Squeezing a bit more from my line?

Hi Guys,

Was just wondering if anyone could offer some hints, Im not very versed in 'good' snr and attenuation ratios so if someone could look at my stats it would be much appreciated Smiley

Had fibre installed about 2 weeks ago, new faceplate and a data extension installed by openreach, The data extension is only about 4 metres long to be fair Im with plusnet on a 40/2 package.

On installation day I was syncing at around 25mbps. Now i seem to have dropped to around 21.8mbps, having tried in the master test socket its around 21.9mbps so fairly sure the data extension is having no major effects.

Sometimes I do hear a slight crackle on the line during calls and wondered if this is something that can be fixed at all? To be fair we had a few issues with our line over the years and the bt test always used to say 'High Loop Resistance' whatever that may mean! Even on ADSL when my line was supposed to be 7mbps we struggled to 5mbps.

Here are my Hub One stats if anyone can tell me what part of it may be causing the issue:

6. Data rate: 2000 / 22398
7. Maximum data rate: 3362 / 26504
8. Noise margin: 7.3 / 6.4
9. Line attenuation: 31.0 / 28.7
10. Signal attenuation: 30.9 / 25.4
11. Data sent/received: 204.9 MB / 2.3 GB
12. Broadband username: ***********@plusdsl.net
13. 2.4 GHz Wireless network/SSID: ISS-2.4GHZ
14. 2.4 GHz Wireless connections: Enabled (802.11 b/g/n (up to 300 Mb/s))
15. 2.4 GHz Wireless security: WPA2
16. 2.4 GHz Wireless channel: Automatic (Smart Wireless)
17. 5 GHz Wireless network/SSID: ISS-5GHZ
18. 5 GHz Wireless connections: Enabled (802.11 a/n/ac (up to 1300 Mb/s))
19. 5 GHz Wireless security: WPA2
20. 5 GHz Wireless channel: Automatic (Smart Wireless)
21. Firewall: Default
22. MAC Address: **:**:**:**:**
23. Modulation: G.993.2 Annex B

Many thanks to all and wishing you all a Merry Christmas Smiley

Lee

You are a little short of the Max Attainable - 22400Kbps compared to 26500kbps (rounded). If the data extension is that close and the results or not significantly different then the extension itself is fine. An HR fault would affect both attainable and atual sync.

What electrical or electronic equipment is located close to the phone wire and data extension?

Have you done a full and total clean reboot?

Power off the hub - leave the line connection in place.
Wait 1 minute then disconnect the line and leave disconnected.
Wait 5 to 10 minutes.
Re-power the hub and allow it to boot fully - say 2 minutes
Reconnect the line and allow to sync.

Plusnet Hub 1 is a rebadged BT HH5A, and no OR modem supplied.

Did you ever report the High Loop Resistance? Does the test still give that? Am I right in saying the engineer did not leave a separate modem between the Hub and the faceplate?

Edited by RobertoS (Sat 19-Dec-15 10:28:58)

Will try the full reboot process and report back now Many thanks In regards to electrical equipment there is none whatsoever, if anything the line comes within about 3 feet of the power adapter for the Hub, It is an all in one so as said no OR Modem involved at all.

The High loop resistance thing was reported and a guy came out some years back and call quality improved slightly, we are talking in the days of dial up where i had frequent disconnects and not only when someone picked up the phone lol.

As to whether or not it reports the same i cannot even remember how or if still possible to do the line test on the phone now

Edited by deleted (Sat 19-Dec-15 11:26:52)

Results after full reboot as described:

5. DSL uptime: 0 days, 00:02:02
6. Data rate: 2000 / 21767
7. Maximum data rate: 3327 / 25616
8. Noise margin: 7.1 / 6.3
9. Line attenuation: 31.0 / 28.7
10. Signal attenuation: 30.9 / 25.4
11. Data sent/received: 0.2 MB / 0.3 MB

Soo not made a great deal of difference it would seem

What speed estimate were you given when you ordered from Plusnet?

What speed estimate ranges (both A and B) do you get by putting your phone number in here:
http://www.dslchecker.bt.com/

What speed estimate ranges (both A and B) do you get by putting your postcode into the address-checker here, followed by selecting your actual address from the subsequent dropdown list:
http://www.dslchecker.bt.com/adsl/ADSLChecker.Addres...

