Kitz Forum
Broadband Related => FTTC and FTTP Issues => Topic started by: Chrysalis on November 12, 2013, 12:09:12 AM
-
Ok so I think I have a HR fault.
I can repeatedly drop my upstream snrm down from 9db+ to under 3db simply by turning on my landline phone dialtone. Its not immediate but I get a largish drop of about 3db after about 5 seconds, then it moves down further slowly/gradually from there until it hits about 3db. The downstream snrm also drops but not by such a huge amount, its about 1-1.5db on the downstream. I am pretty sure if it wasnt for my buffer on my upstream (attainable of 27mbit currently) then the line probably would be dropping out during calls.
BT will assume I dont have access to line stats so the fact the excess snrm is somewhat hiding the fault I guess they will try to tell me nothing is wrong.
The path I plan to take is probably this.
Report as a broadband HR fault, evidence been the dropping snrm during calls, (since hh5 now reports this info I can do so without worrying about a unlocked modem). I insist on engineer visit, I agree to pay fees if no fault, but I will only accept no fault if the following are true.
All joints physically checked and ok.
Junction box physically checked and ok.
Tie pair physically checked and ok.
DP physically checked and ok.
NTE5 swapped out and no improvement.
Pair swap carried out and has gave no improvement.
Lift and shift carried out and has gave no improvement.
I am aware all the above takes time but the fee's charged are high and so as such I am paying for all that to be done.
If all the above done with nothing showing I will agree to pay fee, I think I am been reasonable as BT cant reasonably prove no fault if they not checking everything, I dont consider a JDSU remote test as extensive diagnosis.
any thoughts before I report my problem?
-
Yeah …… you haven't a hope in hell of getting that exhaustive list fulfilled.
Don't worry, I do understand the frustration involved in trying to get a fault resolved, especially a tiny one like yours. But, I'm only here to give you the benefit of my experience.
You need to first choose which path you are going to take, to report your fault ? Network, or Broadband ?
If it's Network, then (as with all reports), they will run a line test that I'm guessing will pass with flying colours, going off the symptoms of your fault. They will then read the script about potential charges. You accept, and they proceed to build a CDTA (Concsious Decision To Appoint) task.
The criteria laid out for this type of task is …... we perform a PQT, we perform an Eclipse/Fast Test, if these both pass and no audible noise is heard on the line, then it's goodbye and you get the bill. No amount of protestation about SNRm fluctuations will cut the mustard. If it is a CDTA task, that's all we are empowered to do.
If you go down the broadband path, you really are in the lap of the gods as to what engineer you get and what they are prepared to accept as an 'issue' with your circuit. They will have to run the mandatory PQT and Eclipse/Fast Test, but can also run a 'DSL Close-out test' ) for 5 or 15 mins), that pings data about for the allotted time looking for errors. If all 3 tests pass, then again I would guess your ISP will throw the bill for work done, your way ??
This may sound harsh, but it's not meant in that manner. All patches have a couple of, how shall I say it …. Konrado's. Chasing down ridiculous fluctuations that are basically part of the systems management tool. I'm not saying this is you, Chrysalis, but there HAS to be a cut-off point where we say the line is performing within expectations. This is a generalisation, and you may well have a decent HR-type fault brewing, I'm just putting over our remit, as agreed with all the ISP's and Ofcom, as you asked for views before you go ahead and report.
By way of a PS ……… if you do go ahead, I think you'd get more bang for your buck if you go down the broadband path. If you do, request that the engineer use 'Quiet Line Test' and also ring your landline whilst the circuit is undergoing the DSL Close-out test. If there is a HR, then chances are the errors will increment under this particular 'loading' test ??
-
So to be clear on this its within BTs expectations for snrm to drop during calls by the equivelent of what is about 15mmbit of US sync speed of signal?
Also there is in my opinion audible noise present on the line also. If I do a QLN test using my handset. As well as the hiss noise there is clicking noises.
I appreciate you laying out BTs internal policies, of course as a customer this isnt in my t&c's. The t&c's simply state if no fault found I can be charged, on a legal ground generally for that to be upheld would require proper extensive diagnosis, in other words more then likely BT would drop charges if I took it to a high level. So I am not that concerned on that side of things, but more about forcing the tests to be carried out which I suspect will be very diffilcult as you said.
The net is littered with stories about people passing JDSU tests and then engineers proceeding to find faults when digging further. I guess this is what you mean by been in the lap of gods as to whether an engineer chooses to ignore test results or not.
