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: Problems solved ... maybe?  (Read 3182 times)

LaurieR

  • Member
  • **
  • Posts: 35
Problems solved ... maybe?
« on: September 08, 2015, 12:32:42 PM »

I'm "kempshott" on MyDSLWebStats.

For many years our FTTC hasn't been great. Downstream slowly dropped from the original estimate of 32 mbps to 24-ish, but upstream has been really flaky, varying randomly from nearly 5 mbps at best, to under 1 mbps for weeks on end.
We've had more than a dozen BT engineers investigate over the years; they've blamed our distance from the cabinet (850m) aluminium wiring, REIN, RFI, split pairs - just about every excuse imaginable.

But yesterday, we finally got an engineer who seems to have sorted it out. He said he'd found "alternative routing" to the cabinet even though we'd been told on many occasions there were no spare pairs anywhere. Now we get a (so far stable):

Max:   Upstream rate = 5420 Kbps, Downstream rate = 41996 Kbps
Bearer:   0, Upstream rate = 5409 Kbps, Downstream rate = 36813 Kbps

(I'm on 40/10). G.INP is off following the DLM reset yesterday. So that's seems all good - maybe even TOO good considering our distance from the cabinet and 23dB line attenuation. The downstream interleave of 721 and the very high downstream FEC error rate of around 4000/min concern me.

It's only been 24 hours but does this look too good to be true?

Logged

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 7382
  • VM Gig1 - AAISP L2TP
Re: Problems solved ... maybe?
« Reply #1 on: September 08, 2015, 04:30:23 PM »

it depends on the error rates of the line which you didnt post, if they low enough then you shouldnt lose performance from DLM changes.

Obviously future crosstalk can never be ruled out.  But most crosstalk tends to occur in the early days for cabinets.
Logged

LaurieR

  • Member
  • **
  • Posts: 35
Re: Problems solved ... maybe?
« Reply #2 on: September 08, 2015, 05:49:21 PM »

it depends on the error rates of the line which you didnt post  ...

I said:
Quote
I'm kempshott on MyDSLWebStats
- which I thought would show the "before" and "after" historical error rates. Hardly any FECs appear before yesterday. Happy to dig deeper - which figures would help?
Logged

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 7382
  • VM Gig1 - AAISP L2TP
Re: Problems solved ... maybe?
« Reply #3 on: September 08, 2015, 06:17:08 PM »

FEC CRC ES and SES
Logged

Black Sheep

  • Helpful
  • Addicted Kitizen
  • *
  • Posts: 5717
Re: Problems solved ... maybe?
« Reply #4 on: September 08, 2015, 06:54:23 PM »

I'm "kempshott" on MyDSLWebStats.

For many years our FTTC hasn't been great. Downstream slowly dropped from the original estimate of 32 mbps to 24-ish, but upstream has been really flaky, varying randomly from nearly 5 mbps at best, to under 1 mbps for weeks on end.
We've had more than a dozen BT engineers investigate over the years; they've blamed our distance from the cabinet (850m) aluminium wiring, REIN, RFI, split pairs - just about every excuse imaginable.

But yesterday, we finally got an engineer who seems to have sorted it out. He said he'd found "alternative routing" to the cabinet even though we'd been told on many occasions there were no spare pairs anywhere. Now we get a (so far stable):

Max:   Upstream rate = 5420 Kbps, Downstream rate = 41996 Kbps
Bearer:   0, Upstream rate = 5409 Kbps, Downstream rate = 36813 Kbps

(I'm on 40/10). G.INP is off following the DLM reset yesterday. So that's seems all good - maybe even TOO good considering our distance from the cabinet and 23dB line attenuation. The downstream interleave of 721 and the very high downstream FEC error rate of around 4000/min concern me.

It's only been 24 hours but does this look too good to be true?

Pleased you've got a result.

Unless you've ever carried out DSL/PSTN network engineering on a large scale, then you won't have an appreciation as to what an engineer faces ?? Simply passing off work done as 'Excuses' highlights the naivety of this.

