Fibre line loss

Can anyone help me understand fibre line loss (dB's) please, this is in the context of FTTP.

For example

Does fibre have a line loss\attenuation per kilometer in the same way that copper does? (if so do we know what that is for fibre used by Openreach).

Does each splice joint introduce line loss (even if its a good splice)?

When an engineer does a light test (say at the CBT or ONT) how do they calculate if the result is as expected? (how do they determine what it should be)

What is a bad line loss result?

The fibre lines if installed to standards should be able to run 20km so you should not be worrying about the light loss figures

It is either constructed well enough to work or not.

Edited by MrSaffron (Sat 08-Dec-18 12:18:53)

I would like to understand fibre line loss for my own technical understanding not for any specific issues.

AFAIK light can travel for up to 40km in fibre without requiring booster(s).

But in answer to the OP, yes there is light loss in a FTTP circuit at various points. During the testing phase of my FTTP line and also later on when there was a fault*, i saw Openreach go around the various points (ONT/CSP/DP/splitter/agg node) measuring light loss. I believe there is a certain tolerance wrt light loss, though I have no idea what the actual numbers are.

But at the end of the day, if your FTTP service works then you have nothing to worry about.

*the "fault" actually turned out to be a human error as someone had messed up my line profile at the exchange - a line reconfig cured this but only after the Engineer had spent 8 hrs outside in the pouring rain tracing the fibre all the way back to the exchange.

Edited by deleted (Sat 08-Dec-18 12:02:05)

https://amzn.to/2PpwiCd

As others have said there is degradation of the light level as the signal travels through fibre. How far the signal goes depends on the optics deployed.

The speed you get over fibre is fixed - so it will always run at the same speed regardless of the loss. When the loss gets to a low enough level then the fibre will stop working - there's no adaptive speed function that means it would continue to work at a lower speed, like there is with xDSL technologies.

Dect

As other have said if it works it doesn't affect FTTP.

There is loss caused by the distance in the fibre and by each splice.

Splices are affected if the fibre was dirty, if not a clean cut etc. just like a torch lens affects light quality.

The loss in the fibre is affected by the SFP used ( the bit that goes on the ends) these are rated for different distances. eg short medium long, SX,MX,LX (say up to 100meters,80m-10k , 8k to 100k etc.) and the quality of fibre. Look online if you want more technical detail on what is available now as my knowledge is slightly out of date in a fast moving field

Using a higher spec one at a shorter distance will blow the distant end and they should match both ends

Over the distances for FTTP this is all irrelevant as it will work or not. There is NO degradation of the service until it breaks, so you get the full speed you buy or nothing over the access FTTP segment. There is the possibility of Contention over the segment from the splitter to the Head end and congestion of backhaul, core and to the content provider will still affect your actual speeds.

Hope this gives you the overview you wanted.

These are a good starting point
http://www.thefoa.org/tech/lossbudg.htm
http://www.thefoa.org/tech/loss-est.htm
http://www.thefoa.org/tech/Linkspec.htm

. I believe there is a certain tolerance wrt light loss, though I have no idea what the actual numbers are.

So for the Openreach PON product measuring @1490 the light loss reading ought to be within -10 to -25db

Have attended a couple of faults where initial tests at the ONT showed readings towards the top end of the scale -23db the reported issue being intermittent lengthy loss of service.

Both were cleared by locating a bum splice, once redone the final reading was -14db

Well done dect for sticking with it. At least you got some sensible replies!