LA batteries between stable and unstable power can be 4 - 5 years down to 1 - 2 years. Mark.
On 4/9/25 08:30, Vasilenko Eduard wrote:
I hope you would agree that it is a clumsy patch to add BMS between the UPS/Inverter and the battery. But we have no choice.
I think you meant "balancer", as the BMS always ships with pre-built batteries. If you are having to add a BMS to your cells, that is already a clumsy patch, as most people don't usually do that. Balancers are usually part of any well-built battery pack. They are only usually necessary for 12V systems. 24V and 48V systems will come with balancers built in.
Why has the industry stopped innovating?
Inverter technology is very stable and reliable, as is LFP and NMC Li-Ion technology. The BMS used to be the most fragile component of a self-generation power system, and even that is now very reliable. In 2025, it often really doesn't matter whether you buy American, European or Chinese for the inverters, as the reliability is that good. Li-Ion cells are primarily made in China, even for battery brands built in Europe or America. So there isn't much to gain there. The only thing left to innovate is battery chemistry, and looking at options that no longer rely on existing Li-Ion technologies. Things like Sodium-Ion, Lithium-Sulfur, Iron-Air, Zinc-Ion, e.t.c., is where innovation is happening. The idea is to find a battery technology that will allow them to be smaller, charge faster, run hotter and last longer.
I recently searched for inverters and/or UPSes, specifically looking for LFP support - I have found nothing. Why do we need to customize it ourselves?
Inverters are built for multiple chemistries, typically LA, AGM and Li-Ion. For Li-Ion, inverters do not generally distinguish between the different chemistries there. That comes down to the actual BMS in the battery. That said, it is not uncommon for inverters to run an LA charging profile for Li-Ion batteries where battery CAN bus communication does not exist.
I would not believe that product managers in APC (and many other vendors) have missed the existence of LFP.
They haven't. They just don't want to have to deal with the increased thermal management and charging profile risk that comes with Li-Ion vs. LA. Also, LA batteries require more frequent replacement. That is a good source of recurring revenue for UPS vendors.
I did assume that any vendor not supporting LFP our days should go out of the market very fast.
While Li-Ion battery deployment is growing rapidly (even when you remove EV's from the equation), the majority of battery deployments are still LA.
It looks like vendors are not developing anything because they are preparing for an inevitable shutdown (losing market to well-known competitors). IMHO: vendors are accelerating their death in this way.
Not really. The self-generation market is a commodity market. Solar panels, inverters and Li-Ion batteries do not cost an arm & a leg anymore. What you really pay for now is a quality engineer, a quality installation and quality support. There is a much larger premium on those things than the actual kit itself. Mark.
On Mon, Apr 07, 2025 at 06:30:42AM +0000, Vasilenko Eduard via NANOG wrote:
lithium is a)expensive, b)dangerous (it could quickly burn your house), c)would be dead in 4 years. We're talking about stationery; it's not wearable, right? Then there are many LiFePO batteries on the market. Many propositions are with a special controller that emulates lead batteries for the respective voltage (it permits connecting it to the old staff). LiFePO is much-much better on a), b), c). But has bigger weight.
Please stop spreading some FUD here, there's reputable lifepo4 systems that have inbuilt fire suppression. Yes there's a lot of cheap stuff out there, but if you are not buying directly you are in good shape with the right UL etc markings. Look at something like what they use in a solar setup, you can purchase a 14.3kW battery + inverter for ~7500 and just wire a L6-30 for your PDU as appropriate. Look at the EG4 for an example: https://signaturesolar.com/eg4-12kpv-hybrid-inverter-48v-12000w-input-8000w-... And the battery has inbuilt fire suppression system: https://eg4electronics.com/backend/wp-content/uploads/2024/03/EG4%C2%AE-Indo... Most things can catch on fire, sure.. but not everything lacks any safety systems. Either way, this is fairly plug and play, and you can get two for 1+1 setup with options to have it control generator or other power input (eg: solar) including controls to test generators etc or to charge battery if you live in an area like me where we have some of the lowest grid reliability in north america. - Jared
On 4/7/25 12:54, Jared Mauch via NANOG wrote:
And the battery has inbuilt fire suppression system:
https://eg4electronics.com/backend/wp-content/uploads/2024/03/EG4%C2%AE-Indo...
Most things can catch on fire, sure.. but not everything lacks any safety systems.
LFP will experience thermal runaway at about 270°c, while NMC will experience the same at 210°c. It's quite a lot to subject your batteries to, especially if you have a reliable BMS. On average, the failure rate of quality LFP cells is about 1 in 10 million. Mark.
