|
|
|
greaneyr:Perhaps the cordless is causing some interference? The filter you mentioned is a master filter wired in with scotchloks? Try swapping the ADSL jack to an RJ45, which is cheaper than the BT jack anyway and should eliminate the BT jack as a source of potential faults.
Unfortunately, it seems the 'crackle' I was hearing on the voice line is only audible on my cordless phone. Swap it out to a standard phone and it's not there at any stage. I swear I heard occasional clicks when I was testing a few days back, but that was probably just from a slightly loose connection as I was inserting the BT plug into the jack. Meanwhile, my SN margin on my router has crept up to 13dB this afternoon.
I guess there's a remote chance it could be a connection on my side of the demarc somewhere causing the issue. I know it's not the line to the filter, or the filter to the cable, but potentially everything between router and the wall jack could be the cuprit. But I'm not holding my breath.
Time to find a new industry!
richms: Hasn't raAined for a while is the problem usually. Chorus seem to like to delay the tech till its been dry for a bit so any crackle has gone.
Are you saying perhaps there may be a connection somewhere along the line that is corroded and noisy (and affecting DSL) but that it's not currently audible because the connection is dry?
If that's the case, I might be wise to delay the cancellation of my landine until after we have had some rain so I can listen and see if that makes any difference. Of course, I might have misunderstood what you meant.
webwat:
Perhaps the cordless is causing some interference? The filter you mentioned is a master filter wired in with scotchloks? Try swapping the ADSL jack to an RJ45, which is cheaper than the BT jack anyway and should eliminate the BT jack as a source of potential faults.
Yes, it's a master filter wired in with scotchloks. I don't think the cordless would be causing any more interference than any other electronic device in the house. If anything, perhaps the wireless router sitting next to the ADSL router might be, but highly unlikely I would have thought.
I checked the ADSL jack today, and there's definitely no signs of corrosion on either of the joints. What I did notice, however, was that the jack has a 105K resistor inside it. Not sure whether that is even wired in the circuit, or whether it is a good or bad thing. It was never the master jack. That one is in another room, with a test termination unit wired in.
Update - I just tried re-terminating the punchdowns on the ADSL jack, and immediately saw my sync rate climb to 9.9, 9.1 and 10.1Mb/s on the first three tries. Not sure if that's just a coincidence, and the change is due to some other factor (such as temperature or time of day) but that does seem promising. Makes me think it's well worth replacing the jack with an RJ45 and doing it properly.
Rain is a common cause of temporary degradation and intermittent faults. If there's a resistor on the xDSL jack, you should be safe to cut it off.
As the day has gone on, the sync speeds have dropped. I'm not seeing anywhere near 10 Mbps like I was this morning. I've downloaded DMT tool, and have been running it since this morning. It's good since it can force a DSL resync, and monitor the SNR margin over time. If I do several consecutive resyncs, the speed will invariably drop each time. This might be the DSLAM making its own decisions about the reliability of my connection, of course. However, I did note that sometimes I would get a connection at a high speed, only to find it disconnected automatically and resynced at a slower speed. Changing the outlet has made no difference, and while this is going on, the voice quality is absolutely fine.
The SNR margin has been fluctuating a lot between 13.0 and 12.5, while earlier it was pretty steady. I'll check again tomorrow when it's cooler.
Each time it resyncs, the noise margin should be exactly 12.0dB - this is a target that it aims for when syncing. Essentially it goes for the highest sync rate it can get, while still maintaining enough noise margin.
If the noise margin then increases (say to 13dB) it means there is less noise on the line than when it last resynced - if you do a resync now, the connection should come up a bit faster. Conversely, if the margin is less than 12dB and you resync, then the new sync rate will be lower in order to maintain the 12dB margin.
If the line is so bad that the margin drops to 6dB or thereabouts, this would most likely force a resync, bringing the margin back up, and dropping the sync rate.
Fluctuating noise margin is a good indicator of a physical line problem - just be aware that physical issues sometimes are out of Chorus's control, like interference from nearby electrical equipment, electric fence etc.
That makes sense. However, that logic isn't quite being followed on my connection at the moment. I'll connect, and it will be 12dB. Then it will invariably climb higher. I'll resync, but instead of connecting faster, it usually connects slower than the previous attempt, all the while, keeping the sync rate at 12dB.
I seem to have managed to steady the fluctuations, however. I read a few places suggesting the last 2-3 feet of cable run still matter, and a flat cable (used on most phones/dsl routers) won't perform as well as a twisted pair. As an experiment, I tried twisting the cable I have quite considerably, and the SNRM is remaining steady. At the moment, I'm at 9.1Mbps with 12.5dB, but I almost lay money if I were to resync now, it would come back slower.
Did this get resolved?
Actually, yes, within the last week. Through completely different circumstances.
I have switched to Slingshot as my ISP. When Spark RQ'd the DSL service on my line, nothing else took its place. I logged a fault with Slingshot who did the same with Chorus, and the following day I had a working DSL connection, presumably on a different DSLAM at the exchange. The vendor model reported is different from the previous, and the SNR margin it uses is 6dB. I now connect at over 13 Mbps consistently.
You have had the line switched to ULL gear in the exchange, most probably with the same copper pair. Quite likely you will see the same problems re-occur, as they pointed to a physical line issue. The lower 6dB margin that is used on the Slingshot gear means a higher sync rate, at the expense of stability on problematic lines.
RunningMan:You have had the line switched to ULL gear in the exchange, most probably with the same copper pair. Quite likely you will see the same problems re-occur, as they pointed to a physical line issue. The lower 6dB margin that is used on the Slingshot gear means a higher sync rate, at the expense of stability on problematic lines.
|
|
|