"  We ended having to build a tool to do the calculations for the techs:"

That's a great idea.

On Tue, Aug 19, 2025 at 5:54 PM Mark Radabaugh <[email protected]> wrote:

> There are two limitations - max distance due to both timing and light that
> are currently 60km but I think it’s theoretically limited at 100km.
> There is also a max difference between the closest and farthest ONT on any
> given port - and that is typically 20km or 40km, but some platforms will
> let you set something in-between 20 and 40km.
>
> If you are using 20km optics the max split doesn’t matter.
>
> If you are using 40km optics and you can put at ONT at 1m and one at
> 20.001km and it should work, but the ONT at 21km won’t - unless you set the
> max differential to 40km.   It costs you a bit of bandwidth to do.
>
> 60km are the same way - 20 or 40km max differential.
>
> The bigger issue with 40 and 60km optics is return loss.   You can keep
> picking better optics at the OLT - A, B, C, C+ and increasing the output
> power, but the ONT’s are all the same output power, and they transmit back
> on the higher loss wavelength.    The C and C+ optics have better receive
> sensitivity and additional error correction (FEC), but ultimately even
> though the ONT can hear the signal fine, it can’t communicate back to the
> head end.
>
> You will likely also find that when running at >30km the difference in ONT
> and OLT receive signals confuses the techs.    They get very used to seeing
> receive light levels that about match - but forget that the 1577nm
> (0.2dB/km) downstream wavelength has much lower loss per km than the 1270nm
> (0.35dB/km) upstream wavelength.    We ended having to build a tool to do
> the calculations for the techs:
>
> Customer:xxxxxxx-ont
> RX at the ONT: -14.9 dB
> TX from the ONT: 6.5 dB
> Expected Loss: 28.5 dB
> Actual Loss: 26.2 dB
> Difference: -2.3 dB
> RX at the OLT: -26.5 dB
> Expected Loss: 34.9 dB
> Actual Loss: 33 dB
> Difference: -1.9 dB
> Distance from OLT: 42.68 km
>
>
>
>
>
>
>
>
>
>
>
>
>
> Without it they tended to freak out - what’s wrong with the fiber - the
> ONT is at -14.9dB and the OLT is seeing -26.5dB?    Do the math at 42.68km
> and they are both about 2dB better than the expected at that  distance.
> This one is within 3dB of theoretical - and good to go.
>
> Mark
>
> On Aug 19, 2025, at 4:55 PM, Chuck <[email protected]> wrote:
>
> There was in the beginning due to the timing and cuing of ONT upstream.
> It was to ensure 32 ONTs could all be transmitting max traffic at the same
> time.  That is what Calix told me many years ago.
> Sent from my iPhone
>
> On Aug 19, 2025, at 11:50 AM, Josh Luthman <[email protected]>
> wrote:
>
> 
> From what I've read, the issue was the timing between the farthest &
> closest ONT.  There was no 20km timing limitation from OLT to anything.
>
> On Tue, Aug 19, 2025 at 12:39 PM <[email protected]> wrote:
>
>> Original GPON specs had a timing limitation of 20 km.
>>
>>
>>
>>
>>
>> *From:* AF <[email protected]> *On Behalf Of *Josh Luthman
>> *Sent:* Tuesday, August 19, 2025 6:31 AM
>> *To:* AnimalFarm Microwave Users Group <[email protected]>
>> *Subject:* Re: [AFMUG] how is underground FTTH done?
>>
>>
>>
>> Class C optics are rated for 60km or 42mi.  That's a hell of a distance
>> when you're only doing 1:32.  We did 1:2 and 1:32 on the sixteenth PON port
>> until we built more to justify another OLT.
>>
>>
>>
>> On Mon, Aug 18, 2025 at 6:13 PM Chuck McCown <[email protected]> wrote:
>>
>> When you work out a 32:1 GPON, the timing distance limitations are
>> exhausted before you run out of light.  Splits are a 3 dB loss.  A good
>> splice will be .02dB.  So don’t worry about splices.
>>
>>
>>
>> *From:* AF [mailto:[email protected]] *On Behalf Of *Dev
>> *Sent:* Monday, August 18, 2025 9:56 AM
>> *To:* AnimalFarm Microwave Users Group <[email protected]>
>> *Subject:* Re: [AFMUG] how is underground FTTH done?
>>
>>
>>
>> The thing with PON is that every time the light hits a splitter it loses
>> a bit of strength, so you sort of have to plan to use a little hotter SFP
>> in your OLT sometimes. Luckily, those are getting way cheaper nowadays, so
>> it’s not the end of the world, but you do have to plan for that.
>>
>>
>>
>> Plus, each splice you do cuts down the signal a bit more. Fusion splicers
>> only lose a tiny bit, but a physical connector can lose a bunch more, like
