Print

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

[Black Sheep]

I think you have more or less pointed at what is the probable issue, Frog ...... (the Exchange equipment) via your astute deductions.  Of course, nothing is ever guaranteed ..... one can only use the info provided and experience, to try and give a ball-park solution. 

The higher-frequency circuits (ADSL, VDSL, Private Wires, ISDN30 etc), tend to use the outer circumference of the wire, sometimes termed 'Skin-effect'. Whereby the phone will transmitting nearer the centre of the wire. So if you can imagine, any kind of physical fault on the wiring (such as corrosion) will obviously occur on the outer circumference first, thus affecting 'BB' services before 'Voice' services.
So, it is highly-unlikely to have a network fault on the wires, that only affects 'Voice' ..... and not 'BB'.

Regarding the question about "checking at 1.6Khz during the PQT" ?? Yes it does, it's actually one of the first checks carried out, after first checking for high voltage on the line.

It is portrayed on the hand-held testers that we use when performing a PQT as 'Insertion Loss'. An average loss when testing from the master socket towards the Main Frame in the Exchange (MDF) is 2dB per km of pair length @ 1.6kHz. The engineer won't need to do any calculations as it is all done for him and like all the other tests performed within the PQT ...... will present itself as a pass or a fail.

The only other thing we are skirting around, but which you have touched upon in your last post, is the fact that everyone's hearing is at differing levels. One man's 'Quiet line' is another man's 'Normal line'. Of course, I'm not suggesting you are 'Mutton' ........ 

[Black Sheep]

No need to mention broadband, or long line etc, to the call-centre advisor or the visiting engineer.

This seems apt:



4/5ths of progressing anything with a carrier is filtering. Don't allow yourself to be incorrectly filtered, if at all possible

Exactly the point I'm trying to make, not just for Frog but anyone reading in interest ?? Well played, Boost. 

[boost]

Thank you, Sir!

[Frogparty]

4/5ths of progressing anything with a carrier is filtering. Don't allow yourself to be incorrectly filtered, if at all possible

Not sure how to stop it. I pressume that unless I can provide some measurements, or have something that BTOR can measure, I am going to be put in the "going deaf but in denial" or "time wasting" category.

The only other thing we are skirting around, but which you have touched upon in your last post, is the fact that everyone's hearing is at differing levels. One man's 'Quiet line' is another man's 'Normal line'. Of course, I'm not suggesting you are 'Mutton' ........ 

What highlights the issue is that for several days a month I am in a different location. Same handset (yes I have carried them with me) but audible conversations, when I call friends they know where I am because of the faint audio or lack of it. Without this regular contrasting experience to compare with  I could begin to think I was mutton!

But I need a measurement. Not in a rush to spring for an expensive decibel meter, but might try with an app on my Nexus 10 - a simple comparison with output from same handset here and elsewhere. Hope it is sensitive to pick up the output from the handset. Any other suggestion?

[Black Sheep]

Not me personally, I'm afraid. 'Faint Transmission' on 'Voice' does exist ……. granted not a lot ….. but it does happen. I would be pressing your CP to request a LIC/EN change. It depends on which type of digital system your 'Voice' is fed over as to whether it's a LIC (AXE10/System Y) …. or EN (System X) ??.

I understand your predicament, but you shouldn't have to be purchasing audio-level meters to have a simple fault repaired, in my humble opinion !!

[burakkucat]

[somewhat off topic]
If only something like an Oscar (or a similar Oscillator/LMS pair) could be deployed at the MDF and another at the NTE5/A . . . 
[/somewhat off topic]

With regards to the DP and cabinets, perhaps you could take pictures of all the relevant "gubbins" and arrange for the images to be viewable? I'm thinking down the route that there may be something non-standard (or perhaps less of the everyday set-up) and by viewing the images one of us may have a sudden "light bulb switched on" moment. 

[Frogparty]

I've had another OR engineer site visit, this time a PSTN guy (rather than BB). I expalined our deductions that the faint transmission was possibly an exchange related fault. He repeated the PQT and, just as before, my line was within parameters. In frustration I phoned a friend and got the engineer to have a chat - yes it was initially OK if a little faint but it got fainter as they talked. So yes, he could see there was problem!! But could he find the cause and fix it?

