POS OC48 interfaces using different wave length
Hello all, I am thinking about using different POS interfaces on dark-fibers. On one side, i already have POS-OC48-LR card (Cisco). This card use 1550 nm wave length. We would like to use a POS-OC48-SR (Cisco). This card use 1310 nm wave length. Are 1310 and 1550 optics comptatible ? If yes, i guess that there is impact on the power Budget available. How-Much ?! Some colleagues reported successful implementation with such config (Cisco POS OC48 Cards) It looks strange to me and i wonder if Juniper and Cisco POS OC48 would also work corretly in such environnement ... I guess that "official" answer from manufacturer would be : NO WAY :-) Some carriers may already tested such config ? Any feedback ? Thanks. V
I am thinking about using different POS interfaces on dark-fibers.
On one side, i already have POS-OC48-LR card (Cisco). This card use 1550 nm wave length.
We would like to use a POS-OC48-SR (Cisco). This card use 1310 nm wave length.
Are 1310 and 1550 optics comptatible ?
You have to have the same wavelength at both ends, otherwise it won't work.
If yes, i guess that there is impact on the power Budget available. How-Much ?!
LR tends to be 1550nm, because 1550nm has less signal loss over the same length of fiber. Simon -- Simon Lockhart | Tel: +44 (0)1737 839676 Internet Engineering Manager | Fax: +44 (0)1737 839516 BBC Internet Services | Email: Simon.Lockhart@bbc.co.uk Kingswood Warren,Tadworth,Surrey,UK | URL: http://support.bbc.co.uk/
So I guess the right answer here is that everyone needs to learn how to use attenuators propperly... On Thu, Jun 07, 2001 at 05:45:21PM +0100, Simon Lockhart wrote:
I am thinking about using different POS interfaces on dark-fibers.
On one side, i already have POS-OC48-LR card (Cisco). This card use 1550 nm wave length.
We would like to use a POS-OC48-SR (Cisco). This card use 1310 nm wave length.
Are 1310 and 1550 optics comptatible ?
You have to have the same wavelength at both ends, otherwise it won't work.
If yes, i guess that there is impact on the power Budget available. How-Much ?!
LR tends to be 1550nm, because 1550nm has less signal loss over the same length of fiber.
Simon -- Simon Lockhart | Tel: +44 (0)1737 839676 Internet Engineering Manager | Fax: +44 (0)1737 839516 BBC Internet Services | Email: Simon.Lockhart@bbc.co.uk Kingswood Warren,Tadworth,Surrey,UK | URL: http://support.bbc.co.uk/
Unless I've severely forgotten most of my physics classes, no; attenuators reduce a signal's strength so as to not overload the input stage of the receiver, if the signal is that strong; or, as mentioned later, to simulate a long-distance run over a short length of fiber. 1310nm and 1550nm wavelengths are entirely different frequencies (not amplitudes) of light. The only way to convert is to use an active repeater that has the one type of transceiver on one side, and the other on the opposite side, or to use Acme Singularity Inc's black hole with ST connectors in a nifty in-line package to slow down the light wave enough (and speed it up in the other direction) that it shifts wavelengths. ( }:P ) -dalvenjah On Thu, Jun 07, 2001 at 10:08:46AM -0700, Wayne Bouchard put this into my mailbox:
So I guess the right answer here is that everyone needs to learn how to use attenuators propperly...
On Thu, Jun 07, 2001 at 05:45:21PM +0100, Simon Lockhart wrote:
I am thinking about using different POS interfaces on dark-fibers.
On one side, i already have POS-OC48-LR card (Cisco). This card use 1550 nm wave length.
We would like to use a POS-OC48-SR (Cisco). This card use 1310 nm wave length.
Are 1310 and 1550 optics comptatible ?
-- Dalvenjah FoxFire (aka Sven Nielsen) Heard recently: "I don't care if you Founder (and CEO) ARE a CEO, nobody under 21 gets in." The DALnet IRC Network e-mail: dalvenjah@dal.net WWW: http://www.dal.net/~dalvenjah/ whois: SN90 Try DALnet! http://www.dal.net/
On Thu, 7 Jun 2001, Wayne Bouchard wrote:
So I guess the right answer here is that everyone needs to learn how to use attenuators propperly...
Attenuation has nothing to do with the reason the two cards won't work together. Think of the cards as color blind. They can only see ONE color. They ignore all other colors. 1550nm and 1310nm are different colors. --- John Fraizer EnterZone, Inc
On Fri, 8 Jun 2001, John Fraizer wrote:
Attenuation has nothing to do with the reason the two cards won't work together. Think of the cards as color blind. They can only see ONE color. They ignore all other colors.
Have you actually tried this? I have personally gotten linkups between 1310 and 1550 GBICs. Just as someone said here before, the receivers are very wide-band.
1550nm and 1310nm are different colors.
Yes, but if your receiver picks up photons in a range from (just guessing) 1200 to 1700nm light, it doesn't matter. -- Mikael Abrahamsson email: swmike@swm.pp.se
On Fri, 8 Jun 2001, Mikael Abrahamsson wrote:
On Fri, 8 Jun 2001, John Fraizer wrote:
Attenuation has nothing to do with the reason the two cards won't work together. Think of the cards as color blind. They can only see ONE color. They ignore all other colors.
Have you actually tried this? I have personally gotten linkups between 1310 and 1550 GBICs. Just as someone said here before, the receivers are very wide-band.
1550nm and 1310nm are different colors.
Yes, but if your receiver picks up photons in a range from (just guessing) 1200 to 1700nm light, it doesn't matter.
