Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Dar2211]

Good morning, could the kitzs experts have a look at my data in mydslwebstats, user name dar2211, please. Zen said to wait following engineer work to see if dlm adjusted the syncs. Nothing happened for three weeks, two of which we were on holiday. Resyned on return but appear to be stuck at 21/1.1, with a line that has hiss & intermittent crackle. It is stable though, show odd patterns of bit swaps per min (10pm to 6am). Appear to be only using band zero cos of line state. Zen says BT won't replace the line or connect me to cab 36 although the data says it serves my road. Some background if you want it, sorry about the length.
Admission of guilt first, took zens & bt's claims at face value & upgraded from zens adsl without doing any research. Both companies claimed 20-27 meg down & upto 3 meg up, bit better than my adsl of 9/1 & worth upgrade for £6 per month. Figures did seem low as I have a fibre cab(36) at the bottom of the road but mention of cross talk blar, blar, made them seem cautionary estimate.
On activation & for 2 days, 22/3 seemed a good increase. Over next few days syn speeds dropped to 12/0.8! Followed zen help soon showed a noisy line. Many emails, calls & reading stuff on this site opened my eyes. Checking cabinets showed my lines into cab 48 1.4 km away, I'd upgraded & "removed" 500m of copper. Zen detected a bridge tap. Three engineer visits, first one didn't want to know, said "line will never give claimed speeds cos there's aluminium in the line, I'll change face plate to see if that helps". Second engineer was better "connected a better down line to pole at end of road", this saw the down syn increase to 17 meg but ignored the tap. Third engineer, young guy much better, changed the socket back plate as it was a screw connector type & should have been changed at first visit, went up the pole outside the house & connected using gel connectors cos box at pole head was screw type. Result more stable line at 21 meg down but no change on the up, stuck at 1.1meg.
What should I say to Zen?
Thanks Dave

[loonylion]

If you have audible noise on the line you should be raising a telephone fault with your telephone service provider.

[j0hn]

OpenReach won't change cabinets I'm afraid, that's a dead end. I'm in a brand new property, in a brand new street, and the vast majority are on the local cabinet at the end of the street. Then there's me and a few other neighbours, who lines run right past said cabinet for a few hundred more yards. knowing my line runs past said cabinet, my initial thoughts were "oh they could just plug me in there" but unfortunately it didn't work like that. Your local cabinet 36 may be full, your line may not pass it, and even if it does your line is within a bigger cable with 99 other lines, and they can't just pop you into the local cabinet.

as for your other issues, the lack of bands being used etc, you need somebody with a bit more knowledge than me I'm afraid.

[Chunkers]

HI, you have my sympathies.  In some ways we are in similar position :

• I am around 1.3 km from my cabinet (more to the exchange) and connect (via Zen) at 9md down / 1Mb up using ADSL 2+
• I have aluminium in my lines and get a relatively high number of errors
• I have had engineers round and have all-new internal wiring / faceplate etc etc

Unlike you, when my local cabinet was FTTC enabled my speed estimate for FTTC was actually less than my current speed on ADSL 2+ so i can understand why you went for it.

I have been led to believe that there is almost no possibility of me convincing BT to change-out out Aluminium line to Copper although you do read some stories where people have done it, so if that's your issue then its not looking good..... I spoke to one of the engineers and he said it never happens unless the voice comms are affected.

Sorry to be the bearer of bad tidings.

Chnks

[Dar2211]

Thanks guys for the quick replies. Will report the noise on the line again & hope that repeated call outs will make OR do something. Hopefully some of you data experts will spot something!

[burakkucat]

Welcome to the Kitz forum. 

I have just configured the viewing frames on MDWS to show some data from your circuit. Not being an expert analyst of G.993.2 (VDSL2) circuits, I tend to view the following five plots -- SNRM, FECs per minute, CRCs per minute, Hlog and QLN.

The SNRM is really quite acceptable, with no apparent fall and rise during the hours of darkness and daylight per 24 hour period. Both DS and US are essentially at 6 dB over a 24 hour period.

The few peaks that are apparent on the FEC plot -- which are DS only -- show that the error detection and correction mechanism is working well.