Resending this without pictures because file size rejection. I'm just catching up on this thread so I have probably missed a bunch but I wanted to share my experience. I originally started to replace old batteries in my UPS's with AGM batteries instead of SLA. More recently I have been going with LiFePo4 (Lithium Iron Phosphate) and with units with 2 batteries (24V) adding in a battery balancer. examples attached. that white, yellow one is like 28 years old. running strong. my observations and research reveal that old UPS units sometimes float at a higher voltage, which makes LiFePo4 a perfect upgrade to lead acid based battery technology. Also should last forever. (10+ years and thousands of cycles) I have 5 UPSs, most I got for free. I also have a solar system in the back yard and here in Florida those LiFePo4 batteries (the cheapest I can find on Amazon at the time) are doing just great keeping all the cameras in the backyard running plus I run power tools etc from them as needed. they're not making lithium iron phosphate batteries with high current cranking capacity. next time I have a car battery to change the prices should drop low enough for that I'll be replacing it with that. On Mon, Apr 7, 2025 at 12:00 PM Mark Tinka via NANOG <nanog@lists.nanog.org> wrote:
On 4/7/25 12:54, Jared Mauch via NANOG wrote:
And the battery has inbuilt fire suppression system:
https://eg4electronics.com/backend/wp-content/uploads/2024/03/EG4%C2%AE-Indo...
Most things can catch on fire, sure.. but not everything lacks any safety systems.
LFP will experience thermal runaway at about 270°c, while NMC will experience the same at 210°c.
It's quite a lot to subject your batteries to, especially if you have a reliable BMS.
On average, the failure rate of quality LFP cells is about 1 in 10 million.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/NMSR7FYJ...
Oh, im interested with more details about your setup. I have 2x APC SmartUPS 1000 units here that I need to replace batts soon. So I slowly think about LFP batts :) This UPS needs 2x 12V 12Ah batts (RBC4). Can you recommend some LPC batts as repleacement? There is enough room in UPS to but small BMS on top of batts. APC SmartUPC 700 is much worse in that regard. I had one such unit and scrapped it. Batt temps where higher due to cramped space, and so they lifetime.. Regards, Borg PS: You can asnwer off-list w/ attachments :) ---------- Original message ---------- From: Javier J via NANOG <nanog@lists.nanog.org> To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Javier J <javier@advancedmachines.us> Subject: [NANOG] Re: Small Capacity UPS Date: Mon, 7 Apr 2025 12:03:08 -0400 Resending this without pictures because file size rejection. I'm just catching up on this thread so I have probably missed a bunch but I wanted to share my experience. I originally started to replace old batteries in my UPS's with AGM batteries instead of SLA. More recently I have been going with LiFePo4 (Lithium Iron Phosphate) and with units with 2 batteries (24V) adding in a battery balancer. examples attached. that white, yellow one is like 28 years old. running strong. my observations and research reveal that old UPS units sometimes float at a higher voltage, which makes LiFePo4 a perfect upgrade to lead acid based battery technology. Also should last forever. (10+ years and thousands of cycles) I have 5 UPSs, most I got for free. I also have a solar system in the back yard and here in Florida those LiFePo4 batteries (the cheapest I can find on Amazon at the time) are doing just great keeping all the cameras in the backyard running plus I run power tools etc from them as needed. they're not making lithium iron phosphate batteries with high current cranking capacity. next time I have a car battery to change the prices should drop low enough for that I'll be replacing it with that. On Mon, Apr 7, 2025 at 12:00˙˙PM Mark Tinka via NANOG <nanog@lists.nanog.org> wrote:
On 4/7/25 12:54, Jared Mauch via NANOG wrote:
And the battery has inbuilt fire suppression system:
https://eg4electronics.com/backend/wp-content/uploads/2024/03/EG4%C2%AE-Indo...
Most things can catch on fire, sure.. but not everything lacks any safety systems.
LFP will experience thermal runaway at about 270°c, while NMC will experience the same at 210°c.
It's quite a lot to subject your batteries to, especially if you have a reliable BMS.
On average, the failure rate of quality LFP cells is about 1 in 10 million.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/NMSR7FYJ...
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/PXYNFGUT...
