Print

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

[Ronski]

. . . a shorter copper wire with grounded coaxial sheilding to CP premises will help reduce these effects.

Please remember that the xDSL circuit is (as close as possible to being) a balanced pair, operating in differential mode. So if you are proposing screening, you will need to use STP (screened twisted pair) cable. Why is differential mode used? Because any spurious, alien, signal induced is very likely to be induced onto both wires of the pair. Common mode rejection of the alien signal occurs whilst the required signal still "gets through".

The cable which runs from my original master socket location (now just a blanking plate with jelly crimps) to it's new location a few meters away is STP, I've often wondered whether the shielding drain wire should be connected to something? What's your thoughts? I may have already asked this, I've a nagging felling I have......

[burakkucat]

Hmm . . . I will say that it should definitely be connected to a "good" earth (and not via the nearest convenient mains socket earth terminal). I am undecided as to whether both ends should be connected to earth. I'm sure there have been discussions on this topic in the past and, I think, Rizla had some convincing arguments -- one way or the other. 

[Chrysalis]

Hi we all know the SNRM drops during the evening and attainable sync rate seems to follow.
Just would like to get and idea on how much it swings from 11am and 6pm don't need any graphs just your DS & US SNRM dB at the times in my post.

thanks

snrm doesnt always drop in evenings, seems more common when loop loss is higher and on overhead lines.

On adsl I had noticeble drops of snrm at night but on my vdsl it barely budges and even sometimes goes up due to outside interference been reduced.

[ryant704]

Hi we all know the SNRM drops during the evening and attainable sync rate seems to follow.
Just would like to get and idea on how much it swings from 11am and 6pm don't need any graphs just your DS & US SNRM dB at the times in my post.

thanks

snrm doesnt always drop in evenings, seems more common when loop loss is higher and on overhead lines.

On adsl I had noticeble drops of snrm at night but on my vdsl it barely budges and even sometimes goes up due to outside interference been reduced.

I have the same, some nights can drop by 3/4DB (DS & US), some nights it won't move at all sometimes it rises to 7/8DB from 6.

[Ronski]

Hmm . . . I will say that it should definitely be connected to a "good" earth (and not via the nearest convenient mains socket earth terminal). I am undecided as to whether both ends should be connected to earth. I'm sure there have been discussions on this topic in the past and, I think, Rizla had some convincing arguments -- one way or the other. 

I could run an earth from one end into the main earth bar in the CU, would that be acceptable? I have some 4mm earth cable in the garage, but that may be a bit  ott.

[roseway]

I have my doubts that earthing the screen will make any difference, because even when it's not earthed it still constitutes a Faraday cage. If you do want to earth it, then I'm fairly certain that one end only is the right way to do it, and the important factor is that the earth connection should have low reactance at DSL frequencies (reactance is frequency-dependent resistance). The gauge of the wire is fairly unimportant, but a run without lots of twists and bends would be best.

[sevenlayermuddle]

Hmm . . . I will say that it should definitely be connected to a "good" earth

With apologies to Eric, I agree with above.

It's a long time since I studied such things, but I kind of thought a Faraday Cage was only effective if it encapsulated the entire system.  In the case of DSL, it would need to encapsulate not just the modem cable, but also the router, all of the BT wiring to the exchange, and the DSLAM.   That would be a very large cage.

I also recall reading arguments that an ungrounded screen could actually make matters worse as it presents a larger surface area to pick up interference, which could then be induced into the conductors within (made possible because the equipment they connect are not entirely within the cage).

As to whether there is any benefit at all from screened cabels to DSL, grounded or not, I would be highly sceptical.

But I do confess it's a long time since anybody expected me to be a reliable source on things, I may be mistaken.   Talking 'Twaddle', in other words 

[NewtronStar]

its seems very hard to judge the SNRM's swings and sways as it looks like every line has it's own unique pattern going on, but kind of getting this is more down to weather conditions and common local interference and the Ionosphere only plays a small part, so that's three random number generators that effects your SNRM day in & day out.

anyway on a lighter note i had my first SES error, may have had loads before but Ronskies GUI now is able to display them 

[bbnovice]

I wasn't intending to post but your comment about SES struck a chord.

Attached are recent 24 hour graphs for the US/DS SNRM of my connection. This is a typical set of graphs ever since OR fixed a lot of cable problems that previously existed between the FTTC cabinet and the home.  As you can see the SNRM hardly seems to vary, but I do have concerns over my ES and SES counts.

Over the past 19 days I have had 3240/1592 down/up ES, but more wrorringly 127/1 down/up SES. Yet the connection never seems to be significantly affected (touch wood).



 

[NewtronStar]

I wasn't intending to post but your comment about SES struck a chord.

Attached are recent 24 hour graphs for the US/DS SNRM of my connection. This is a typical set of graphs ever since OR fixed a lot of cable problems that previously existed between the FTTC cabinet and the home.  As you can see the SNRM hardly seems to vary, but I do have concerns over my ES and SES counts.

Over the past 19 days I have had 3240/1592 down/up ES, but more wrorringly 127/1 down/up SES. Yet the connection never seems to be significantly affected (touch wood).

Thats a nice SNRM graph straight lines are good and if connection is good no need to worry untill the next modem resyncs to see if any settings have been forced upon your line.

[burakkucat]

I could run an earth from one end into the main earth bar in the CU, would that be acceptable? I have some 4mm earth cable in the garage, but that may be a bit  ott.

Having "slept on it" I'll now say "why bother?", as 7LM has nicely summarised the situation --

It's a long time since I studied such things, but I kind of thought a Faraday Cage was only effective if it encapsulated the entire system.  In the case of DSL, it would need to encapsulate not just the modem cable, but also the router, all of the BT wiring to the exchange, and the DSLAM.   That would be a very large cage.

In other words, will you see any benefit from (correctly) screening just a few yards (or metres, if you wish) of your circuit?    I think not. 

That is, of course, discounting any placebo effect. 

[Ronski]

Thanks for the replies about the screening, I won't worry about it, One less job to do.

I'll post a screenshot tomorrow of our ses errors at work, well over 100.

[burakkucat]

Over the past 19 days I have had 3240/1592 down/up ES, but more wrorringly 127/1 down/up SES.

How about putting that into some form of perspective?

Nineteen days is (19 * 24 * 60 * 60) seconds. So, 127 seconds of that total is a whopping (127 * 100) / (19 * 24 * 60 * 60)%. I.e. 0.0077%

I wonder if the "excessive, obsessive Eagle effect" has mutated into something extremely contagious? 

b*cat goes and hides from the attentions of that beak! 

[Chrysalis]

its seems very hard to judge the SNRM's swings and sways as it looks like every line has it's own unique pattern going on, but kind of getting this is more down to weather conditions and common local interference and the Ionosphere only plays a small part, so that's three random number generators that effects your SNRM day in & day out.

anyway on a lighter note i had my first SES error, may have had loads before but Ronskies GUI now is able to display them 

ronski need to make new version with wider column to support your FEC soon

[Ronski]

Here's mine from our works line. This was on fast path until the firmware update, then a day or two later it all went a bit pear shaped!

PS. I'll tweak the column width a bit 

Edit: Actually whilst posting that my FEC's went up to 1002598178, and I think there is still room to spare, as you can see I get around half million FEC errors per minute.

Seems NewtronStar connection has been up 6 days, but he has four times as many FEC errors than me, but his delta figure is only 171,000 odd where is mine is regularly half a million, he must have a lot of errors sometimes to get that high that quickly. Or did his total not reset at the last resync, I know BE1 said some do and some don't.