Print

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

[NewtronStar]

I am not a very happy person on FTTC at the moment with SNRM swinging from 5.5 evening to 7.8 in the afternoon and have discovered when I ring the land line up stream errors occur could anyone offer a resaon why this may happen, thinking of calling this in as a fault and it will be my first time ever doing this for broadband.

In the Graph you will see the US green spikes thats me calling the landline from my mobile.

[JGO]

It may be a HR fault on the line.

Sometimes using 17070 to "Ringback" offers a temporary cure/diagnostic but it sounds to me as if it does need reporting as a fault.

P.S.  If a Quiet Line test shows noise that strengthens your hand with BT !

See what other Kitzen advise.

[Bald_Eagle1]

What happens to SNRM:-
a) During the phone ringing
b) when it is picked up
c) when it is put back on hook?

When there was a dodgy joint at my DP, SNRM would increase while the phone was ringing, decrease when it was answered & temporarily stabilise when it was put back on hook.

After hanging up, SNRM would remain stable for a few hours & then gradually reduce until smoeone dialled in again.

It became so bad (particularly during warm, dry weather) that simply picking up the phone to dial out would reduce SNRM to negative values & often cause the connection to resync at lower speed.

All classic symptoms of a HR fault.

[NewtronStar]

What happens to SNRM:-
a) During the phone ringing
b) when it is picked up
c) when it is put back on hook?

The SNRM does not change during a: b: and c: just US CRC spikes when the phone rings, the only thing I have done at my end is install the MKII faceplate three weeks ago but the SNRM swings were there before hand.

[burakkucat]

   Hmm . . . Interesting. A definite fault in the metallic pathway, for sure. Which type of fault, however, is not clear.

As recommended, above, please:

• Perform a QLT (17070, option 2) with a classic, wired, telephone from the "test" socket at the NTE5/A. Listen for any audible noise.
• Perform a "ringback" (17070, option 1) with DSLstats monitoring the circuit. The SNRM should be closely watched.

You may need to wait for the problem to become worse before you finally obtain the "smoking gun" evidence, thus justifying an Openreach engineering appointment.

[NewtronStar]

I dont have a wired telephone to test on 17070 just a dect cordless one will ponder down to a pound shop and get one and run through the QLT option 2

Yeah I have been suspicious of my line now for well over a month and if you remember my line has been picking up alot of RFI in the evening and causing large chunks of tones to disappear and it's getting worse, and the phone errors has deffo confirmed this and I think it's the start of worse to come 

[burakkucat]

I certainly do recall your experiences of RFI during the hours of darkness.

The circuit (DSLAM to CPE) operates in differential mode. For optimum performance and maximum common-mode rejection of unwanted signals, the circuit should have a "good" AC balance. (Black Sheep will be able to give an example of what value and above can be considered to be good . . .) A circuit that has a defect (HR or semi-conducting joint, for example) will not show as "good" AC balance as it should. With less than "good" balance, the common-mode rejection is degraded. More RFI is experienced. If a joint is showing semi-conducting tendencies there is opportunity for cross-modulation (wanted with unwanted signals, unwanted-1 with unwanted-2 signals, etc) and basic rectification of the RF signals present (both wanted & unwanted) in the circuit. Into that mix now add the ringing voltage (80+ VAC at 25 Hz), coupled with a degree of loading on the far-side of the low-pass filter when the telephone is used . . .

End result? One scenario would be exactly what you are experiencing! 

I think it's the start of worse to come 

b*cat does not like to see unhappy kittehs. So view it this way. If it gets worse, it becomes easier for even the less-experienced Openreach technician to trace and fix. Unfortunately Black Sheep does not cover Northern Ireland . . .

[kitz]

It does sound like you have the beginnings of a HR fault.

I hate to be the bearer of bad news, but I very much doubt that BT will do anything about it whilst its just generating errors.  Trying to get anyone to take note is a hard task.   I was getting CRCs and a loss up of to 15dB on the upstream, and even line drops, yet none of their tests seemed to show anything.  It wasnt until the voice failed too that it got fixed. 