Please, keep this discussion in the alias (for me to see). Thanks! Ed/ -----Original Message----- From: borg--- via NANOG <nanog@lists.nanog.org> Sent: Tuesday, April 8, 2025 10:30 To: nanog@lists.nanog.org Cc: borg@uu3.net Subject: [NANOG] Re: Small Capacity UPS Oh, im interested with more details about your setup. I have 2x APC SmartUPS 1000 units here that I need to replace batts soon. So I slowly think about LFP batts :) This UPS needs 2x 12V 12Ah batts (RBC4). Can you recommend some LPC batts as repleacement? There is enough room in UPS to but small BMS on top of batts. APC SmartUPC 700 is much worse in that regard. I had one such unit and scrapped it. Batt temps where higher due to cramped space, and so they lifetime.. Regards, Borg PS: You can asnwer off-list w/ attachments :) ---------- Original message ---------- From: Javier J via NANOG <nanog@lists.nanog.org> To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Javier J <javier@advancedmachines.us> Subject: [NANOG] Re: Small Capacity UPS Date: Mon, 7 Apr 2025 12:03:08 -0400 Resending this without pictures because file size rejection. I'm just catching up on this thread so I have probably missed a bunch but I wanted to share my experience. I originally started to replace old batteries in my UPS's with AGM batteries instead of SLA. More recently I have been going with LiFePo4 (Lithium Iron Phosphate) and with units with 2 batteries (24V) adding in a battery balancer. examples attached. that white, yellow one is like 28 years old. running strong. my observations and research reveal that old UPS units sometimes float at a higher voltage, which makes LiFePo4 a perfect upgrade to lead acid based battery technology. Also should last forever. (10+ years and thousands of cycles) I have 5 UPSs, most I got for free. I also have a solar system in the back yard and here in Florida those LiFePo4 batteries (the cheapest I can find on Amazon at the time) are doing just great keeping all the cameras in the backyard running plus I run power tools etc from them as needed. they're not making lithium iron phosphate batteries with high current cranking capacity. next time I have a car battery to change the prices should drop low enough for that I'll be replacing it with that. On Mon, Apr 7, 2025 at 12:00˙˙PM Mark Tinka via NANOG <nanog@lists.nanog.org> wrote:
On 4/7/25 12:54, Jared Mauch via NANOG wrote:
And the battery has inbuilt fire suppression system:
https://eg4electronics.com/backend/wp-content/uploads/2024/03/EG4%C2%A E-Indoor-280Ah-Battery-Specifications-Sheet.pdf
Most things can catch on fire, sure.. but not everything lacks any safety systems.
LFP will experience thermal runaway at about 270°c, while NMC will experience the same at 210°c.
It's quite a lot to subject your batteries to, especially if you have a reliable BMS.
On average, the failure rate of quality LFP cells is about 1 in 10 million.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/NM SR7FYJX472CXEMUUQEYJNJJJJUWQG3/
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/PXYNFGUT... _______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/BOT5ZVIX...
On 4/8/25 09:29, borg--- via NANOG wrote:
Oh, im interested with more details about your setup. I have 2x APC SmartUPS 1000 units here that I need to replace batts soon. So I slowly think about LFP batts :)
This UPS needs 2x 12V 12Ah batts (RBC4). Can you recommend some LPC batts as repleacement? There is enough room in UPS to but small BMS on top of batts. APC SmartUPC 700 is much worse in that regard. I had one such unit and scrapped it. Batt temps where higher due to cramped space, and so they lifetime..
12V Li-Ion batteries will come with a BMS installed as standard. If you can, buy the ones with a Bluetooth controller included. They are usually drop-ins, so they'll be the same size as the LA ones you are replacing. What you really want the space for is the balancer. Mark.
If you don't mind building you own UPS from COTS parts, you may want to look at Xantrex (https://xantrex.com/). I have one of their UPS units (the FREEDOM XC PRO 2000) in an RV with 2 100AH LiFePo4 batteries in parallel. I've considered using them for IT deployments but haven't pulled the trigger just yet. On Tue, Apr 8, 2025 at 2:30 AM borg--- via NANOG <nanog@lists.nanog.org> wrote:
Oh, im interested with more details about your setup. I have 2x APC SmartUPS 1000 units here that I need to replace batts soon. So I slowly think about LFP batts :)
This UPS needs 2x 12V 12Ah batts (RBC4). Can you recommend some LPC batts as repleacement? There is enough room in UPS to but small BMS on top of batts. APC SmartUPC 700 is much worse in that regard. I had one such unit and scrapped it. Batt temps where higher due to cramped space, and so they lifetime..
Regards, Borg
PS: You can asnwer off-list w/ attachments :)
---------- Original message ----------
From: Javier J via NANOG <nanog@lists.nanog.org> To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Javier J <javier@advancedmachines.us> Subject: [NANOG] Re: Small Capacity UPS Date: Mon, 7 Apr 2025 12:03:08 -0400
Resending this without pictures because file size rejection.