How does OR go about finding the cause of an issue the test equipment does not register? Contractually (BTOR to BT) the line passes the tests (big green tick on test kit) so there is nothing to be done. But thankfully the OR engineer liked challenges and helping people, and I do hope his boss is OK with the 4 hours he spent on my "unmeasurable" but very real problem.

First he replaced the NTE5, 2 measurements had been boarderline and they improved a tadge - A/c Balance = 57, Dis Capa = 470nF. Then he went out along the line to cab etc. testing and recrimping some of the contacts. No measurable problems on the line were detected at any point. That kind of seems obvious to me, but I hope I have benifited from the tidy-up. Finally he left saying there was nothing more he could do but if something ever came to him he would come back.

I appreciate his efforts, but I wish he had looked at the exchange equipment rather than the line. Are linesmen allowed into the exchange? Who maintains that kit? Sadly it seems that there is no way forward with this fault. It is not the line, it does not show on a PQT, but use the connection for conversation across the PSTN and it is faint.

I am exploring an alternative - BT SmartTalk. Turns out they offer a free Andriod app, intended for use on 3G phones, that enables you to make VoIP calls at your land line tariff from an Andriod device. Caller ID shows an associated mobile number, but the app does not have to be on that device. I registered and made a call using it on my Nexus 10 - audible conversation, oh joy! The mic is crap so will get something to plug in, but it seems a way forward.

And finally an apology. I may not be deaf, but I could be going blind! Turns out that the FTTC ECI cab is next to the orginal cab.

[boost]

Sorry to hear about your continued strife on this!

I wonder if someone can clarify something for me.

21CN = let's send everything over a broadband line. Voice. Data. The works!

Do we know if what we presume to be a stone age voice switch that your PSTN handset attaches to, is actually a stone age style voice switch/PBX or a new fangled 21CN style device?

Let's assume stone age voice switch. How are voice calls currently being 'back hauled' across ze network? Actually, how was it done 20 years ago? How do I get a voice call from Lands end to John o Groats?

Is there a test for voice faintness that someone could do and then process? Would help to know what to ask for, otherwise I will just get more of the same (and pay as a time waster).

Funny you should mention it. Yes there is. There's a catch though... voip traffic only OR native PSTN audio converted to some digital format and transported across ze internetz?

Code: [Select]

Mean Opinion Scores (MOS)

The quality of transmitted speech is a subjective response of the listener. Each codec used for transmission of Voice over IP provides a certain level of quality. A common benchmark used to determine the quality of sound produced by specific codecs is MOS. With MOS, a wide range of listeners have judged the quality of voice samples sent using particular codecs, on a scale of 1 (poor quality) to 5 (excellent quality). The opinion scores are averaged to provide the mean for each sample. The table below shows MOS ratings and the corresponding description of quality for each value.
Table 2 MOS Ratings
Score
Quality
Description of Quality Impairment
5
Excellent
Imperceptible
4
Good
Just perceptible, but not annoying
3
Fair
Perceptible and slightly annoying
2
Poor
Annoying but not objectionable
1
Bad
Very annoying and objectionable


From: http://www.cisco.com/c/en/us/td/docs/ios-xml/ios/ipsla/configuration/15-mt/sla-15-mt-book/sla_udp_jitter_voip.html#GUID-6507B56A-F809-46D6-BC51-780E325A3782



What do I mean?
http://www.broadbandbuyer.co.uk/store/voip-adapters/

Devices have been available for donkeys to auto-convert your voice calls into an RTP stream to be sent over the internet. I am making the wild assumption that this is what BT do, on a grand scale, for any exchanges that don't satisfy the commercials for a rip and replace to a shiny MSAN etc?

Where am I going with this?

There is a tiny chance, this is not a voice port issue at all but the transport network making a pigs ear of sending your voice traffic across it. Voice traffic is always top priority, assuming the source device is applying the correct markings to ensure it gets the red carpet treatment across the entire network.


Anyway, you can probably ignore this post completely because we're going to find out you actually have a stone age switch coupled to stone age backhaul

Still, it's always good to consider alternatives?

It's always worth remembering why VoIP will never replace PSTN completely.
Emergency services.
You may want to pursue this angle when you relog the fault for the 14th time

[d2d4j]

Hi

I hope you don't mind, but we had a client who had similar issue, voice fading etc and it was resolved by a new phone.