I recall the black-cat saying to me something like "never-mind at some point it will fail completely, THEN they will fix it". 
Its so damn annoying when WE know there is something wrong

[NewtronStar]

Thanks BC and Kits I can see where your going with this and will hold out as long as I can at my end.
PS typing this from the Hudl its fun but have to put on the reading glasses it all looks so small compared to desktop 23" monitor 

[NewtronStar]

Just some observation info when the Desktop moniter (23" led) is on the the SNRM will drop 0.5dB and the moniter is 1.5 meters from the HG612 and 0.5 meters from Data socket, the PC desktop on/off has no effect on SNRM just the moniter which is audible at 7283 Khz on the Radio Scanner, my US attainable has been dropping over the past three weeks it's lost 400 Kbps.

[burakkucat]

Hmm . . . I'm not really the person to comment upon radiated energy from a monitor.

However, I have a couple of questions. Are you sure it is LED or could it be TFT? As for the frequency at which it radiates, do you notice a corresponding peak in the QLN graph?

7283 kHz is around tone 1688 - 1689.

[NewtronStar]

Hi it's an LED monitor AOC i235FH, can't distinguish the spikes from the HLOG as it's mixed up with the other RFI spikes and it's frequency has a very wide band.

Here is a short SNRM graph with the monitor on/off you will see the SNRM climb when on and drop when off, only noticed this a few days ago when the monitor went into sleep mode.

[Black Sheep]

As B*Cat has pointed out, it may indeed require your line fault to develop, before positive action can be taken with OR.

Not ideal, as Kitz has pointed out, but as I've said many, many times before ...... there has to be a cut-off point between acceptable faults and not acceptable. For example, one of our esteemed members on this very forum would probably try and raise a fault if his attenuation shifted by 0.3dB. I'm not saying he's wrong, it's just not acceptable from a business perspective.

Likewise, the amount of errors you are receiving now, or the 2.2dB swing in SNR will probably not be classed as 'Service affecting'. Again, I apply the caveat I don't make the rules. The ISP's, Ofcom and BTOR collectively decided on what the rules would be.

Regarding AC Balance, the higher the number, the better. There are various communiques within OR giving differing opinions as to what should try to be achieved, but a general rule of thumb would be 45-50dB or over for PSTN, and 60-65dB or over for digital services. As an on-site engineer though, you have to take into account all the test results and form an educated opinion.

It does appear you have a fault of some kind brewing, if the graphs are a true representation, but I wouldn't be expecting too much at this time. Of course, it's your own shout and you can demand an engineering visit as long as you agree to the potential TRC's (Charges). 

[NewtronStar]

I appreciate your comments Black Sheep and understand fully what your saying and time after time when reading forums relating to possible faults on EU line the OR engineers gives it the all clear and I want to avoid that hassle and wait until the fault is so obvious, though I do have a higher threshold to faults when they occur than some others yet I am no push over when a serious fault needs fixed  but not there yet.

Yep doing all the rounds in house to make sure this issue is not on my side to avoid potential TRC's charges but i think the ISP will still read this out loud from the hymn sheet. 

[Chrysalis]

classic HR symptons, however if you want to try something quickly get a nte5 of ebay and swap it out.

I had pretty much the same thing, a large drop of upstream snrm during calls (didnt matter if in or out) along with occasional SES, swapped the nte5 and now 0 ses after weeks of uptime and no snrm fluctuations.

of course with my upstream attainable been almost 30mbit my connection held out, but for you I expect its more rough.

4+ engineers failed to diagnose my issue as well, is a nightmare dealing with them.

By the way before my first engineer visit plusnet gave me a disclaimer of fees, I replied I accept as long as satisfactory proof is provided is no line fault, plusnet accepted those terms, and I wasnt charged for multiple visits even tho the engineers claimed they had nothing to fix. (first engineer said was HR but then left).