I'm just catching up on this thread so I have probably missed a bunch but I wanted to share my experience. I originally started to replace old batteries in my UPS's with AGM batteries instead of SLA. More recently I have been going with LiFePo4 (Lithium Iron Phosphate) and with units with 2 batteries (24V) adding in a battery balancer.
examples attached. that white, yellow one is like 28 years old. running strong.
my observations and research reveal that old UPS units sometimes float at a higher voltage, which makes LiFePo4 a perfect upgrade to lead acid based battery technology. Also should last forever. (10+ years and thousands of cycles)
I have 5 UPSs, most I got for free. I also have a solar system in the back yard and here in Florida those LiFePo4 batteries (the cheapest I can find on Amazon at the time) are doing just great keeping all the cameras in the backyard running plus I run power tools etc from them as needed. they're not making lithium iron phosphate batteries with high current cranking capacity. next time I have a car battery to change the prices should drop low enough for that I'll be replacing it with that.
On Mon, Apr 7, 2025 at 12:00˙˙PM Mark Tinka via NANOG < nanog@lists.nanog.org> wrote:
On 4/7/25 12:54, Jared Mauch via NANOG wrote:
And the battery has inbuilt fire suppression system:
https://eg4electronics.com/backend/wp-content/uploads/2024/03/EG4%C2%AE-Indo...
Most things can catch on fire, sure.. but not everything lacks any safety systems.
LFP will experience thermal runaway at about 270°c, while NMC will experience the same at 210°c.
It's quite a lot to subject your batteries to, especially if you have a reliable BMS.
On average, the failure rate of quality LFP cells is about 1 in 10 million.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/NMSR7FYJ... _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/PXYNFGUT... _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/BOT5ZVIX...
I'm actually a sizable fan of cheap but decent quality used commercial UPSes, for a lot of scenarios like this. In my office gear/rack, I have two tripplite SMART1500CRMXL that I got used when a donation was made to a location from an office decommission back in *2011* - so having already long been in service - and the only thing I've had to do was batteries. (the much older model than currently under those part numbers, with just LED indicators instead of the LCD - though it does take an SNMP card) - https://tripplite.eaton.com/support/SMART1500CRMXL And for that, the recent deluge of drop-in LiFePo4 cells has been great. Instead of the design three 12v 9Ah internally, it took well to pre-made 12v 10Ah lifepo4 cells. Beyond that, it has your standard SB50 for expansion battery pack runtime, which was easy enough to wire up an external pack with a "cheap" BMS and a string of LiFePo4 cells, giving me 105Ah at 36V all said and done for the external pack.... I've even gone so far for network gear as to taking a small Belkin desk-side UPS and drilling out a hole into its casing and running an external battery pack on the side. Instead of the built in 2x 12v 5Ah, it suddenly had 4x 6V 50Ah cells. That thing ran a few switches for a very long time, though I never took any measurements, before one of the batteries gave out I had a switch going on 5 years uptime in that setup, with multiple extended power outages in that timeframe (I'd re-charge it off generator to full at least once a day, and leave it on if I could, but I never got anywhere near testing the limits of it from just raw voltage checks) Obviously, not solutions for bigger and more enterprisey scenarios, but...... options can be had. And with the battery being wired up externally, adding something like a regulated solar output to the battery terminals directly would be able to keep the system alive/running longer and re-charge faster too. -----Original Message----- From: Dave Phelps via NANOG <nanog@lists.nanog.org> Sent: Tuesday, April 8, 2025 9:06 AM To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Dave Phelps <tippenring@gmail.com> Subject: [NANOG] Re: Small Capacity UPS If you don't mind building you own UPS from COTS parts, you may want to look at Xantrex (https://xantrex.com/). I have one of their UPS units (the FREEDOM XC PRO 2000) in an RV with 2 100AH LiFePo4 batteries in parallel. I've considered using them for IT deployments but haven't pulled the trigger just yet. On Tue, Apr 8, 2025 at 2:30 AM borg--- via NANOG <nanog@lists.nanog.org> wrote:
Oh, im interested with more details about your setup. I have 2x APC SmartUPS 1000 units here that I need to replace batts soon. So I slowly think about LFP batts :)
This UPS needs 2x 12V 12Ah batts (RBC4). Can you recommend some LPC batts as repleacement? There is enough room in UPS to but small BMS on top of batts. APC SmartUPC 700 is much worse in that regard. I had one such unit and scrapped it. Batt temps where higher due to cramped space, and so they lifetime..
Regards, Borg
PS: You can asnwer off-list w/ attachments :)
---------- Original message ----------
From: Javier J via NANOG <nanog@lists.nanog.org> To: North American Network Operators Group <nanog@lists.nanog.org> Cc: Javier J <javier@advancedmachines.us> Subject: [NANOG] Re: Small Capacity UPS Date: Mon, 7 Apr 2025 12:03:08 -0400
Resending this without pictures because file size rejection.
