Print

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

[edward]

This is what I'm thinking. I've already made a couple of inquiries along the lines you've suggested. Even if the effect on the line is minimal, from a cosmetic point of view I'd like it done now.

Thanks for all the replies and advice, not much more to be said, I'll let you know in due course what happened.

[waltergmw]

@ Edward,

We really do have a quandary here as I have already indirectly testified. When my wheelbarrow happens to stop adjacent to an enquirer's house I often find that the EU is quite bemused. Sometimes it is suggested that I call after reviewing the achieved throughput speed from the BT Speed tester or another one. Having walked the patch quite extensively now, I am aware of the expected speeds whereas the EU usually doesn't have a clue nor does (s)he have a veritable encyclopaedia of performance statistics.

A prime example, which caused much mutterings of veiled aspersions towards the hapless subcontract installation engineers, happened when an EU brightly said that his speed was so much faster than it was before (it was 1.98 Mbps) at 14.52 Mbps even though it started at 19.12 Mbps but was estimated at 22.6 Mbps. Once I had entry it was quite obvious that all was not well. I adjusted the internal (customer-side only wiring) and was able to observe a significant improvement in the Max rate but the achieved rate remained stuck at 14.52. I heaved a great sigh of relief when told that the ISP was Zen internet. There was no question that an Openreach engineer visit would be arranged ASAP. Upon the arrival of the O/R engineer, his JDSU produced very similar results to those I had observed; this perplexed him as he had not seen this anomaly before. A call was made and by magic the DLM was reset resulting in a speed of 26.78 Mbps - well above the VDSL estimator figure.

Sorry for the diversion but back to your problem. If you get a retired engineer who does a fantastic job, you will almost certainly have a very frustrating experience with India unless and until you can convince the operative to refer the matter to a level two engineer. Thankfully I've only had one instance of a BT supplied service. On that occasion I had just completed exactly the same exercise next door resulting in a new Maximum throughput of around 42 Mbps, yet this one was was stuck at a speed of 19.99 Mbps. The Level two engineer would not accept that a cap had been applied as BT never cap their services, but he did admit that there seemed to be a restriction. Eureka after much meandering !!! The O/R engineer arrived and again "they all lived happily ever-after !

In your case you can state that you have taken advice from many contributors on the fantastic Kitz web site (including the venerable one !) who all state that you have a sub-standard installation requiring remedial action by an expert O/R engineer. Once you have "caught your fish" it should then be relatively easy to remind him to RESET the DLM after he has corrected the wiring anomaly.

RE your own internal wiring, which you are at liberty to adjust as you wish, I suggest it would be prudent to separate any voice component from the incoming VDSL signal. However as I always want to achieve the absolute maximum I would still recommend installing the modem directly into the real master socket's SSFP and running an external ethernet cable to your required computing area with the router.

Kind regards,
Walter

[Black Sheep]

Wise words indeed, from Walter.

[edward]

It was a quandary, the only one to decide how to resolve it being me, and one I can give an update to now.

I wasn't going to chance an OR engineer visit again, so I found an independent telecoms engineer. I was very lucky and found the type of 'old school time served clever and skilled sort who was clued up and knew what he doing' engineer. Emailed him a link to this thread, then rang and talked on the phone, agreed between us what to do, and made an appointment, which was for this morning.

He was just the kind of engineer you want, showed me pics on his phone of some of the wiring horrors he's been called out to resolve, so by comparison my job was simple. The redundant NTE5 and filters now gone, nicely crimped, and blanked off.

After reconnecting the max attainable rate went up by approx. 1.5Mb/s (35248-36212) and actual connected line rate from 29809 to 31381, the acid test (download speedtest), went up from 28.32 to 29.95 Mb/s. This latest result was the same figure we got when the OR engineer did tests during the installation, and it had just dropped a little since then, and today its gone back up and the only time since then that I've seen at that rate again. Regardless of that it's neater and now wired as it should be.

I think they are very good results anyway for 800-900 meters from the cab, and having had Infinity installed for going on for 3 weeks now I'm hoping that's a 'stable' result.

I've attached two sets of graphs, one taken yesterday, and one today after the rewiring. The one from yesterday seems to be missing a group of tones, I'm wondering if that would be due dropping them over the time the modem was up, around 15 days or more, when the line quality at some point wasn't so good, maybe night time interference?

Thanks for all the tips and advice, glad I got the work done.

[burakkucat]

Yay. Result! 

I think it would be best to wait for Feathers to swoop down and comment upon your rather peculiar bit loading graph, recorded yesterday. I certainly have never seen such an absence before . . . 

[Bald_Eagle1]

I've attached two sets of graphs, one taken yesterday, and one today after the rewiring. The one from yesterday seems to be missing a group of tones, I'm wondering if that would be due dropping them over the time the modem was up, around 15 days or more, when the line quality at some point wasn't so good, maybe night time interference?

I have seen that happen once or twice, using both the scripted and program versions.

Re-running the script/program a few seconds later usually provides more valid data.
I can only assume it is due to trying to harvest the data just at the same time as the modem is updating its own logs.

When watching the GUI, it seems to update every 10 seconds.
Maybe you just caught it at the 'wrong' time?

The connection now also looks quiter & has a better SNRN since the wiring work.

[edward]

I see, more of little glitch in the program rather than actual line state. Thanks, if I get a reading like that again I'll redo it.

Again thanks for the info and guidance, all I got on the 'official BT forum' was..."if it aint broke, don't fix it" ha ha!