As G.Inp is active on your circuit, the CRC plot shows predominately US errors at a very low level and one DS spike in the low twenties . . . but nothing of concern.

The Hlog plot shows no evidence of a bridging tap but is quite clear that your circuit's poor overall performance is a result of a "decayed" metallic pathway. Your circuit is relying on the US0 and DS1 bands only. (The might be a small contribution from the DS2 band but as it shows an increasing perturbation with frequency rather that a smooth curve I would be inclined to ignore it.)

The QLN plot is fairly typical for the average xDSL circuit. Perhaps the average level of the "noise floor" is about 5 dB greater than what would be liked but there is really nothing of significance present.

We must remember that both the Hlog and QLN plots are "snapshots" of the prevailing conditions at the time when the DSLAM and modem underwent (re)synchronisation. They are not "ongoing" plots like the other three, to which I have referred.

For a more thorough analysis of your circuit's performance you really need the attention of an analytical WWWombat or Bald_Eagle1. Let's see who hears my whistle . . . 

Having been told that there is a significant quantity of aluminium cabling in your circuit, the Hlog and QLN plots show that it is possibly the best that the circuit can achieve. 

When making that call to your telephony service provider to report an audibly noisy line, do not make any mention of the broadband service.

Ultimately, if nothing can be done to improve the quality and performance, you might consider discussing, with Zen, its reversion back to a G.992.5 (ADSL2+) circuit.

[WWWombat]

(My post crossed with b*cat's, but I don't think I can add more detail)

Do you think you still have the tap? The Hlog graph doesn't seem to show it currently.

The quiet line graph seems to not show much noise either.

The G.INP retransmission appears to be handling errors OK, and there is little variation in snrm., except for one event last week that looks suspiciously like a thunderstorm. Not a lot can be done with that, but the rest seems OK. With audible noise on the line, I'd have expected a little worse.

On balance, the speeds are likely to be entirely down to the combination of length and aluminium. But it is certainly worth getting the audible noise sorted.

Copper rearrangement doesn't happen much, especially for individual lines. It is just too costly. However, we are starting to see it a bit more within the most recent BDUK projects, where there are full groups of people who would be better served by a closer cabinet. This happens with new all-in-one cabinets more, but sometimes by re-parenting to a closer cabinet.

To be funded by BDUK, you'd need to be part of a group that would benefit, and be cheaper to subsidise than other potential beneficiaries... and as rearrangement is more expensive than the norm, it remains relatively rare.

[Dar2211]

Many thanks guys for that excellent explanation of my lines performance.
What you both say confirms what the first engineer said, I shouldn't have doubted him! He seemed pretty tee'd of about going out and explaining why super fast broadband often couldn't be achieved.
I'll keep an eye on the charts that you set up burakkucat. I'll leave the monitoring on. The last QLN followed my disconnection at night. Is it worth disconnecting for 30 mins during the day & seeing if the QLN changes. My QL test have all been done during the day, I have a few excellent recordings of the noise which I'll see if I can post next week when we're home.
The bit swapping behaviour is interesting & I wonder what is causing it as its happening when there is no data activity on the line.
As too, G.inp, if that was set on the up could that improve its sync? Uploading data & images takes the same time as dsl. Or is it simply that as the lines on the noise floor that higher frequency bands cannot be used.
Zen said BT wouldn't replace the lines due to cost, but that was rich I said, they've been getting & will continue to get around 20 quid a month of all the residents in our road for lines that cannot guarantee superfast broadband. A bit of a scam really, pulling a length of copper cable can't cost that much. Moreover the last engineer said there was plenty of spare capacity in cab 36 for all of our road, so why not re route. That's another argument though.
wwwombat, the tap was still showing on the third engineers line tests. He replaced the 80's wall plate with the drop cable screw connectors with a new press fit type one & the Bridge tap disappeared. He said there were capacitors etc in the old one that were responsible, said it was known issue & first engineer should have replaced it straight away. They have an instruction to change any they find. You're completely right about the thunderstorm! I expected it to cause a resync but it must have affected lots of lines so no action was taken by the DLMs. (I learnt this from the tech pages  )
My line is provided by Zen but I'll still report the line noise. Am I right in thinking quiet line means absolutely no audible noise at all?
Again many thanks & I'll report what Zen & BT say.