However, they were using wireless phones, so not sure if it applies here or if another phone has been tried.

Many thanks

John

[boost]

I forgot to mention on my last, largely pointless, post that choosing an inferior target codec is more likely to produce quiet and crap audio.

The markings thing is usually associated with choppy voice calls.

[Frogparty]

There is a tiny chance, this is not a voice port issue at all but the transport network making a pigs ear of sending your voice traffic across it.

Interesting idea. I presume that my phone call transmission gets digitised at some point? How does a standard PSTN voice call get around the country?

It still leaves me with the problem of getting BTOR to investigate a rare problem that does not show on a PQT, and is not a line or user equipment issue. It seems that there is no obligation for them to do anything line related or otherwise if the line tests within parameters. Aggh!!

Oh yes John, not a handset problem, I have tried several as it was the first thing I thought could be the cause. In fact the engineer experienced the faintness on his test handset. But always happy for input and this thread is getting long to read all the details.

Even if BT SmartTalk via Android tablet, home BB and phone tariff enables me to make (outgoing) audible VoIP calls for no extra cost, I still want to pursue this issue if I can find a way to do it. When I get the chance I intend to trace every line from the DP and ask the all the houses on the same pole if they have a faintness issue. One way to meet the neighbours - nearest is 400m away and others further, so it is not that I have been unsociable not talking to them yet.

It's always worth remembering why VoIP will never replace PSTN completely.
Emergency services.
You may want to pursue this angle when you relog the fault for the 14th time.

In an emergency rest assured I would make myself heard!! I have had cause to call each of the emergency services at some time or other, in all cases I used a phone box. The last couple were in the mobile phone era but in locations without signal. Rural call boxes do have an essential use even if they are a loss maker for BT. But that is another topic.

[burakkucat]

Perhaps Black Sheep will be able to assist with further comments but, as I understand things, Openreach deal with the local network from the EU's NTE5/A to the exchange MDF, inclusive. Operate deal with everything exchange based, past the MDF deeper into the system (and also the fibre cabinets out in the streets).

The local network (the "first mile" or "last mile", which ever way you care to view it) for telephony is purely analogue. Once beyond the MDF, it finally meets an ADC in the System X or System Y exchange equipment. The analogue telephony signal is now a PCM "signal" and remains in that format either across a junction to another local exchange or to a GSC/SSC and then onto the trunk network. Eventually that PCM "signal" arrives at the destination exchange (System X or System Y) and a DAC finally converts it back to analogue. The analogue signal then travels through the MDF, the local network and reaches the NTE5/A of the called EU.

[boost]

Interesting b*kat!

Not familiar with PCM but a quick Google reveals it is just G711
I did have a dabble with transcoding G711 to G729 once. I was not impressed with the audio quality, though.

If you look at this table:

http://www.en.voipforo.com/codec/codecs.php

Some of the codecs have MOS scores attached to them. Lower = worse audio quality, I believe.

[burakkucat]

My brain does not retain Gnumbers. Hence the reason why I always refer to G.DMT, ADSL2, ADSL2+, VDSL2, PCM, etc, etc.

If you are right re: MOS numbers (and I see no reason why not to believe you    ), then going from PCM (G711) (with MOS 4.1) to G729 (with MOS 3.92) will cause some degradation is audio quality.

The bandwidth of PCM is 3.1 kHz (300 Hz the lower limit and 3.4 kHz the upper limit) and the sampling rate is 8 kHz. So definitely not Hi-Fi, just ordinary telephony. 

[Berrick]

Just to clarify?

'Skin-effect'. Whereby the phone will transmitting nearer the centre of the wire. So if you can imagine, any kind of physical fault on the wiring (such as corrosion) will obviously occur on the outer circumference first, thus affecting 'BB' services before 'Voice' services.
So, it is highly-unlikely to have a network fault on the wires, that only affects 'Voice'

Makes sense but the way I read this staement it assumes no joints or connections.

Where theres a joint/connection there is impedance and if dis similar metals are involved corrosion both these factors, where joint/connection are present, could surely effect higher and lower frequencies in different ways. IE little or no apparent effect on the BB whilst attenuating voice?