In reply to a post by Pheasant:

This might help...

https://info.support.huawei.com/AccessInfoTool/PON_B...

And very interesting it is - I can see a ~0.8dB increase in downstream attenuation changing the 1:n splitter to a 2:n splitter.

What I still can't get my head around is how the upstream path doesn't take a >3dB loss changing from a 1:n to a 2:n.

In reply to a post by MaryHinge:

Thinking about it logically if I'm right and they are using dual input splitters then the second input could feasibly be for other wavelengths e.g. from an XGS-PON OLT?

Each splitter has a single lit fibre connected.

OpenReach use this single fibre to carry both the GPON and XGS-PON signal.
No need for a 2nd fibre or 2nd input to do this.

As you pointed out yourself OpenReach don't have the OLT port capacity available for splitters to have a backup fibre connected.

Each splitter has a single lit fibre connected.

OpenReach use this single fibre to carry both the GPON and XGS-PON signal.
No need for a 2nd fibre or 2nd input to do this.

As you pointed out yourself OpenReach don't have the OLT port capacity available for splitters to have a backup fibre connected.

@John83 since you seem to be familiar with this would you be able to confirm if this part of my original post is correct?

I'm not sure but my assumption is that at each SPN they break into one of the bundles of 12 fibres and the rest just pass straight through to the next SPN, so a 36 fibre cable would support a chain of 3 SPNs, and a 72 would support a chain of 6 SPNs.

Of course each SPN would not likely need all 12 individual fibres as the ones I've seen take a maximum of 4 individual 32-way splitters (= max 128 CBT ports per SPN) so you would only need 4 fibres per SPN, but the remaining fibres from the bundle of 12 are stored for future (non-GPON?) use.

In reply to a post by ft247:

In reply to a post by Pheasant:

This might help...

https://info.support.huawei.com/AccessInfoTool/PON_B...

And very interesting it is - I can see a ~0.8dB increase in downstream attenuation changing the 1:n splitter to a 2:n splitter.

What I still can't get my head around is how the upstream path doesn't take a >3dB loss changing from a 1:n to a 2:n.

Have a look at the physics of passive optical splitters here:
https://www.physics-and-radio-electronics.com/blog/f...

Insertion Loss for passive splitters goes up logarithmically based on the number of *output* ports. Forward and return losses are near enough identical - looking at the Huawei PON budget calculator above, there is only a small difference in optical loss between down stream and upstream. This difference accounted by the change in downstream and upstream optical wavelengths - just as it would be if you were measuring loss on a single piece of fibre.

The actual difference in IL between a 1:n splitter and 2:n splitter should (in theory) be no more than 0.2 dB. The loss curve is effectively the same, being driven by the number of outputs, just shifted slightly. I just plugged some numbers into the budget calculator above and it predicted an overall end-to-end 0.3dB difference in loss budget swapping (both forward and reverse) between a 1:4 splitter and 2:4 splitter. So hardly anything.

I've no idea how many fibres are in each fibre tube that goes to the Splitters.

On my development they pulled a single fibre tube (containing who knows how many fibres) to each jointbox containing a splitter.
They put 2 x 32 Splitters in each jointbox they pulled the fibre to.

Some images of the Prysmian outdoor plant used by OpenReach...

https://ibb.co/0qDTdYS

Edited by j0hn83 (Tue 19-Jan-21 12:11:58)

In reply to a post by j0hn83:

Some images of the Prysmian outdoor plant used by OpenReach...

https://ibb.co/0qDTdYS

Very interesting, thanks. The yellow joint is marked as GPON + FIRS. FIRS presumably stands for Fibre Integrated Reception System, which I didn't think would be of interest to Openreach.

In reply to a post by ft247:

In reply to a post by j0hn83:

Some images of the Prysmian outdoor plant used by OpenReach...

https://ibb.co/0qDTdYS

Very interesting, thanks. The yellow joint is marked as GPON + FIRS. FIRS presumably stands for Fibre Integrated Reception System, which I didn't think would be of interest to Openreach.

It isn't.
I should have said some of the plant used by OpenReach.

In reply to a post by j0hn83:

It isn't.
I should have said some of the plant used by OpenReach.

There was an Openreach FIRS product once - withdrawn in late 2019. It's visible in their price list section 5.1.2.

Edited by ft247 (Tue 19-Jan-21 21:14:06)

Never spotted that. Interesting.

I've uploaded the Prysmian document those images came from:

https://drive.google.com/file/d/13ZL5jbyzAOT4g1WXfVq...

Edited by j0hn83 (Tue 19-Jan-21 23:54:58)

In reply to a post by ft247:

In reply to a post by j0hn83:

It isn't.
I should have said some of the plant used by OpenReach.

There was an Openreach FIRS product once - withdrawn in late 2019. It's visible in their price list section 5.1.2.

Interesting. Probably was quite niche and not many takers. Looks like IRS still could've happily worked (@1550nm) alongside GPON and even future XGSPON.