Unfortunately, we don't have access to the previous engineers notes and actual work completed, and I'm not naïve enough myself to realise we do have poor engineers out there, but they are in the low percentages as opposed to high.
With that in mind, percentage-wise there's a bl00dy good chance you have had a decent engineer, who hasn't been able to detect a fault at that particular time in history. Again, this can happen to the very best of engineers.

The 'Excuses' listed (Split-pairs, REIN, Ali cable etc etc), are all symptomatic of DSL problems. The fact the engineer today found a 'Spare', but the other engineers couldn't ......... doesn't mean they were being lazy or incapable ...... it just means that today there happens to be a 'Spare'.
Distribution point (DP) capacity is forever changing, for example Mrs Smith who used to have 2 phone lines as she worked from home, now only has the 1 phone line as she's since left the company. That now means there is a 'Spare' line.

As I say, I'm pleased success has been found for you ...... but, for the new readership of this forum who may themselves have an issue, (lets face it, not many will be GOOGLING 'Broadband fault' if they haven't got one), I feel I need to balance the input so they aren't of the opinion most engineers make excuses and do little else.  :)
Logged

LaurieR

  • Member
  • **
  • Posts: 35
Re: Problems solved ... maybe?
« Reply #5 on: September 09, 2015, 10:14:21 AM »

G.INP kicked in at 04.40 and now I've hit my 40/10 cap. FECs are way down and interleave now 16 / 1.

Max:   Upstream rate = 5319 Kbps, Downstream rate = 42068 Kbps
Bearer:   0, Upstream rate = 5359 Kbps, Downstream rate = 40000 Kbps

I simply didn't think this was possible given my 850m distance from the cabinet and apparent aluminium cabling. When I did my IBM telecomms training back in the 70's we were assured 4800 baud was the most POTS would ever handle.

I wasn't having a go at the individual BT engineers, most of whom let slip they were under enormous managerial pressure to close calls as quickly and cheaply as possible when many of the problems (round here, anyway) were down to lack of investment in infrastructure.

Logged

Dray

  • Kitizen
  • ****
  • Posts: 2361
Re: Problems solved ... maybe?
« Reply #6 on: September 09, 2015, 10:25:52 AM »

Excellent :)
Logged

b4dger

  • Reg Member
  • ***
  • Posts: 214
    • HmmmUK
Re: Problems solved ... maybe?
« Reply #7 on: September 12, 2015, 01:11:33 PM »

Great outcome :)
Sorry if I missed it, but what router/modem are you using?

Logged

LaurieR

  • Member
  • **
  • Posts: 35
Re: Problems solved ... maybe?
« Reply #8 on: September 12, 2015, 01:43:55 PM »

Great outcome :)
Sorry if I missed it, but what router/modem are you using?

Unlocked Huawei HG612 modem and (blush!) PlusNet-supplied N150 router. I had been looking for an excuse to get a combined modem/router but wifi is OK and, with my line now sorted, I'm not sure I need one.

Logged

b4dger

  • Reg Member
  • ***
  • Posts: 214
    • HmmmUK
Re: Problems solved ... maybe?
« Reply #9 on: September 12, 2015, 03:40:33 PM »

Thanks - so nothing special.
Sounds like it must just be a decent line now it's sorted.

My line is of a similar distance to the cabinet.
Started at 30Mbps which soon settled at 20.

For the last week I've had a fault - no phone and VDSL down to 10Mbps - So I'm guessing I'm down to one wire somewhere along the line.

BT initially thought it would be fixed without access to my house, but when they attempted to fix things during the week that didn't work out. So my next engineer is visiting on Monday (after carrying out numerous test socket tests for Zen my line provider and having to accept the normal charges if it's my fault etc.).  It's been a painful week, but it would cheer me up if at the end of it my line improves and behaves similar to yours :)
Logged
 

anything