-
Its worse than that - getting BT to accept a fault
About 6 months ago I had a no dial tone fault with associated appalling broadband
Checked at master socket and with different phone: same result
So this is most likely a single line of the pair disconnected.
I ring the BT call centre
They test the line:- Line Test OK - there is no fault sir
Oh really I reply - then why is there no dial tone?
Ah, yes, well, ummm, errrr, do you want us to send an engineer
(Skip the bit about how I will be charged if it is my equipment)
Engineer arrives
What seems to be the problem sir?
(Clearly the BT engineer has been told nothing and has done no checks on the line at the exchange)
Well there is no dial tone I reply
Gets out his BT test phone - checks - indeed there is no dial tone
Ah, yes, mmmmmm
...and the fault
yes, single line dis' in the PCP
I despair!
-
To be clear …… I've never received, seen, discussed, been witness to …….. any form of communication that lays out SNRm parameters, in either its dormant, or active state, with regards to landline usage. So I can't inform you what BT expectations are.
I don't need to go any deeper as to the calibre of engineer you may end up with, it's been done to death and we all know the problems. But, yes, the PQT/Eclipse are not the ultimate conclusive test of a fault present. Although, they are the agreed and defined parameters to which we are tasked to work to. It really is down to whether the fault is present when 'we' are on-site. If it isn't, but raises its ugly head again after we've left, then raise another fault. Painful ?? Yes. But it will get to a stage where the fault will be present whilst we are there, and can then be rectified. All previous visits, however many, can then be written off in terms of billing issues.
Just like spurious faulting techniques littering the net, I'm sure there's probably quite a few 'Konrado's' knocking around as well ?? That's why there are defined parameters in which to test to. Most faults will be captured using these systems, others might need a few more visits before success can be claimed.
If you are adamant you have audible noise on the line, then go for it, raise a network fault.
-
Its worse than that - getting BT to accept a fault
About 6 months ago I had a no dial tone fault with associated appalling broadband
Checked at master socket and with different phone: same result
So this is most likely a single line of the pair disconnected.
I ring the BT call centre
They test the line:- Line Test OK - there is no fault sir
Oh really I reply - then why is there no dial tone?
Ah, yes, well, ummm, errrr, do you want us to send an engineer
(Skip the bit about how I will be charged if it is my equipment)
Engineer arrives
What seems to be the problem sir?
(Clearly the BT engineer has been told nothing and has done no checks on the line at the exchange)
Well there is no dial tone I reply
Gets out his BT test phone - checks - indeed there is no dial tone
Ah, yes, mmmmmm
...and the fault
yes, single line dis' in the PCP
I despair!
Unless he was actually in the Exchange at the time of picking your fault up, he would have no reason to go to the Exchange to test your line. If the remote systems have shown your line to be a 'LTOK', then the most obvious place to start is the EU's premises. As mooted many times before, the remote line test systems quite often won't 'see' a HR, and will return 'LTOK'.
-
Unless the fault is obvious, I wouldn't bother reporting it.
I have a similar fault, SNRM dropping with phone use and intermittent phone noise, although I am also getting related dsl diconnects which is adding to the noise i.e. making the phone very noisy. Reported as a broadband fault as I was concerned the intermittent noise may not be detected and a charge raised. Broadband engineer duly arrived and did some tests which came back normal, as far as I know, which isn't surprising as the fault only occurs with phone use. Fortunately by this time the noise and disconnects were obvious and the engineer proceeded to unsuccessfully look for the fault between my house and exchange.
-
You will only be charged if there is a problem with your equipment. Even if they turn up to your house perform all there tests and they pass, they then checked your equipment is setup correctly. They leave and report it back to your ISP (BT I pressume), your ISP doesn't have the right to charge you. You would only occur the charge if there is a problem with your equipment, I've had a good 5 - 10 engineers turn up who have turned performed the tests and gone. Not once Have I been charged this is because my equipment is absolutely fine and I'm not causing the problem. If the problem isn't with you BT are under a legal obligation to pay for the visit even if the fault wasn't fixed or found.
As BS mentioned it highly depends on the engineer you get, I now know a good majority of good engineers in my area because the amount of times they have had to come out. Also one of the engineers was profounded you could see line stats from the modem. I explained you could unlock the modem and use software/scripts to plot data into graphs if you wish. After a few more different engineer visits I got the same one who was good friends with the chief of the area I live in and he came round with him. Who was also shocked you could get line stats from the modem, they then gave me a HG612 which I showed them how to unlock. If i get one of these engineers I know now they will perform the tests required to mark the job as completed then complete any work I wish to be done to my line without question. This comes down to engineers either enjoy/respect there job to go and find the fault if the tests fail, the others just want the easy money.