Receivers convert light into electricity when a photon bumps an electron through the semiconductor band gap. There is a sharp cutoff at the low energy side. (If the photon energy is less than the band gap energy, it can't push the electron to the higher energy state.) The cutoff on the high energy side is very broad. Since shorter wavelengths means higher energy photons, the 1310 nm transmitter will almost certainly work with the 1550 nm receiver. It is possible that the 1310 nm receiver might not work with the 1550 nm transmitter. This is because the efficiency of the reciever is best near the band gap of the material. It is possible to choose the material and do some other structural things (multiple quantum wells) to the semiconductor to boost the sensitivity at the desired wavelenth. The manufacturer may have elected to use a receiver that was optimized for 1300 nm. In that case, it may have no sensitivity to 1510 nm signal. But it would still work for 800 nm. BTW, the attenuators are only important for making sure you don't burn out a receiver when you test the unsupported configuration. -steve
Hello all,
[ snip ]
Are 1310 and 1550 optics comptatible ?
Can humans see ultraviolet light? These are two different wavelengths. The equipment is specifically designed and built to _not_ allow interaction between different wavelengths. One can run 1310 and 1550 on the same fiber with WDM, but 1310 and 1550 won't talk to each other. Eddy --------------------------------------------------------------------------- Brotsman & Dreger, Inc. EverQuick Internet Division Phone: +1 (316) 794-8922 Wichita/(Inter)national Phone: +1 (785) 865-5885 Lawrence ---------------------------------------------------------------------------
Are 1310 and 1550 optics comptatible ?
Can humans see ultraviolet light? These are two different wavelengths. The equipment is specifically designed and built to _not_ allow interaction between different wavelengths.
One can run 1310 and 1550 on the same fiber with WDM, but 1310 and 1550 won't talk to each other.
As someone already pointed out, the receivers are often wide enough to see both wavelengths. (Cheaper to build that way.) So it might work, depending on the loss and dispersion characteristics of the specific fiber installation. But I'd be surprised if Cisco would actually support that configuration... An WDM demultiplexer typically has a passive narrowband optical filter in front of the receiver to allow only the desired wavelength through. Steve
On Thu, 7 Jun 2001, Steve Feldman wrote:
Are 1310 and 1550 optics comptatible ?
Can humans see ultraviolet light? These are two different wavelengths. The equipment is specifically designed and built to _not_ allow interaction between different wavelengths.
One can run 1310 and 1550 on the same fiber with WDM, but 1310 and 1550 won't talk to each other.
As someone already pointed out, the receivers are often wide enough to see both wavelengths. (Cheaper to build that way.)
So it might work, depending on the loss and dispersion characteristics of the specific fiber installation. But I'd be surprised if Cisco would actually support that configuration...
An WDM demultiplexer typically has a passive narrowband optical filter in front of the receiver to allow only the desired wavelength through.
Steve
The receivers are probably wide enough to see both wavelengths, but maybe not. The 1550 transmitter might not be seen by the 1310 receiver, depending on choices made by the manufacturer. You might need to use different attenuators in the two directions. (You are planning to use two fibers, right?) You should do no harm by testing it, but check the transmit powers before you do. Make sure that you add attenuators if one of the transmitters is spec'd at a higher output power than the other. I come from an optics background and did some optoelectronics. My gut tells me it will probably work. Steve Schaefer Dashbit - The Leader In Internet Topology www.dashbit.com www.traceloop.com
You can do that. The issue really is whether the receivers are going to accept signals in the other band and you will find that they will. You can also do el-cheapo CWDM (CoarseWDM) using inexpensive filters so you can run a 1310 and a 1550 at one end to their respective pair-mates at the opposite end to get 2 seperate signals on the same fibers without full-blown and expensive DWDM gear. One color could be POS and the other GIG-E, if that is handier. A filter to split 1310 from 1550 is inexpensive campared to ITU grid spacing. Nothing labeled xWDM needed except perhaps the simple filters you need at each end. In a closer to each other color wise but still mismatched situation, if you happen to be using cisco's 15454 family (Cerent) ITU grid OC-48 cards not for their different colors, but for their much higher power than the plain LR card, any two different colors cards you happen to have on hand will play fine to each other - just don't put any filters inbetween. Also note that some of cisco's published GBIC configurations include cross band applications. ----- Original Message ----- From: "Vincent Gillet" <vgi@opentransit.net> To: <nanog@nanog.org> Sent: Thursday, June 07, 2001 9:39 AM Subject: POS OC48 interfaces using different wave length
Hello all,
I am thinking about using different POS interfaces on dark-fibers.
On one side, i already have POS-OC48-LR card (Cisco). This card use 1550 nm wave length.
We would like to use a POS-OC48-SR (Cisco). This card use 1310 nm wave length.
Are 1310 and 1550 optics comptatible ? If yes, i guess that there is impact on the power Budget available.
How-Much ?!
Some colleagues reported successful implementation with such config (Cisco POS OC48 Cards) It looks strange to me and i wonder if Juniper and Cisco POS OC48 would also work corretly in such environnement ...
I guess that "official" answer from manufacturer would be : NO WAY :-) Some carriers may already tested such config ? Any feedback ?
Thanks.
V
participants (10)
-
Barton F Bruce -
Dalvenjah FoxFire -
E.B. Dreger -
John Fraizer -
Mikael Abrahamsson -
Simon Lockhart -
Steve Feldman -
Steve Schaefer -
Vincent Gillet -
Wayne Bouchard