I'm just catching up on this thread so I have probably missed a bunch but I wanted to share my experience. I originally started to replace old batteries in my UPS's with AGM batteries instead of SLA. More recently I have been going with LiFePo4 (Lithium Iron Phosphate) and with units with 2 batteries (24V) adding in a battery balancer.
examples attached. that white, yellow one is like 28 years old. running strong.
my observations and research reveal that old UPS units sometimes float at a higher voltage, which makes LiFePo4 a perfect upgrade to lead acid based battery technology. Also should last forever. (10+ years and thousands of cycles)
I have 5 UPSs, most I got for free. I also have a solar system in the back yard and here in Florida those LiFePo4 batteries (the cheapest I can find on Amazon at the time) are doing just great keeping all the cameras in the backyard running plus I run power tools etc from them as needed. they're not making lithium iron phosphate batteries with high current cranking capacity. next time I have a car battery to change the prices should drop low enough for that I'll be replacing it with that.
On Mon, Apr 7, 2025 at 12:00˙˙PM Mark Tinka via NANOG < nanog@lists.nanog.org> wrote:
On 4/7/25 12:54, Jared Mauch via NANOG wrote:
And the battery has inbuilt fire suppression system:
https://eg4electronics.com/backend/wp-content/uploads/2024/03/EG4%C2%A E-Indoor-280Ah-Battery-Specifications-Sheet.pdf
Most things can catch on fire, sure.. but not everything lacks any safety systems.
LFP will experience thermal runaway at about 270°c, while NMC will experience the same at 210°c.
It's quite a lot to subject your batteries to, especially if you have a reliable BMS.
On average, the failure rate of quality LFP cells is about 1 in 10 million.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/NM SR7FYJX472CXEMUUQEYJNJJJJUWQG3/ _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/PX YNFGUTBFBZJE6QFA4ZDC52QPJMKMZC/ _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/BO T5ZVIXAQCTFOLPZDIVWXTDMR64P3E2/
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/Q3MA3OKZ...
On Tue, Apr 8, 2025 at 6:06 AM Dave Phelps via NANOG <nanog@lists.nanog.org> wrote:
If you don't mind building you own UPS from COTS parts, you may want to look at Xantrex (https://xantrex.com/). I have one of their UPS units (the FREEDOM XC PRO 2000) in an RV with 2 100AH LiFePo4 batteries in parallel.
Yes, an integrated (inverter + charger + power supply) connected to batteries is the simplest, most compact, plug and play. Another example product: https://www.aimscorp.net/product/2000-watt-pure-sine-inverter-charger-24-vol... Options exist to additionally combine solar MPPT inputs. Sealed Lead Acid AGM 12v 100Ah batteries still work, I don't rule them out. If you can avoid converting to AC, then you avoid the inefficiency of an inverter running with undersized loads. You need an integrated (charger + power supply), Iota is very good. https://www.iotaengineering.com/products/dls-power-converters-battery-charge... Call Iota and speak with an engineer to correctly size the charger/batteries and 12/24/48v, I learned a lot in 5 minutes. A power supply+charger can charge the battery all the way up, drop to a maintenance voltage, and still source all the current to feed your gear without overcharging the battery, similar to an alternator in a car. The power supply must have capacity to source enough current at maintenance voltage. A device like this can replace AC power supplies for ATX boards: https://www.mini-box.com/DC-DC Here is an excellent 1 port POE Injector with 12v UPS, connect a battery of your choice and enjoy: https://mikrotik.com/product/mups Consider using a battery balancer for arrays with many batteries. https://www.victronenergy.com/batteries/battery-balancer
Wanted to reply to this thread and add the pictures I couldn't send before, but here is an example of what I am doing with APC UPSs I usually get for free. People throw them out when the batteries die. https://photos.app.goo.gl/xUiz28cDtwTygw2m9 With 2x 12 volt LiFePo4 batteries, you should have a balancer like the one in the black 1500 UPS. I drilled a hole and used 3m double sided tape to mount it. I like that I can see the voltage of each battery. All these lithium based batteries have an internal BMS (battery management system) but if you put 2 or more in series, you want a balancer as well so they don't fall out of line. I still have an old UPS with 2 external AGM batteries, I modified it to use the battery disconnect as the battery plug. I use a balancer there as well. 7 years still working good for my needs. When the batteries go, ill replace them with LifePo4. Pay attention to the BMS specs on the batteries to make sure it is powerful enough for your use case. So far I have not had any trouble with my upgrade to these old / free UPSs. Anything I can use that doesn't end up in a landfill is good for our children, and our children's children. On Tue, Apr 8, 2025 at 2:29 PM Dylan Ambauen via NANOG < nanog@lists.nanog.org> wrote:
On Tue, Apr 8, 2025 at 6:06 AM Dave Phelps via NANOG < nanog@lists.nanog.org> wrote:
If you don't mind building you own UPS from COTS parts, you may want to look at Xantrex (https://xantrex.com/). I have one of their UPS units (the FREEDOM XC PRO 2000) in an RV with 2 100AH LiFePo4 batteries in parallel.