-
ryan the policies have changed, it used to be charge if EU equipment fault found. Now its charge if no fault found 'or' EU equipment fault found.
As I said tho I am not concerned about the charge as BT would never uphold it when pushed. But more so i planned to use me paying the charge as bait to make BT do some proper work. I suspect if given the choice 1 - quick JDSU test no charge or 2 - extensive diagnosis 2+ hours checking joints etc. but I pay fee (if no fault found), BT choose #1 every time.
I am pretty sure there is noise and not only noise but a HR fault (unless something else can lower snrm during calls), bear in mind lowering snrm during calls is the same thing as dropping connections during calls except its harder to notice, both are the same thing snrm lowers, except when the connection drops its because its ran out of snrm to maintain the connection.
I think what I will try to do before reporting is leave an upload going and start a call, to see if the throughput is actually affected.
-
Ok BT have found a fault remotely but it was kind of funny.
I rang them up on my mobile, and the test passed, so I ended the call and rang back using my landline meaning the line would be in use. This time the test failed, but the guy said he doesnt know what the failure message is as he never seen it before on any other line so I have been escalated and awaiting a call back. He wouldnt tell me what the error is.
-
Ryan,
Black Sheep pretty much states above that Openreach charges if no fault found. I would have thought that most of these charges do get passed onto the customer by the isp including BT retail.
-
Chrysalis,
I didn't know a remote test could be carried out on a line when in use?
-
Call centre ran a test whilst I was on the phone, so I guess yes.
Also its possible to even do tests on BT's website whilst in use but I dont think is as good as tech support tests.
Anyway I have an engineer appointment arranged now for 2nd week of december.
In regards to charges yes BT retail will most of the time likely pass on charges by default unless the visit is done by the exec office (then its waived by default) but the charges are weak legally since the t&c's state in vague terms if no fault is found, in a court of law a judge isnt going to accept a quick jdsu test as proof of no fault so it would be a no brainer result, and BT wouldnt let it go that far either, they would bail out and drop the charge,.
BTr would still be paying openreach,we have to consider openreach's contract with BTr is completely irrelevant to BT's contract with the end user.
-
Chrysalis,
I didn't know a remote test could be carried out on a line when in use?
A remote PSTN test can not be carried out whilst the landline is in use. A WHOOSH VDSL/ADSL test can be run when the landline is in use. :)
-
I somehow managed to get through to technical support in Milton Keynes. What a difference! My faith in BT has been restored. ;D
-
ryan the policies have changed, it used to be charge if EU equipment fault found. Now its charge if no fault found 'or' EU equipment fault found.
If it's changed within the last 3 weeks but I still believe it is true from my experience.
-
Same here I was under the impression anything inside the customers premises if found to be the fault was chargeable (excluding faulty modems & routers) and anything ouside the premises the ISP picks up the bill.
Would like to see these new ISP chargable conditions :o
-
I think folk are getting confused, which is understandable, as we ourselves can't keep abreast of the ever-changing rules that apply to TRC (Time Related Charges).
The same basic rules still apply, i.e.: You are covered by your line-rental agreement for anything up to and including your NTE5. As long as it's not malicious damage, vandalism etc. We can't be held responsible for the jealous 'Ex' who puts an axe through the drop wire. Yes … it happens.
The changes that have happened, where 'sometime confusion' may occur, is on the CDTA (Concsious Decision To Appoint) tasks. Basically, if the EU raises a fault, the recent 'Line test' history will also be looked at. If there's been evidence that a fault has existed recently, but everything tests ok whilst we are on-site, then the TRC will not apply.
Conversely, if the test history only shows 'LTOK', and still does whilst we are on-site, then TRC will apply.
Please, I can't help anymore than that, as I don't get involved in TRC. It's always going to be a grey area.
-
If you do, request that the engineer use 'Quiet Line Test' and also ring your landline whilst the circuit is undergoing the DSL Close-out test. If there is a HR, then chances are the errors will increment under this particular 'loading' test ??
