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:

Pages: 1 2 [3] 4 5

Author Topic: HCD again, beginning in line 4  (Read 12358 times)

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: HCD again, beginning in line 4
« Reply #30 on: August 01, 2022, 01:10:06 PM »

Line 4 has been dead for a few days until mid-morning when A&A gave it a kick by initiating a test and somehow brought it back to life. Downstream sync rate dropped from 1.4 Mbps to approx 0.4 Mbps. AA is arranging another engineer visit. Quote from AA today:

Quote
I'll get a fault raised. I think we should approach BT on what the long plan solution is here as we can't keep going on like this. We've had countless engineers out with no long term fix out in place.
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: HCD again, beginning in line 4
« Reply #31 on: August 01, 2022, 06:18:02 PM »

Hmm . . . We can only wait and see what will be the Openreach response.  :-\
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.

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: HCD again, beginning in line 4
« Reply #32 on: August 01, 2022, 10:26:50 PM »

HCD error report, tested just now, 21:30 UTC:

--
* HCD:❗So-called ‘hollow curve disease’ (HCD) defect detected in the downstream SNR-vs-tones curve. Serious fault. 🔺
   Link #4: Summary:  tone 62 is 8.25139 dB below chord  1: (tones 40-85); ❗
   Link #4: Details: chord  1: [ tone 40: 22.75 dB; curve midpoint test tone 62:  15.8125 dB; tone 85: 25.4375 dB ]; chord midpoint test tone 62: 24.06389 dB.

Latest picture of horror, taken earlier this afternoon (so numbers don’t match the above):

« Last Edit: August 01, 2022, 10:30:56 PM by Weaver »
Logged

tubaman

  • Senior Kitizen
  • ******
  • Posts: 12507
Re: HCD again, beginning in line 4
« Reply #33 on: August 02, 2022, 09:25:22 AM »

The problem here is that Openreach could throw in the towel and give up trying to fix the line. Although you have the right to request a 'decent' broadband service the obligation for one to be supplied appears somewhat limited - https://www.ofcom.org.uk/phones-telecoms-and-internet/advice-for-consumers/broadband-uso-need-to-know.
Logged
BT FTTC 55/10 Huawei Cab - Zyxel VMG8924-B10A

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: HCD again, beginning in line 4
« Reply #34 on: August 02, 2022, 03:12:03 PM »

I didn’t think that that meant starting up a dispute with AA though. Do you follow?

Is this costing AA, OR, or BTW ? (From RevK’s postings of many years back, he would wish for BTW to pay for repairs because AA is paying BTW for ‘broadband’ and its supposed to be working. Actually a thought occurs to me, if that’s correct, then it would be OR vs BTW in a shouting match then?)

Mind you, even if OR had a strop and refused to fix the line, then I would just order a new line and then do a cease. That worked last year fine.



AA told me "SFI task has been raised for 03/08/2022 PM slot." I’ve heard this term many times but I don’t know what exactly an SFI engineer is.
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: HCD again, beginning in line 4
« Reply #35 on: August 02, 2022, 04:38:32 PM »

AA told me "SFI task has been raised for 03/08/2022 PM slot." I’ve heard this term many times but I don’t know what exactly an SFI engineer is.

SFI -- Special Faults Investigation. More correctly, SFIA -- Special Faults Investigation Assure. An Openreach service available to ISPs to investigate problematic DSL circuits.

Also see the Kitz Glossary 'S' Page.
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.

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: HCD again, beginning in line 4
« Reply #36 on: August 02, 2022, 04:50:31 PM »

Does that mean a different grade of OR engineer? What’s the difference in tasks performed?  The problem is that some of the OR engineers I’ve had in the past don’t (I suspect) understand DSL thoroughly, but I’m perhaps being very unfair.
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: HCD again, beginning in line 4
« Reply #37 on: August 02, 2022, 04:59:35 PM »

Does that mean a different grade of OR engineer?

Yes. I would expect to see a Black Sheep equivalent. Someone who has a good understanding of the DSL technology.

Quote
What’s the difference in tasks performed?

Sorry, I can't answer that. Perhaps others will be able to provide some input? (b*cat looks towards . . . the usual suspects.  ;)  )

Quote
The problem is that some of the OR engineers I’ve had in the past don’t (I suspect) understand DSL thoroughly, but I’m perhaps being very unfair.

Understood. And yes, there are those who are experts in telephony circuits but not DSL circuits.
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.

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 7382
  • VM Gig1 - AAISP L2TP
Re: HCD again, beginning in line 4
« Reply #38 on: August 03, 2022, 05:01:05 PM »

Good luck on line 4 weaver, let us know how it goes. :)
Logged

Black Sheep

  • Helpful
  • Addicted Kitizen
  • *
  • Posts: 5717
Re: HCD again, beginning in line 4
« Reply #39 on: August 03, 2022, 05:49:10 PM »

Does that mean a different grade of OR engineer? What’s the difference in tasks performed?  The problem is that some of the OR engineers I’ve had in the past don’t (I suspect) understand DSL thoroughly, but I’m perhaps being very unfair.