Yes, an integrated (inverter + charger + power supply) connected to batteries is the simplest, most compact, plug and play. Another example product:
https://www.aimscorp.net/product/2000-watt-pure-sine-inverter-charger-24-vol... Options exist to additionally combine solar MPPT inputs. Sealed Lead Acid AGM 12v 100Ah batteries still work, I don't rule them out.
If you can avoid converting to AC, then you avoid the inefficiency of an inverter running with undersized loads. You need an integrated (charger + power supply), Iota is very good.
https://www.iotaengineering.com/products/dls-power-converters-battery-charge... Call Iota and speak with an engineer to correctly size the charger/batteries and 12/24/48v, I learned a lot in 5 minutes. A power supply+charger can charge the battery all the way up, drop to a maintenance voltage, and still source all the current to feed your gear without overcharging the battery, similar to an alternator in a car. The power supply must have capacity to source enough current at maintenance voltage.
A device like this can replace AC power supplies for ATX boards: https://www.mini-box.com/DC-DC Here is an excellent 1 port POE Injector with 12v UPS, connect a battery of your choice and enjoy: https://mikrotik.com/product/mups Consider using a battery balancer for arrays with many batteries. https://www.victronenergy.com/batteries/battery-balancer _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2SJFSA...
On Tue, Apr 8, 2025 at 11:20 PM Javier J via NANOG <nanog@lists.nanog.org> wrote:
but if you put 2 or more in series, you want a balancer as well so they don't fall out of line.
There are some vendors (for example, PowerSonic with their PSL-SC line) which design their LiFoPo4 batteries to be direct replacements for classic SLA batteries, even for high discharge applications such as UPS's, for both parallel and series applications. As with all else, do your own testing in your own lab.
One thing to think about when designing a UPS solution (constant on type) is that your AC->DC supply must be able to carry your max load + the max charge load. Think about instances where you’re running on batteries for a sufficient period that your charge current drain may be equal to your actual steady-state load. -- John Fraizer LinkedIn profile: http://www.linkedin.com/in/johnfraizer/ On Tue, Apr 8, 2025 at 2:29 PM Dylan Ambauen via NANOG < nanog@lists.nanog.org> wrote:
On Tue, Apr 8, 2025 at 6:06 AM Dave Phelps via NANOG < nanog@lists.nanog.org> wrote:
If you don't mind building you own UPS from COTS parts, you may want to look at Xantrex (https://xantrex.com/). I have one of their UPS units (the FREEDOM XC PRO 2000) in an RV with 2 100AH LiFePo4 batteries in parallel.
Yes, an integrated (inverter + charger + power supply) connected to batteries is the simplest, most compact, plug and play. Another example product:
https://www.aimscorp.net/product/2000-watt-pure-sine-inverter-charger-24-vol... Options exist to additionally combine solar MPPT inputs. Sealed Lead Acid AGM 12v 100Ah batteries still work, I don't rule them out.
If you can avoid converting to AC, then you avoid the inefficiency of an inverter running with undersized loads. You need an integrated (charger + power supply), Iota is very good.
https://www.iotaengineering.com/products/dls-power-converters-battery-charge... Call Iota and speak with an engineer to correctly size the charger/batteries and 12/24/48v, I learned a lot in 5 minutes. A power supply+charger can charge the battery all the way up, drop to a maintenance voltage, and still source all the current to feed your gear without overcharging the battery, similar to an alternator in a car. The power supply must have capacity to source enough current at maintenance voltage.
A device like this can replace AC power supplies for ATX boards: https://www.mini-box.com/DC-DC Here is an excellent 1 port POE Injector with 12v UPS, connect a battery of your choice and enjoy: https://mikrotik.com/product/mups Consider using a battery balancer for arrays with many batteries. https://www.victronenergy.com/batteries/battery-balancer _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2SJFSA...