I'd back this one up. The thing is you need a decent engineer though. One of the early ones I had said that CRCs didnt matter because they auto-corrected and since I didnt have any FECs my line was ok (hmmm my line isnt interleaved ???). Anyhow once I got a decent engineer and by the time voice had started to fail too, the engineer did the above test and said my CRCs were way above what he considered the threshold for a line of this length.
-
if the test history only shows 'LTOK', and still does whilst we are on-site, then TRC will apply.
So basically if you report a line fault to many times and nothing has been found as a fault then your classified as "Cry Wolf" that I can understand wasting resources & time when others need it more.
have a nice link -> http://www.openreach.co.uk/orpg/home/updates/briefings/generalbriefings/generalbriefingsarticles/gen10412.do
-
I can't say, 'Cry wolf', as half the forum would be slating me and OR in no time. ;)
As I say, TRC isn't my forte, but I have just glanced over the 'War & Peace' document that covers this awkward beast. My lord, it's deep stuff.
Of course, there are going to be many genuine 'intermittent faults', that are going to fall under our TRC criteria, that may not be found instantly. But in the same breath, there's many more that fall under the TRC criteria that cover 'EU mis-operation', 'Faulty EU equipment', 'Vandalism', 'Damage on EU's curtilage', 'Flood', 'Fire' etc etc etc ...........
The genuinely faulty ones that are covered by the rental T&C's, will usually get repaired first time, but if it requires a couple of visits to nail it, the TRC's will be waived. A ball-ache, yup, but a necessity to minimise manpower wasteage.
I feel I may have left myself open here to the pedants that will scrutinise every possible scenario. It won't make any difference, the processes in place are there for a reason. To protect both the business and the EU. That's why we all get sick to death of being asked if we've plugged the phone in the test socket, blah-de-blah, but it's asked for a reason.
-
ryan the policies have changed, it used to be charge if EU equipment fault found. Now its charge if no fault found 'or' EU equipment fault found.
If it's changed within the last 3 weeks but I still believe it is true from my experience.
in your case either your isp may have waived the fee or the engineer put in report he found an issue and fixed (I think some do this to prevent charges).
Its in the openreach doc's somewhere regarding the charges, aaisp director revk has also blogged about the charges. It is also in zen's t&c's in more detail than most isp's they specifically state if the engineer cannot find a fault a charge is likely to be raised.
The change was much more than within last 3 weeks, I cant remember the date but its been for a while now.
-
if the test history only shows 'LTOK', and still does whilst we are on-site, then TRC will apply.
So basically if you report a line fault to many times and nothing has been found as a fault then your classified as "Cry Wolf" that I can understand wasting resources & time when others need it more.
have a nice link -> http://www.openreach.co.uk/orpg/home/updates/briefings/generalbriefings/generalbriefingsarticles/gen10412.do
yeah that says it quite clearly, thanks for finding it.
PQT pass and access gained - charges apply.
Although I respect also what black sheep has said that if there is faults on the recent history the fees can be waived, maybe thats what happened to ryant.
Also plusnet absorb some of these fees as I beleive when they do charge they only charge £60 whilst openreach charge the isp more then that.
My point of view in the original post was that for a fee of £115 I want value for money, I think £115 merits more than a 5 minute JDSU test, even if they diagnosing for 2 hours they still getting over £50 an hour, and that the contract between openreach and isp is frequently different from isp to end user.
-
I done the test and its a fault finder.
My thoughput crawls during calls on the upstream, downstream not really affected tho.
That seems more conclusive now black sheep?
19mbit/sec not call.
2-5mbit/sec during call. but takes about 5 secs to start slowing down.
-
I would absolutely say so, based on your findings.
However, (and I have to put this), to all intents and purposes, 'throughput' doesn't fall under our remit. Now that's out of the way, I would hope that you get a decent broadband engineer who is willing to accept you have an issue there ??
Lap of the Gods, I'm afraid. Broadband should have remained a 'Closed User Group' (CUG), as it was during its infancy. The guys who were on it back then, had a genuine interest and were very capable. By moving it into the 'multi-tasking' queue's, it lost that edge.
-
who's remit does it fall under when the slow throughput is caused by line errors?
-
Chrysalis, please try and see through what I'm trying to say here, without me having to type down every scenario.
Throughput is not our remit. If the line is 'erroring', we have to perform various tests as laid out in the agreements, of which throughput is not one of them, and if they pass, then we are at liberty to close the job.
REIN causing extreme amounts of errors, and can slow the throughput down to 128Kbps, but it's jack to do with us.