An SFI engineer is exactly as B*Cat describes, skilled in DSL faulting.

As in any group of people though, you'll have the good, the bad and the ugly .... from fully conversant in all things DSL/Network, to new starters freshly trained. It is the luck of the draw who you would get.

An SFI engineer though, should absolutely have all the skill-sets that other engineers may only have one or two of ....... such as MDF/HDF accredited, UG accredited, OH and internal accredited ....... along with DSL accredited. In a nutshell, he/she should be able to perform any task required if a fault is found.

There's the rub .... if a fault is found. They are given strict instructions to carry out the mandatory suite of tests (agreed with all ISP's) and if they all pass, it's adios amigo. They are requested to do so as there will be other customers requiring their services.

This is why it was agreed by all involved that there is a 'cone of acceptance' with regard to the tests. If your circuit falls in that CoA, then it's job done as far as OR and your ISP are concerned. If there is a fault found internally beyond the NTE, then the engineer can repair it if it falls within the 2hr total time paid for by the ISP. If the fault is found to be on OR's network, then there is no time limit on repairing it.

Faults never mend themselves, so if it isn't found on this visit due to being miniscule and in its infancy, then over time it will degrade for sure to the point the tests will find something untoward that can be worked on.

Of course, this will cause debate but it is what it is ....... if there's a nothingness kind of fault, that only by actively searching for it would you know it was even there, chances are the SFI visit will be 20mins long.
Cone of Acceptance, ISP's agreement to the CoA ...  and other fault/install volumes to deal with are to be borne in mind.

I haven't read this thread in its entirety, so unsure as to what level of service affecting issues you are currently experiencing, Weaver.  :)
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: HCD again, beginning in line 4
« Reply #40 on: August 04, 2022, 05:15:57 AM »

The downstream sync rate just goes steadily down and down until it lands at something less than 1 Mbps (normal downstream is 2.7 Mbps @3dB SNRM) while the downstream SNRM goes way up, currently it’s at 12 dB. The ES and CRC values are way too high as part of the sickness, normally the ES count over 15 mins is zero and I like to keep it that way. Right now the most recent downstream speed is 350 kbps. I say ‘most recent’ because the link has been down for about 20 hours and up for 3 hours before that.

Over the last week the link has been up and down for long periods repeatedly.

So I am hoping that given that there’s no DSL at all at the moment this should be a million miles outside the CoA.



OR engineer was booked to come this afternoon, but never showed so I emailed A&A.
Logged

Black Sheep

  • Helpful
  • Addicted Kitizen
  • *
  • Posts: 5717
Re: HCD again, beginning in line 4
« Reply #41 on: August 04, 2022, 09:35:51 AM »

The downstream sync rate just goes steadily down and down until it lands at something less than 1 Mbps (normal downstream is 2.7 Mbps @3dB SNRM) while the downstream SNRM goes way up, currently it’s at 12 dB. The ES and CRC values are way too high as part of the sickness, normally the ES count over 15 mins is zero and I like to keep it that way. Right now the most recent downstream speed is 350 kbps. I say ‘most recent’ because the link has been down for about 20 hours and up for 3 hours before that.

Over the last week the link has been up and down for long periods repeatedly.

So I am hoping that given that there’s no DSL at all at the moment this should be a million miles outside the CoA.



OR engineer was booked to come this afternoon, but never showed so I emailed A&A.

An engineers dream is that, Weaver - give me a circuit that isn't working at all, over an intermittent one any day of the millennium !! Fingers crossed for success, mate.
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: HCD again, beginning in line 4
« Reply #42 on: August 09, 2022, 04:12:38 AM »

AA tell me that they have been chasing BTW re line 4 and other earlier chaos. No news yet though.

Here’s the very latest picture of horror, at 288 kbps downstream, 350 kbps upstream. Weird shape, slightly, required a small revision of the HCD detection algorithm to v5.2 to also handle this. Notice the cut-off on the right, above tone ~110. Can anyone elucidate? The normal highest tone is at ~150-160, very approximately.

Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: HCD again, beginning in line 4
« Reply #43 on: August 09, 2022, 03:33:07 PM »

Notice the cut-off on the right, above tone ~110. Can anyone elucidate? The normal highest tone is at ~150-160, very approximately.

My guess is that the faultly metallic pathway is acting as a band-pass filter, only allowing a signal of limited frequency range to reach the ATU-R.
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.

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: HCD again, beginning in line 4
« Reply #44 on: August 11, 2022, 03:40:42 AM »

My guess, based on nothing much, was that it was some kind of conscious decision made by the ZyXEL when the higher tones are judged to be full of garbage and so need to be disabled. One thing in favour of this, and something that perhaps works against your metallic path theory, is that the cut-off is at a ‘significant number’ tone, here tone 110, if I’m reading it correctly. It’s a multiple of ten, and that, I think, is suspicious.
Logged
Pages: 1 2 [3] 4 5