On 4/6/25 14:55, Mike Hammett via NANOG wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
This is still cheaper than doing a DC plant. A basic telco rectifier, fuse panel and batteries will be in the 5k+ range. Do you need network monitoring too? A UPS will have a standard SNMP interface which can be a huge plus. At my house I had a 400w load between routers, cameras, telco, and was able to get a 240v APC UPS which was rated at 6kva/5kw. The issue was, this equipment all ran with .7 power factor at best, which is about 550 kVA. I discovered the UPS has a 16s of 12v 5Ah batteries with no real balancing between them. I can see why you have to replace them every 3 years in normal service. What I did was rework the external connector to an external string of 16 50a (9.6kw) and added a BMS to level the charge between the batteries. This would give me about 7 hours to 20%. Recharging took about a day after this rundown. I had about 800 for the UPS, 1200 in batteries, 330 for the BMS, and about 250 for cables/connectors/etc to set it up. $2,600 total, I'm sure it would be a bit more today given inflation. In general it worked well and provided the runtime needed, but it was a bit bulky. The real advantage is the lifetime extension of the batteries, as at full load the battery would see a 26 amp draw, which on a 50a/h battery is .5C discharge rate, vs 5.2c on the stock pack. At my load this was .1C, which is just a trickle, so between this and the BMS, I greatly expanded my battery life, and expected 5-7 years out of the SLA. As I had extra power on this, I ran a 10/3 romex with an L14-30 inlet from the network closet to my office and then powered the office from 240v UPS. This worked very well, and the breaker in the UPS tripped when the water flooded my office, but still kept the network up in the main house which was at a higher level. Eventually we were able to watch empty 5 gallon water jugs float down the hallway and until the water rose over batteries which were on the floor. IMHO you can't beat a used APC UPS with an external pack. -- Bryan Fields 727-409-1194 - Voice http://bryanfields.net
Not trying to hijack your sourcing/design discussion but: In your maintainability calculation, you also need to think about battery replacement/maintenance/repair. You are going to need to roll a truck to every location every few years to replace batteries. Who pays for that? If your service fails during a power outage due to a bad battery you have to explain that. What is your SLA on battery runtime? What happens if you roll a truck and the customer has a blown circuit breaker? Alarm companies have two different battles: Cellular technology upgrades and battery replacement cycles. Often they have to eat the cellular upgrades (Like 2G and 3G sunsets, and at some point 4G will happen) Since monitoring depends on the cellular radio. They can roll that into a contract renewal like “Sign a two year renewal contract and we will replace ‘Your’ cellular radio for free”, other wise, your system will not be monitored after this date. I have seem some alarm companies offer to sell you the batteries (or you can probably source them online) and you swap it out yourself, or they charge for a service call and bill you for the battery as part of that. Just something to think about if you haven’t already done so! Hope this helps G On 6 Apr 2025, at 13:55, Mike Hammett via NANOG wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com
Midwest-IX http://www.midwest-ix.com
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/7P2J2YK3...
Hi,
On 6 Apr 2025, at 20:55, Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
For places which are not proper IT cabinets, I’d go with something like https://us.ecoflow.com/ - most (if not all) support charging while output is on, and you get the extra benefit of being able to add a solar panel if you want to. Not sure about the efficiency though. BR, ic
ic, Alas, although some of these units advertise this capability, they don’t reliably operate this way. I’ve tried several brands as solar-charged UPSes at remote radio antenna sites, and all eventually failed within just a couple months of the batteries didn’t make it through a long gray spell. In my experience, they may initially work as a UPS for a few power outage cycles, but then suddenly fail permanently with burned components. Some vendors actually say operating as a UPS — drawing power while charging — voids the warranty, despite appearing to work. For mission-critical operations, it’s best to use a name-brand self-contained UPS designed for the purpose. In a small space you won’t get more than an hour or two of runtime, but that’s the physics we’re stuck with at this time. -mel via cell
On Apr 8, 2025, at 5:30 AM, ic via NANOG <nanog@lists.nanog.org> wrote:
Hi,
On 6 Apr 2025, at 20:55, Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
For places which are not proper IT cabinets, I’d go with something like https://us.ecoflow.com/ - most (if not all) support charging while output is on, and you get the extra benefit of being able to add a solar panel if you want to.
Not sure about the efficiency though.
BR, ic
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HSXYBNXQ...
Agreed. In particular I've tried to use EcoFlow for this task and none have ever survived a year in service. Some got an initial repair, but all failed again just out of warranty and ended up discarded. -Dorn On Tue, Apr 8, 2025 at 7:57 AM Mel Beckman via NANOG <nanog@lists.nanog.org> wrote:
ic,
Alas, although some of these units advertise this capability, they don’t reliably operate this way. I’ve tried several brands as solar-charged UPSes at remote radio antenna sites, and all eventually failed within just a couple months of the batteries didn’t make it through a long gray spell.
In my experience, they may initially work as a UPS for a few power outage cycles, but then suddenly fail permanently with burned components. Some vendors actually say operating as a UPS — drawing power while charging — voids the warranty, despite appearing to work.
For mission-critical operations, it’s best to use a name-brand self-contained UPS designed for the purpose. In a small space you won’t get more than an hour or two of runtime, but that’s the physics we’re stuck with at this time.
-mel via cell
On Apr 8, 2025, at 5:30 AM, ic via NANOG <nanog@lists.nanog.org> wrote:
Hi,
On 6 Apr 2025, at 20:55, Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-)
For places which are not proper IT cabinets, I’d go with something like https://us.ecoflow.com/ - most (if not all) support charging while output is on, and you get the extra benefit of being able to add a solar panel if you want to.
