Print

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

[konrado5]

I've edited my above post.

[kitz]

Konrado Im sure sure what youre wanting here.  B*cat has several times said that he's helped as much as he can.   You are asking him to perform remote diagnosis on your line without any tools.

Whilst I appreciate that there may be some tiny anomalies on your line - it is actually running well within the acceptable parameters and this is why no telco will get involved.  Every single line has some anomaly or other . I dont think Ive ever seen a perfect line when it comes to the real world. Its not the same as lab testing - once those lines are in the ground then there are all sorts of variants that can occur. Its not unusual for lines to have different gauge wires along the length.  If b*cat feels that the dip in your hlog is down to different poundage, then I'd tend to take his word for it. 

I honestly don't know anyone that would have gone into so much depth and be able to give you so much knowledge than what he has.. and I'm talking on any forum not just the kitz one. If he says he doesn't know and is at the limit of his knowledge then please leave it at that

[konrado5]

But University of Technology engineer perahps would say something more. Could someone give any contact to the University of Technology engineer who knows much about ADSL? The anomaly is tiny but very curious.

EDIT: I've contacted with John Hearfield (auhtor of following web page) and he says it is rather not gauge changing because gauge changing would give step changes in Hlog but not undulations.
http://www.johnhearfield.com/Telephone/Cable.htm

[JGO]

Kitz, your mention of anomalies reminds me of something which used to give trouble on coax.( eliminated by better manufacture.)

A slightly off-centre wheel in the extrusion machine resulted in slight sinusoidal undulations in thickness & cable impedance.
Could something like this happen in drawing copper wire ?
 NB pure speculation !!   

[boost]

But University of Technology engineer perahps would say something more. Could someone give any contact to the University of Technology engineer who knows much about ADSL? The anomaly is tiny but very curious.

EDIT: I've contacted with John Hearfield (auhtor of following web page) and he says it is rather not gauge changing because gauge changing would give step changes in Hlog but not undulations.
http://www.johnhearfield.com/Telephone/Cable.htm

Did you actually get a reply from some random, in 10 minutes, about a web page he appears to have put together 2 years ago?

[konrado5]

I've get reply two weeks or week after I've sent mail.

[konrado5]

If you are asking me for my further thoughts on the observation I will say that, although impedance changes do occur with changes in wire gauge, it could be that the undulating roll towards the higher frequencies in your Hlog plot are the secondary (and higher) effect of the presence of a bridging tap at the lower frequency end of the plot. However it is inconclusive due to the dip that is often apparent when transitioning from the US frequencies to the DS frequencies.

I've made more accurate graph of only first 100 tones. Upstrem Hlog is more accurate. Is bridge tap visible on upstream? I know that notch is not caused by bridge tap.

EDIT: My observations indicates that correct downstream attenuation value (on all lines) is from around tone 40.

Best regards
konrado5

[burakkucat]

I really do not know what else I can provide as material input to your quest. That said, I have just a couple of comments.

Firstly, in relation to the second of the two latest attached pictures, above. When viewing that second picture in its native resolution I see that there is a significant length of the cable with the outer sheath removed and the internal wires exposed. In that condition the two twisted pairs of the cable can become untwisted, thus being a region of the cable that is not according to its specification. If that length of the cable was just reburied by the replacement pavement and was not replaced, then any number of "peculiar" effects my be found in the operation of the circuit.

Secondly, looking at your latest Hlog graph of the first 100 tones. It is quite easy to figuratively "remove" the notch (between US & DS tone) by eye. Once that has been done, I do not see anything that resembles the (initial, primary) notch which would be the characteristic "fingerprint" of a bridging tap. Such a notch would show a distinct "V" profile . . . and that is not present in your Hlog graph.

You have presented me with sufficient evidence such that I have to rule out cable gauge changes (i.e. essentially impedance changes) as the origin of the "undulations" present in your Hlog graph. Now you have shown me that there is no evidence for a bridging tap. What else is there to account for the appearance? I have one suggestion . . . Could it be due to the cable having its two twisted pairs being effectively untwisted in one or more regions of its length? (As visible in the second of the two above pictures.) I really do not know. 

Purely for reference, I attached two Hlog graphs obtained from the same circuit. The first with a bridging tap present, the second with the bridging tap removed.

[konrado5]

Firstly, in relation to the second of the two latest attached pictures, above. When viewing that second picture in its native resolution I see that there is a significant length of the cable with the outer sheath removed and the internal wires exposed. In that condition the two twisted pairs of the cable can become untwisted, thus being a region of the cable that is not according to its specification. If that length of the cable was just reburied by the replacement pavement and was not replaced, then any number of "peculiar" effects my be found in the operation of the circuit.

This region of cable was damaged by heavy equipment what in turn resulted in the lack of Internet access during 24 hours. Engineers fixed it. However I've noticed no change in the Hlog graph after it. This cable was damaged some days later than temporary one day change in the attenuation (23.5 dB/20.2 dB instead 25.0/14.8 dB). The cable from the photo is not significant.

Once that has been done, I do not see anything that resembles the (initial, primary) notch which would be the characteristic "fingerprint" of a bridging tap. Such a notch would show a distinct "V" profile . . . and that is not present in your Hlog graph.

Notch around tone 100 resembles for me bridge tap notch. Only difference is that notch is not as deep as in your photo.

Could you look once again at the my usual Hlog graph to tone 100 and one day hlog to tone 100. This second graph has unfortunately false upstrem Hlog datas (at that time I didn't know about DELT mode) but I know at that time I've had significantly higher upstream attenuation (20.2 dB instead 14.8 dB) and significantly lower attenuation at the lower downstream frequiences (significantly lower notch around tone 100) and lower attenuation before the notch. I'm not sure if temporary change was caused by pavement workers. They then didn't damaged cable on the photo. Perhaps it was accidence. Perhaps my ISP switched something in Central Office for one day.

Best regards
konrado5

[roseway]

Konrado, you're pushing too hard again. Please take note of what Kitz said above. You are still demanding too much from one helpful member.

[konrado5]

Konrado, you're pushing too hard again. Please take note of what Kitz said above. You are still demanding too much from one helpful member.

OK, I'll try to contact with some University of Technology engineers.

[burakkucat]

Konrado, you're pushing too hard again. Please take note of what Kitz said above. You are still demanding too much from one helpful member.

OK, I'll try to contact with some University of Technology engineers.

Please do. And then let us know what you discover, in due course.

(I really can not think of anything else that may assist you in your investigations.)

[konrado5]

Could you give e-mail addresses? I've contacted with Slawomir Kula (polish engineer who written book about xDSL). He once replied that it is need to check router and DSLAM impedance but I've written that I've had it on two different routers but he has not replied again. Other polish engineer written me that it is required to TDR test. However it is very expensive.

[konrado5]

I made more accurate graph for my typical Hlog and one day temporary Hlog: first two graphs for tones 33-100, second two graphs for tones 100-200, third two graphs for toes 200-300, fourth two graphs for tones 300-400 and fifth two graphs for tones 400-511. I see significant differences not only on the 33-100 graph. Something what causes my signal reflections is such that it at one day caused slightly other whole Hlog.

[konrado5]

Attachments