>> the equivalent of a mile or more of distance, and a piece of crud on an
>> uncleaned connector can lose 5 miles distance, so make sure you clean them
>> with those cheap cleaner tools.
>>
>>
>>
>> On Aug 18, 2025, at 8:39 AM, Josh Luthman <[email protected]>
>> wrote:
>>
>>
>>
>> Splitters are waaay small.  Smaller than a standard house key.
>>
>>
>>
>> What you are looking at is an MST terminal, looks like 8 ports.  There
>> can be a splitter inside of that yes.  You can have the MST with 8 fibers
>> splice to another 8 fibers or you can have what is in your picture have 1
>> fiber in, split 1x8, and then have 8 ports out for the installers to simply
>> plug in to.
>>
>>
>>
>> If that MST is a 1x8, you can have a 1x4 before it, between the MST and
>> OLT.  That makes for OLT -> 1x4 splitter -> 1x8 splitter/MST.  That is
>> still a 1x32 split.
>>
>>
>>
>> On Mon, Aug 18, 2025 at 11:34 AM Ken Hohhof <[email protected]> wrote:
>>
>> I thought PON used like 16:1 or 32:1 splitters, and in this photo, I
>> assumed that’s what the black boxes were.
>>
>>
>>
>> *From:* AF <[email protected]> *On Behalf Of *Josh Luthman
>> *Sent:* Monday, August 18, 2025 10:16 AM
>> *To:* AnimalFarm Microwave Users Group <[email protected]>
>> *Subject:* Re: [AFMUG] how is underground FTTH done?
>>
>>
>>
>> Don't assume that about aerial.  That's not how it works.  Don't think
>> about it in terms of taps.
>>
>>
>>
>> Generally speaking, installations are PON.  What we do is design the
>> fiber so we can hook up 100% of homes.  We assign a color to every house.
>>
>>
>>
>> The first thing to think about is that you have to access the individual
>> strand out of the cable, be it 12/24/48/144/etc.  That is done with a
>> SpliceCase or you splice on an MST for an ez mode plug.  At Imagine we only
>> splice - no connectors, no MST, no plugs, etc.
>>
>>
>> Second thing is that when there's a cable up and down the road, you just
>> need access to it through the case/MST from the house.  This can be from
>> the house to the handhole (concrete box in the ground) or you can run it
>> from the house to the handhole through some 1.25" duct to the next handhole
>> where there is one case.
>>
>>
>>
>> I can show you what it looks like if you don't get it yet.
>>
>>
>>
>> On Mon, Aug 18, 2025 at 11:11 AM Ken Hohhof <[email protected]> wrote:
>>
>> The fiber train left without me, so maybe someone here can help me
>> understand how the physical installation is typically done.
>>
>>
>>
>> I’ve seen aerial fiber and it’s pretty straightforward, I see splitters
>> up on poles maybe at each intersection, and to hook up a customer, they run
>> a drop wire from the nearest splitter to the house.  If take rate is better
>> than expected or a new house is built, worst case I assume they just add a
>> splitter.
>>
>>
>>
>> But I also see FTTH deployments going in where they are boring for duct
>> in the ROW and putting a little handhole in front of every house.  How does
>> this work?  Are they using taps instead of splitters?  If not, when they
>> get a customer install order, do they pull his drop cable through all the
>> handholes to a splitter?  That doesn’t seem feasible.  Are they dedicating
>> a strand to each house and pulling the main cable out each time and
>> splicing to that strand?  And what if they estimate the take rate wrong, or
>> a new house is built?
>>
>>
>>
>> There’s probably a simple explanation and once someone enlightens me it
>> will be a Duh! moment.
>>
>> --
>> AF mailing list
>> [email protected]
>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com
>>
>> --
>> AF mailing list
>> [email protected]
>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com
>>
>> --
>> AF mailing list
>> [email protected]
>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com
>>
>>
>>
>> --
>> AF mailing list
>> [email protected]
>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com
>>
>> --
>> AF mailing list
>> [email protected]
>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com
>>
> --
> AF mailing list
> [email protected]
> http://af.afmug.com/mailman/listinfo/af_af.afmug.com
>
> --
> AF mailing list
> [email protected]
> http://af.afmug.com/mailman/listinfo/af_af.afmug.com
>
>
> --
> AF mailing list
> [email protected]
> http://af.afmug.com/mailman/listinfo/af_af.afmug.com
>
-- 
AF mailing list
[email protected]
http://af.afmug.com/mailman/listinfo/af_af.afmug.com

Reply via email to