Not sure about the efficiency though.
BR, ic
_______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/HSXYBNXQ... _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/TLTIGR22...
On 4/8/25 15:03, Dorn Hetzel via NANOG wrote:
Agreed.
In particular I've tried to use EcoFlow for this task and none have ever survived a year in service. Some got an initial repair, but all failed again just out of warranty and ended up discarded.
These Ecoflow jobs seem ideal for a drive out to the camp for a night. And maybe as emergency backup if your main backup fails. I wouldn't use them for prime time. Mark.
Maybe https://www.mini-box.com/OpenUPS is worth mentioning? I haven't used it YET, but it looks to be capable. On Tue, Apr 8, 2025 at 9:15 AM Mark Tinka via NANOG <nanog@lists.nanog.org> wrote:
On 4/8/25 15:03, Dorn Hetzel via NANOG wrote:
Agreed.
In particular I've tried to use EcoFlow for this task and none have ever survived a year in service. Some got an initial repair, but all failed again just out of warranty and ended up discarded.
These Ecoflow jobs seem ideal for a drive out to the camp for a night. And maybe as emergency backup if your main backup fails.
I wouldn't use them for prime time.
Mark. _______________________________________________ NANOG mailing list
https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/2EP36OAK...
On 8 Apr 2025, at 14:57, Mel Beckman <mel@beckman.org> wrote:
ic,
Alas, although some of these units advertise this capability, they don’t reliably operate this way. I’ve tried several brands as solar-charged UPSes at remote radio antenna sites, and all eventually failed within just a couple months of the batteries didn’t make it through a long gray spell.
In my experience, they may initially work as a UPS for a few power outage cycles, but then suddenly fail permanently with burned components. Some vendors actually say operating as a UPS — drawing power while charging — voids the warranty, despite appearing to work.
Interesting, thanks for your feedback. Also overhead of 41W sounds like a feal breaker. BR, ic
On 4/8/25 14:29, ic via NANOG wrote:
Hi,
On 6 Apr 2025, at 20:55, Mike Hammett via NANOG <nanog@lists.nanog.org> wrote:
I'm trying to find something that keeps my customer's network gear online for a meaningful amount of time. The challenge is that an ONT, firewall, switch, AP, and some IP phones doesn't add up to be very much load. Most normal UPSes get terribly inefficient at lower load ratings. Add up all of the network devices a customer may have and we rarely break 50 watts of load. Normal, small UPSes are lucky to break 50% efficiency at those loads whereas they may be 95% efficient at say 100 or 200 watts. Get a bigger unit with a bigger battery and now you're even less efficient. Get a big enough unit to have extendable batteries and now you're spending thousands of dollars for such a small request.
I've gone asking, but haven't really gotten anywhere. The best technical solution was from some electronics parts nerds that was basically to build my own small rectifier and battery system. Great. I can achieve high efficiencies with small loads, letting me have say 4 or 8 hours of battery. However, I've got a science project, not something I can deploy at a customer.
I'm hoping one of you has the magic bullet in what product a service provider should use in this scenario.
Oh, and of course, being able to centrally manage them from my own iron would be great too. :-) For places which are not proper IT cabinets, I’d go with something like https://us.ecoflow.com/ - most (if not all) support charging while output is on, and you get the extra benefit of being able to add a solar panel if you want to.
I recall spending some time reviewing the Ecoflow Delta 2, which was about US$799 a year ago, but now seems to be going for US$479 today. Not bad for a solar-supported, pre-built 1.2kWh LFP battery system: https://us.ecoflow.com/products/delta-2-smart-extra-battery?variant=40573812...
Not sure about the efficiency though.
Not very good, from a number of reports I found of people who bought it. Numbers are in the range of 83% @ 200W to 90% @ 1kW. Not very good numbers. Worse if you have a lower load, because feedback is that this unit has an operating overhead of 41W, which is super high. That said, may just be what the OP is looking for. Mark.
participants (30)
-
borg＠uu3.net -
Brandon Butterworth -
Brandon Svec -
Bryan Fields -
Christopher I. Sinclair -
Colin Constable -
Craig Smith -
craigesmith＠gmail.com
-
Dave Phelps -
Dorn Hetzel -
Dylan Ambauen -
Eric C. Miller -
Gary Buhrmaster -
Gary Sparkes -
Gerry Boudreaux -
ic -
James R Cutler -
Jared Mauch -
Javier J -
John Fraizer -
jordi.palet＠consulintel.es -
Mark Tinka -
Mel Beckman -
Mike Hammett -
Tim Burke -
TJ Trout -
Tom Beecher -
Travis Garrison -
Vasilenko Eduard -
William Herrin