Maximum Length for 1000Base-SX
Bookmark this page:
 Yahoo!
 Google
 Windows Live
 del.icio.us
 digg
 Netscape
|
|
|||||||||||||
|
Posted by Wrolf on September 14, 2006, 6:10 pm
Please log in for more thread options Can I calculate the maximum run length for 1000Base-SX based on: Max dB Loss (7.5dB) > # patch panels * loss per mated pair + distance
attenuation
Where loss per mated pair is nominally < .75 dB, but we test for < .3 dB; and distance attenuation is 3 db / km And thus exceed 220 m over 62.5/125 =B5m? Do I count the switch to patch cable as a mated pair, or is that included within the How about modal dispersion? Is there a calculation for that? Or do I have to run replacement 50 =B5m fiber? TIA Wrolf | |||||||||||||
|
Posted by on September 18, 2006, 1:46 pm
Please log in for more thread options Wrolf wrote: > Can I calculate the maximum run length for 1000Base-SX based on:
> > Max dB Loss (7.5dB) > # patch panels * loss per mated pair + distance > attenuation > > Where loss per mated pair is nominally < .75 dB, but we test for < .3 > dB; > > and distance attenuation is 3 db / km > > And thus exceed 220 m over 62.5/125 =B5m? > > Do I count the switch to patch cable as a mated pair, or is that > included within the > > How about modal dispersion? Is there a calculation for that? > > Or do I have to run replacement 50 =B5m fiber? How far do you want to go? | |||||||||||||
|
Posted by stephen on September 18, 2006, 6:55 pm
Please log in for more thread options
Wrolf wrote: > Can I calculate the maximum run length for 1000Base-SX based on:
> > Max dB Loss (7.5dB) > # patch panels * loss per mated pair + distance > attenuation > > Where loss per mated pair is nominally < .75 dB, but we test for < .3 > dB; > > and distance attenuation is 3 db / km > > And thus exceed 220 m over 62.5/125 µm? the limit i learnt was 260m. No - the limit isnt power loss, it is model dispersion (ie pulses go via different paths thru the cable and overlap). So it actually depends on your m/mode cable type - higher bandwidth / distance product means you can stretch the limit. 1 easy way to get further is to use LX GBICs and mode conditioning patch leads - that gives you 550m (the same as 50 um fibre). Any more than that then you need a switch at an intermediate point, single mode - or you just "wing" it and hope there is enough slack in the spec for your install that it works. FWIW i have seen 1200m work on old 50 u cable (over a decade ago now, when we put the 1st Bay Networks layer 3 switches in - we didnt have the extra slot for the single mode card). >
--
> Do I count the switch to patch cable as a mated pair, or is that > included within the > > How about modal dispersion? Is there a calculation for that? > > Or do I have to run replacement 50 µm fiber? Regards stephen_hope@xyzworld.com - replace xyz with ntl | |||||||||||||
|
Posted by Wrolf on September 28, 2006, 9:43 pm
Please log in for more thread options
Bod43@hotmail.co.uk wrote: > How far do you want to go?
Turns out one run is 1500', with a single SC-SC patch panel at each end. The other run is about 2000', with a single SC-SC patch panel at each end, and a pair of SC-SC panels in the middle. The patch panels will all be tested for maximum 0.3dB loss per mated pair. Wrolf | |||||||||||||