[Dar2211]

Should have also said that I'd have been completely lost & at the mercy of Zen tech support & BT if it wasn't for this site, forum & it's experts!

[burakkucat]

My advice would be not to deliberately disturb the circuit's synchronisation status on the off-chance that there might be a difference observed in the QLN (or Hlog) plots.

The Hlog plot tells us that there is just no usable signal above the highest sub-carrier in the DS1 band. The Hlog plot can be regarded as a representation of the attenuation per individual sub-carrier. So with the significant attenuation, the overall bandwidth of the circuit is severely limited.

The QLN plot is a visual representation of the electrical noise that is detected per sub-carrier. The more positive the value, the more noise detected. (For clarification, the QLN plot is displaying noise across the radio frequency spectrum. For audible signals carried by the telephony network the lower and upper limits are approximately 300 Hz to 3400 Hz respectively. As each sub-carrier of a broadband signal is 4.3123 kHz wide, it can be seen that the audio pass-band for a telephone can be accommodated within one sub-carrier. All sub-carriers from 0 to 31 are left unused by the broadband services. Those 32 sub-carriers provide plenty of space for the telephony service, along with a buffer-zone.)

As for a correctly operating telephone service, there should be no hum, no clicks, crackles, plops, buzzing, overhearing other users, etc, etc. The only sensation one should detect when performing a Quiet Line Test (from a wired, not cordless, telephone establish a connection to 17070 and take the relevant menu option) is that the telephone appears to be "electrically alive". If you were to gently blow into the microphone, you should hear it in the earpiece (sidetone) -- but nothing else.

[Dar2211]

My advice would be not to deliberately disturb the circuit's synchronisation status on the off-chance that there might be a difference observed in the QLN (or Hlog) plots.

The Hlog plot tells us that there is just no usable signal above the highest sub-carrier in the DS1 band. The Hlog plot can be regarded as a representation of the attenuation per individual sub-carrier. So with the significant attenuation, the overall bandwidth of the circuit is severely limited.

The QLN plot is a visual representation of the electrical noise that is detected per sub-carrier. The more positive the value, the more noise detected. (For clarification, the QLN plot is displaying noise across the radio frequency spectrum. For audible signals carried by the telephony network the lower and upper limits are approximately 300 Hz to 3400 Hz respectively. As each sub-carrier of a broadband signal is 4.3123 kHz wide, it can be seen that the audio pass-band for a telephone can be accommodated within one sub-carrier. All sub-carriers from 0 to 31 are left unused by the broadband services. Those 32 sub-carriers provide plenty of space for the telephony service, along with a buffer-zone.)

As for a correctly operating telephone service, there should be no hum, no clicks, crackles, plops, buzzing, overhearing other users, etc, etc. The only sensation one should detect when performing a Quiet Line Test (from a wired, not cordless, telephone establish a connection to 17070 and take the relevant menu option) is that the telephone appears to be "electrically alive". If you were to gently blow into the microphone, you should hear it in the earpiece (sidetone) -- but nothing else.

Right got it, this burakkucat.
My lines defo not quiet!

[burakkucat]

My lines defo not quiet!

Then that defect is the one upon which to focus. Report it as a telephony fault; audible noise.

[Dar2211]

Quick update
After 5 engineer visits, last two being telephone guys I now have a line that only has a faint hum.
Both telephone eng when testing the line said there was no noise or hum; I disagreed. So first guy went to the second junction point and replaced the connectors but couldn't do the next ones as it was underground & a special engineer was needed. Last telephone engineer was one of these special underground engineers but as he detected no faults he wouldn't do anything, it was throwing it down at the time.
Interestingly both line engineers put in their report that they had changed my faceplate but they had not.
At this point Zen said they could do no more as the line was performing as per BT line checker i.e. 22/2.2 mbps. Partially correct, down manages 22mbps but up is fixed at 1.2mbps
I've continued to monitor the modem stats (dar2211) & since the engineer visits the QLN has changed, there is data for all the bands & the Hlog plot is a little different, there are 4 points at higher frequency now! The error correction looks to be working & keeps mostly the service running.
I still get random modem issues, disconnects & no service although all the lights appear normal. I had the latter this morning but the stats don't show an issue.
Still not happy, the figures suggest that the service is OK but when using IPTV I see random pixilation on HD channels. I cannot stream 4K movies due to constant buffering, the service provider (oober tv) reports that at 22 mbps there should be no issues on any service. Sadly I don't know if this is true.
I'm still of the opinion that BT/Openreach should do something about the line quality so I've emailed Clive Selley, CEO Openreach. My comments are being investigated by my local area manager but as yet I've had no details. Will post again when I receive a reply.
It bugs me that BTOR can spend thousands of pounds on national advertising telling us that 96% of the population can get super fast broadband but won't replace crappy lines or do cab changes.
Any suggestions and comments would be appreciated.