If your line is faulty, and tests faulty, then by repairing that fault one would hope the throughput would be restored. But I will say it again, throughput (OFFICIALLY) is not our concern.
-
I understand what you saying but its an eye opener.
High errors will reduce throughput, its certianly not the isp's remit as they dont own the line.
I appreciate you posting the internal policies but from a customer point of view this isnt a good situation, openreach sidestepping what should be their problem to deal with, obviously at this stage we have no idea if its REIN, HR fault or something else but seems a HR fault is most likely given my symptons. As I understand tho a HR fault is hard to diagnose, JDSU tests often fail to diagnose it so engineers would typically look for other symptons hence getting back to this dipping snrm, reduced performance and so on. But I believe you are trying to state openreach will only recognise such a fault if the problem is extreme (ie. disconnections). So its like a car sales man selling a car he knows has faults but not telling the customer.
Hopefully i get a good engineer with his head screwed on right and this doesnt blow up. As I wont sit back accepting this if openreach just throw it back saying tough luck deal with it.
If I am wrong and there is such a tests you know off thats reliable for detecting HR, maybe you can tell me and I will ask the engineer to carry it out.
I wish kitz in her own thread before revealed what her engineer told her was an acceptable error rate as that could help other people if they get fobbed off with error rates above that error rate.
-
" As I wont sit back accepting this if openreach just throw it back saying tough luck deal with it."
IF it proves to be RFI, then sorry, that is what you are liable to have to do, since it is outside BT's control. ( HMG seem to have washed their hands of RFI - you might hope that FFTP is their long term plan but I doubt it, and that won't help other services anyway. )
In my professional experience (in another field) the biggest problem with RFI is getting the complainant to consider the possibility that their own installation could be at least contributing to the trouble.
Way back when the GPO investigated complaints of interference to TV for HMG, each year some 80% of cases were due to faulty TV receivers or aerials !
An explanation. I deliberatly use the standard radio term "RFI" because it doesn't have to be impulsive ; you can b-- most things with a white noise jammer !
-
ok but it would be a weird RFI for it to only occur when I am using the phone :)
I have tested different handsets including none wireless ones.
Generally speaking tho its not a good way to do business, if I said to one of my customers its out of my control deal with it (even if speaking the truth), I dont think they would be my customer for much longer.
-
When checking for a suspected HR fault, I have been taught to connect my TDR to the end of the pair (temporarily disconnected from the NTE5/A) and "look" towards the exchange battery. From a mobile phone initiate a call to the circuit under investigation. As a result of the AC "ringing" voltage being superimposed upon the 50V DC bias of the exchange battery, any HR or semiconducting joint will reveal itself by a "movement" of the anomaly observed in the trace occurring in synchronisation with the "ringing" cadences. :)
-
It has to be frustrating to ISP customers when they can't talk to the people directly responsable for the upkeep and quality of ever line in the UK, and you want answers to what the problem is and would like it fixed yesterday.
I myself would like Openreach to be more accountable to every customer that use's there equipment instead of hiding behind ISP's and do think they provide a good service to ISP's but for the end user as in any business transparency works better for all.
-
When checking for a suspected HR fault, I have been taught to connect my TDR to the end of the pair (temporarily disconnected from the NTE5/A) and "look" towards the exchange battery. From a mobile phone initiate a call to the circuit under investigation. As a result of the AC "ringing" voltage being superimposed upon the 50V DC bias of the exchange battery, any HR or semiconducting joint will reveal itself by a "movement" of the anomaly observed in the trace occurring in synchronisation with the "ringing" cadences. :)
thanks I will ask for this test if its not already carried out and the engineer says cant find anything.
-
Ok I dont want to flood kitz forums with new threads so will post this here, and may also post this in the new firmware thread or vectoring thread.
I flashed my modem today with the new firmware. The one with btagent removed and gui added.
Made some graphs, and something interesting has come to light since these graphs now show upstream data I wasnt seeing before.
They show.
Upstream attenuation lower than adjacent downstream tones (BT artifically increasing my downstream attenuation?)
Massively lower QLN on U1 tones, easy to see why, see next.
Massively reduced SNR on U1, power cutback for sure.
SNR on U2 almost as high as end of D1. Can clearly see my U2 has more snr than adjacent downstream tones. This is even same QLN on U2 as downstream this is a reasonable indication the downstream has power cutback on it.
-
Just for curiosity, what do your bitloading graph & pbParams data look like for the same period?
-
I will post them but deliberatly didnt before as they are a way to get sidetracked.