It is worth knowing these numbers to judge the rest of the data against. You might also want to check whether your neighbours get similar estimates.

When you perform the last two tests, the screen should identify the exchange name and cabinet number. Can you let us know those?

In reply to a post by Bales1983:

On installation day I was syncing at around 25mbps. Now i seem to have dropped to around 21.8mbps, having tried in the master test socket its around 21.9mbps so fairly sure the data extension is having no major effects.

First, please understand that FTTC suffers more from an effect known as crosstalk than ADSL did; crosstalk is noise induced on your line from other FTTC subscribers. As take-up increases, crosstalk increases, and speeds drop somewhat.

So, annoying as it seems, sometimes we can do nothing about the drop in speeds. This is why BT provide a range - and we should expect variation within that range.

In reply to a post by Bales1983:

6. Data rate: 2000 / 22398
7. Maximum data rate: 3362 / 26504
8. Noise margin: 7.3 / 6.4
9. Line attenuation: 31.0 / 28.7
10. Signal attenuation: 30.9 / 25.4

The statistics here are something of a bare minimum, and don't help us identify whether any problems are going on.

SNR is always set to target 6dB, and should only be higher when your sync speed is already the maximum.

31dB attenuation is quite high. Do you know how far from the cabinet you are? I'd guess at a line length of more than 1km.

Anyway, the fact that your "actual sync speed" differs from your "maximum sync speed" by around 20% suggests that DLM has intervened on your line because of the error rate, and has turned on both interleaving and forward error correction (FEC) as a means to reduce the error rate. Unfortunately, this reduces your sync speed (usually by between 10% and 20%, but sometimes more) and increases latency (by 8ms or more).

Have you noticed any difference to latency?

This kind of DLM intervention used to be common on all types of line, but BT has introduced new error protection mechanisms on *some* of the cabinets (ones made by Huawei) known as G.INP, or re-transmission, which doesn't have the same impact on speed.

That means we can't be sure whether your really have suffered intervention in this way without knowing what type of cabinet you are connected to.

If DLM has intervened, the usual first recommendation is to start monitoring the line statistics - but I'm not sure if you can do that with the Hub One.

Sometimes I do hear a slight crackle on the line during calls and wondered if this is something that can be fixed at all? To be fair we had a few issues with our line over the years and the bt test always used to say 'High Loop Resistance' whatever that may mean! Even on ADSL when my line was supposed to be 7mbps we struggled to 5mbps.

They're all indications that suggest some form of problem on the line, that could, in turn, cause enough errors to have caused a DLM intervention.

Thank you so much for a very informative reply, Im a bit of a nerd when it comes to reading up on things before taking the plunge so im fully expecting crosstalk to kick me at somepoint if not already

Speed estimate from plusnet was around 33mbps if i recall correctly.

Ok the phone number test gives me (Exchange GORLESTON is served by Cabinet 21)

Clean range of 33.2 to 23.9 and an Impacted range of 26.2 to 13.1 ( I might add the OR engineer said there were no bridge taps or anything as i asked this lol)

The Postcode check when i select my address gives me the exact same figures (although the ADSL figures were higher than for my direct phone number search?)

Im unsure about the neighbours as dont particularly mingle with them

About a week ago i did experience random resyncs during the night so i would suspect DLM is kicking in as i seemed to lose a little more sync speed after each reboot, in regards to my line length as far as i know i am about a mile or more from the cabinet unfortunately! My ping to begin with was around 14ms and currently it is at 21ms.

Unfortunately my Hub One is fairly limiting in regards to accepting ICMP and stats it will give lol


Lee

Unfortunately the HH5A/Hub one doesn't support G.INP but even if it did you are connected to an ECI cabinet which doesn't offer G.INP anyway.

I don't use an HH5A, and have never encountered a Hub One, but I recall the world has stepped on...

Wasn't the original problem with the HH5A that it didn't support G.INP in the upstream, and when it was asked to by DLM, it resulted in major speed and latency impacts on the downstream?

And the BT solution was to take away G.INP in the upstream, as a default setting. For every modem.

So, isn't the current status this:
- BT's default DLM setting (on Huawei cabs) is to turn on G.INP downstream only
- The HH5A currently will run G.INP downstream (on a Huawei cabinet) provided DLM sets it that way.
- DLM won't subsequently turn on G.INP upstream for the HH5A, even if the error rate increases.
?