[Dar2211]

Should have attached these graphs to my last post

[Dar2211]

Good afternoon, finally I have had a response from BT high level investigation team following an investigation by my local area manager.
As you can see their investigation and conclusion is what others on this forum have said:

Dear Mr Ridley
 
Further to my email of 1 November, I am very sorry for the delay in getting back to you.
 
I have been in contact with the local engineering manager and we have carried out a full review of the local network serving your property.   Your line is served from Arrowebrook exchange via street cabinet 48.
 
I appreciate that the close proximity of another fibre cabinet is frustrating.  However, it does depend on where the boundaries are for each individual cabinet as to which premises are served by them.  These boundaries are established historically. Also each cabinet has a dedicated number of distribution points which are fed via dedicated ducts which restricts capacity to each cabinet  and  therefore it is not possible to move to a different cabinet. 
 
Having looked in detail at your line and recent faults the engineering manager  has found that the problems you have been having are due to the length of your line rather than any inherent defects in the network. Unfortunately there is a marked degradation in performance of the Fibre to the Cabinet (FTTC) product depending on the length of the copper line from the fibre cabinet to the premises.  Your line is 1.4km and this is the distance where issues unfortunately start  to occur.
 
Please continue to report any future problems you have to your communication provider (CP) who can carry out their own checks.  All CPs have dedicated contact points and escalation paths within Openreach if they need any assistance on behalf of their customers. This is the correct industry process. I have highlighted your comments regarding the engineers notes and behaviour. Thank you for bringing this to our attention. This has been raised with the engineering manager for the relevant coaching and feedback.
 
Kind Regards

Sarah Reeves
Customer Resolutions,High level complaints
Openreach
Web: www.openreach.co.uk

So no mention of any possible fix to the lines poor performance, i.e. tough and make do!
Interesting that they make no mention of an impacted line & presence of aluminium cable in the line from cabinet 48. Their own dsl checker shows that the line is "impacted", with a particularly poor upstream rate. I've attached the dsl checker results.
I've emailed again Clive Selley, saying how dissatisfied I am with this response but so far I haven't had any comments back.  I think this says a lot.
I've forwarded the BT reply to Zen Tech Support and received this:

Good Afternoon Dave,

Thank you for the update. Regarding your upstream speeds. The result you're seeing is within the estimate that was already advised to yourself by my colleague Chris on 24/10 and can also be confirmed by visiting the BT broadband checker site: https://www.dslchecker.bt.com/adsl/adslchecker.TelephoneNumberOutput

The current profile you're on for fibre is 0.128M-40M Downstream, Retransmission Low - 0.128M-10M Upstream, Error Protection Off. so a DLM reset would not change this as you are already achieving the speeds the line is capable of. The overall broadband connection is also very stable with no adverse interference or faults being detected.

We will be soft closing this fault ticket in 24 hours. This removes the fault from our work stack to keep reviewing however if a fault does re-occur within 7 days we can reopen the ticket and pass back to BT.

Please feel free to get back in touch if you have any further queries.

Kind Regards

So basically "get lost" we can't do anything. They have also stated that as the line is performing within the BT estimates that they have complied with their own terms and conditions so I will have to see the contract out.
You'll see the quoted profile has a spec that is quite unbelievably wide where the downstream minimum is only 0.32% of the max and the upstream minimum is only 1.3% of the max!
Unless I'm missing something, this says they can supply whatever they like and still comply with the T&C's.  I can't see that this can be acceptable.

[Moderator edited to remove the telephone number from the attached image.]