Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Author Topic: G.fast fluctuating upstream SNRM and disconnections - suggestions, opinions?  (Read 601 times)

re0

  • Reg Member
  • ***
  • Posts: 703

Hiya guys and gals,

To summarise what this post will be about, I have been having some large fluctuations (in the form of spikes) on the upstream SNRM along with up to a couple of disconnections a day for the past month especially. Not only are there disconnects, but there can be huge amount of errors and upstream sync speed can spike downwards massively (sometimes spiking below 30 Mbps).

Looking back a few months and beyond, there have been a few occurrences here and there where the SNRM margin on the upstream was a bit impacted. So, while it was not too unusual to see occasionally, it could be even weeks between the occurrences and the impact was not noticed as disconnects never really happened.

What has changed now is that it is happening almost every day recently, especially since around the 8th February. I understand we have had some wet, windy weather. But it looks like that may not be the exact cause since the weather does not seem to always overlap in a meaningful way and may be the contrary some dry days may have the issue, while wet days may not (correlation ≠ causation, anyway). However, I am not ruling it out completely since during heavy rain one night I had experienced popping on the phone line while in the test socket (though I have been unable to replicate it since).

I have reported it to Zen, but the tests keep coming back clear and they cannot see the fluctuations like I can. However, they did note that there had been a lot of errors but do not want to raise the fault to Openreach due to the risk of no fault being found. I understand their position since knowing my luck they will come out and it will not be occurring at the time, but I think I have finally hit the point of it looking like a problem somewhere since the DLM has finally set the target SNRM to 10 dB and it is slowly eating into my upstream (~47 Mbps sync. and attainable, down from ~50 Mbps sync. and ~64 Mbps attainable yes, it is hardly anything in terms of actual sync., but signifies the point of impact).

I have done a few things, such as switching the filters and cables no change. The phone is never plugged in so it cannot be that. The drop cable is new since January 2018, and the socket received some attention (upgraded the faceplate) when G.fast was provisioned last year and checked again when a fault was raised a little while after. I cannot really switch the modem if I want to be able to log since the only spare modem I have is the MT992 (though if the DLM raises the SNRM more I could use it to see if the sync. improves via Zen support or their panel once the modem is switched).

I have attached a few images as examples to what I am seeing my end - just attached 29th February as an example. There have been worse and better days, just wanted to pick a more recent one.

Any suggestions and opinions?

TIA.
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 29432
  • Over the Rainbow Bridge
    • The ELRepo Project

Having seen the three plots and going on nothing more than the "tingles experienced in the whiskers", I will say that there does appear to be a faulty joint "developing". Probably beginning to show high resistance and/or semi-conductive tendencies.
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 32444
  • Trinity: Most guys do.
    • http://www.kitz.co.uk

I would tend to agree with what b*cat says.

G.fast is more likely to be susceptible to early signs of developing HR faults than VDSL...  in exactly the same way VDSL is compared to ADSL..   because you are using higher frequencies.   
I had a HR type fault on ADSL2+.   I knew it was there because I could see certain spikes, but as it wasn't affecting my line rate or sync speed I didn't do anything.   However when on VDSL it immediately affected the line rate and started causing disconnections when the phoneline was in use.   
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

re0

  • Reg Member
  • ***
  • Posts: 703

Thanks for both your inputs, b*cat and kitz.

I've got more graphs I can upload when I'm next at the computer with access to them. Perhaps you'll see something that I'm missing.

Since the connection doesn't meet the fault conditions (according to Zen), I could still request an engineer but with the potential financial penalty. Or I could wait until it manifests into a complete failure, but who knows how long.

I was surprised that they didn't raise the fault given that I had supplied them with evidence showing that the line had audible popping at one stage.

Anyway, is it not weird that the upstream is mostly impacted?
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 32444
  • Trinity: Most guys do.
    • http://www.kitz.co.uk

Because its on the upstream, both the ISP and Openreach seem to care far less than if it was downstream. 

>>  is it not weird that the upstream is mostly impacted?
Not at all - HR faults impact upstream first and usually the most, which is why I suspect it is an emerging HR fault   
I know you say the phone isn't plugged in, but as a test are you able to do so..  and does it get worse if the phone is in use or if the phone rings?
If you do a Quiet Line Test and there is audible noise, you may stand a better chance of getting it fixed sooner as a phone line fault than a dsl fault.   
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

re0

  • Reg Member
  • ***
  • Posts: 703

I haven't tested for a week or so, but last time I checked I didn't notice any difference while a call was in progress or if the phone rang.

As for the QLT, I did notice popping/crackling on the line while doing the test approx. two weeks ago. However, this was during heavy rain and I have been unable to replicate it since.

I would have to test it again later or sometime this week.
Logged

re0

  • Reg Member
  • ***
  • Posts: 703

Last couple of days, SNRM has been really solid. Not really much rain at all, so makes me wonder about water ingress. Either way, there is a problem somewhere. Will continue to monitor.
Logged

re0

  • Reg Member
  • ***
  • Posts: 703

Disconnects have been rare and errors have been less for the last week or so, but the problem persists. DLM dropped SNRM to 9 dB for a few days, but clearly it didn't like it so it's back up to 10 dB again. Fluctuations still present, and I am finding it difficult to put any exact conditions as causation.

Just attached SNRM from the 12th, which was probably one of the worst days in the last week.

I just do not know how to approach it right now, so I am in the same situation as when I started. I just believe that this amount of fluctuation cannot be regarded as normal. Sure, the upstream sync. speed is mostly within the clean and impacted ranges, though the extremes of these fluctuations (spikes) can make it fall below these ranges (for mere seconds).

Perhaps unrelated, but I wonder if it is something to do with the E-side since my estimate for WBC ADSL2+ was around 11-12 Mbps since June last year until sometime around January this year when it tumbled all the way down to 7 Mbps. Perhaps noise could be making its way into the D-side?
Logged