2-Wire Logs

[Tacitus]

Can anyone tell me what "VCXO Frequency Offset -120.0 ppm   Suspicious - hardware frequency mismatch" means.

It's in the management and diagnostic console at:  http://home/xsltPAGE=J42&THISPAGE=J21&NEXTPAGE=J42

[Rik]

Is this firmware 6, Tac? I can't get that page.

[Steve]

It may be something to do with a dodgy filter???

[Tacitus]

Is this firmware 6, Tac? I can't get that page.

Firmware 5 5 Rik.  Wouldn't have thought it was a dodgy filter as it's plugged directly into the XTE faceplate.

RStats recognises the 2 wire but tells me that javascript is not enabled.  Just to be sure what should I specify as the 'stats' page?  V5 firmware

[Rik]

I use:

http://192.168.1.254/xslt?PAGE=B02&THISPAGE=B01&NEXTPAGE=B02

[Tacitus]

That seems to be producing something as RStats now showing some figures. 

The graph is currently showing an RX SNR margin of between 1 & 0 whilst the 2-Wire's own page shows a steady 15dB down and 18dB up.  Nothing shows up on the RStats speed graph so I'm wondering if v6 is not entirely compatible.

My theory is that during early morning there is so much noise the modem can't lock on to the signal.  Once the noise goes down sufficiently it then syncs.

[Rik]

Have you tried RouterStats Lite, Tac, it claims to be compatible with 2Wire BT2700HG-V (firmware 6.1.1.48).

[Glenn]

I'm confused, is it a V5 or V6?

Firmware 5 5 Rik.  Wouldn't have thought it was a dodgy filter as it's plugged directly into the XTE faceplate.

RStats recognises the 2 wire but tells me that javascript is not enabled.  Just to be sure what should I specify as the 'stats' page?  V5 firmware

That seems to be producing something as RStats now showing some figures. 

The graph is currently showing an RX SNR margin of between 1 & 0 whilst the 2-Wire's own page shows a steady 15dB down and 18dB up.  Nothing shows up on the RStats speed graph so I'm wondering if v6 is not entirely compatible.

My theory is that during early morning there is so much noise the modem can't lock on to the signal.  Once the noise goes down sufficiently it then syncs.

[Rik]

I'm always confused.

[Glenn]

It's your job.

[Steve]

Memory fails me but I thought that you could customise the data on router stats and point it at the downstream margin figure

[Rik]

It's your job.

[Rik]

Memory fails me but I thought that you could customise the data on router stats and point it at the downstream margin figure

That's how I do it.

[Tacitus]

Right now RouterStats has got into a loop and is using 100% CPU.....  Something is not right  

Bit late now but it might be worth going the V6 firmware and RSLite route.

Thanks for the help guys - I'll get there in the end.  If only they did a version for Macs - or Linux 

[MisterW]

Can anyone tell me what "VCXO Frequency Offset -120.0 ppm   Suspicious - hardware frequency mismatch" means.

The 2Wire MDC displays a value for VCXO Frequency Offset which represents the difference in crystal frequencies between the devices at either end of the line.
The spec is 0-150ppm but normally you would expect values up to around the 20ppm mark, A value of 120ppm indicates a problem somewhere.

[Tacitus]

The 2Wire MDC displays a value for VCXO Frequency Offset which represents the difference in crystal frequencies between the devices at either end of the line.
The spec is 0-150ppm but normally you would expect values up to around the 20ppm mark, A value of 120ppm indicates a problem somewhere.

Very interesting    The big question is exactly where the problem is.  I can't see that it's the XTE-2005 faceplate since the signal goes straight through without being filtered as the router is plugged into the RJ11 on the main socket. 

That leaves a router fault but the same fault - takes a long time to sync on first power up - appears to exist in both my regular Zyxel and the 2-Wire that replaced it.  James in support ran a full set of tests and found no problems.  However, if we accept the same fault is unlikely, although not impossible to appear in two routers, then perhaps their is a problem at the exchange.  Frankly I'm out of my depth with a lot of the tech stuff surrounding DSL, but that diagnosis seems logical to me.

[Steve]

Filterplates can become faulty, have you tried the test socket and looked at the time difference if any it takes to obtain a sync

[Tacitus]

Filterplates can become faulty, have you tried the test socket and looked at the time difference if any it takes to obtain a sync

I get more or less the same result using the test socket Steve, although I suppose I could try a different faceplate.  TBH I'm not all that keen on the XTE faceplate filters as they appear rather more complex electronically (aka more to go wrong) than the equivalent from Clarity, although they are supposed to be better at filtering.  

A normal sync from powerup should, at most, be within 5 minutes.  Once you head towards an hour or more, something somewhere must be wrong.

What I will do over the weekend is to try the original router on Sis iDNet line.  If it syncs normally it would suggest the problem is not with the router itself.

That said, I left the 2-Wire running overnight; intriguingly it's still synced and, the noise margin has dropped, so I think I'll leave well alone for a while.

Oh the joys of ADSL...  

[EDIT]  Hmm 10 minutes later the SNR has shot up.  It is breakfast time I suppose.

[Tacitus]

I'm confused, is it a V5 or V6?

For the record it's V 5.29.107.12

[MisterW]

However, if we accept the same fault is unlikely, although not impossible to appear in two routers, then perhaps their is a problem at the exchange.  Frankly I'm out of my depth with a lot of the tech stuff surrounding DSL, but that diagnosis seems logical to me.

That would be my guess too. Given that it appears to be the same from the test socket and with different routers, then that eliminates filters,faceplates,house wiring and equipment, that only leaves a line or exchange fault. In my experience with 2700's if the DSL diagnostics indicates that any of the parameters are not 'Ok' its usually right!! 

[Tacitus]

That would be my guess too. Given that it appears to be the same from the test socket and with different routers, then that eliminates filters,faceplates,house wiring and equipment, that only leaves a line or exchange fault. In my experience with 2700's if the DSL diagnostics indicates that any of the parameters are not 'Ok' its usually right!! 

James ran a full series of tests yesterday which showed no faults and the line was fine.... 

Once I have sync, both routers seem to hold on to it, the 2-Wire perhaps slightly better.  The problems starts when I switch them off and they need to sync from poweron.  A difference in frequency at each end could well explain that. 

If I leave the router on then things seem fine, so I guess support would not want to pursue this with BT in case of the possible charge.  Problem with all BT resellers, they aren't the masters of their own domain.  If BT stick up two fingers there's not a lot they can do...

[Polchraine]

James ran a full series of tests yesterday which showed no faults and the line was fine.... 

Once I have sync, both routers seem to hold on to it, the 2-Wire perhaps slightly better.  The problems starts when I switch them off and they need to sync from poweron.  A difference in frequency at each end could well explain that. 

If I leave the router on then things seem fine, so I guess support would not want to pursue this with BT in case of the possible charge.  Problem with all BT resellers, they aren't the masters of their own domain.  If BT stick up two fingers there's not a lot they can do...

The original post with "VCXO Frequency Offset -120.0 ppm   Suspicious - hardware frequency mismatch" is pointing to a hardware fault/issue that could stay the same or get worse.    The exchange will use a very high precision and stable oscillator that is likely to be referenced against a national standard source.       The oscillator in the modem will be fairly cheap but can be tuned to match the exchange by applying a small voltage to the oscillator and the modem should remember what voltage when switched off .    The modem has detected that the oscillator centre frequency is way off but has corrected it and it may be drifting around which will cause problems on reboot when the VCXO setting previously saved is not appropriate and the frequency is mismatched and it is possibly unable to stabilise quickly enough resulting in a reduced sync speed and increased errors.

On the V6 it is possible to see the history - I can go back 20 events and see the VCXO offset - mine varies from -12.7 to -13.6.   Is yours always -120 or does it go up and down?

If you are having initial sync problems on reboot and maybe slow speeds then I would suggest a reboot to establish a session and allow a few minutes for the modem to settle down and components to warm up.   Then do a reset http://aaa.bbb.cc.dd/xslt?PAGE=login_post and select Reset Broadband           Reestablishes your broadband link (physical and logical)    That should re-establish the link fairly quickly with your VCXO running at a frequency close to the exchange

[Rik]

Thanks for that. 

[Tacitus]

Thanks for this very informative post 

On the V6 it is possible to see the history - I can go back 20 events and see the VCXO offset - mine varies from -12.7 to -13.6.   Is yours always -120 or does it go up and down?

It appears to be stuck at -120

If you are having initial sync problems on reboot and maybe slow speeds then I would suggest a reboot to establish a session and allow a few minutes for the modem to settle down and components to warm up.   Then do a reset http://aaa.bbb.cc.dd/xslt?PAGE=login_post and select Reset Broadband     Reestablishes your broadband link (physical and logical)    That should re-establish the link fairly quickly with your VCXO running at a frequency close to the exchange

Just tried that.  Reboot, leave for a few minutes to allow the session to establish and then did the Reset.  Apart from the sync going down a little everything remained the same.  Sync was down a little but the sync was re-established quite quickly, certainly within normal parameters. VCXO remained at -120

So it seems the long sync time only happens when the router is powered up.

[Tacitus]

I've managed to get RouterStats Lite running with the 2-Wire; it doesn't give anything beyond SNR and sync speed but at least it lets you do a plot. 

Been running overnight and, as of 7AM the SNR has varied from 15dB with occasional drops to 14dB.  Connection speed has remained rock solid at 2140 which is very good for my line, although once it rains that will go to pieces in short order. 

The original problem of why it should take so long to sync remains unresolved.  Unfortunately I have to assume that something has failed in both modems.  I'll be round at Sis place this weekend so I'll try using it on her iDNet line and see what happens.

[Rik]

 

[Tacitus]

To update this thread, I tried the router on Sis line and it synced in 65 seconds!  On that basis it doesn't appear to be the router at fault but something to do with the line or the exchange. 

I left it off overnight as suggested with the BT work going on and again it took some 46 minutes to establish sync at 1280.  At this point it was reporting a SNR of 25dB.  A reboot brought the connection rate up to 2112 and the SNR dropped to 15dB.  Again it reports VCXO at -120 and suggests a hardware frequency mismatch. 

Once the connection is up it generally stays up although the speed is slowly dropping.  Whether this is due to excessive noise or the frequency mismatch I have no idea, but the problem is unlikely to go away.

[Rik]

Tell support what's happening and what you've done to eliminate the router, Tac.

[Tacitus]

Tell support what's happening and what you've done to eliminate the router, Tac.

Will do Rik