DLM period now over....any tips to boost sync + speed... (ADSL2+)

[#1]

Sorry we've not been in to this thread this week yet!

Quote

Ah, ok, I was unaware of the 3db profile conditions. Looks like I'm gonna have to try locate this cause of noise on the line. Regarding the target SNRM, I thought I was on the 6db profile, but I'll ask the DCT tomorrow.

I'll check, however if there is interference of any kind (sorry if that's been eliminated already, there's a lot of info to get through in here!) then I'd suggest going down the fault process if you're confident it's not internal to your premises.

[#2]

Just to confirm, your target SNR on the downstream is 6db.

[#3]

Hi Chris,
Thanks for getting back to me.  So in a nutshell, I've been reading other threads where PlusNet customers with 29db of downstream att. are syncing at around 16000 Mbps and d/loading at around 88% of that. I also have a 29db attn, but cannot understand why I only sync @ 13500 Mbps.
So far
- quiet line test is, quiet!
- changed micro filter
- changed adsl cable to a filtered one (resulted in an increase of approx. 1 Mbps)
- tried another router (BT2700 HGV, original not hacked with any firmware)
- Only have a BT master socket  (BT engineer came over to reactivate the line a month or so ago and fitted a BT openreach printed faceplate. Its not an I-plate, I think its a basic NTE5)
- No extensions at all so no need to try the master socket.(right?)
- Wired BT handset which I have even left unplugged for a day on the advice of others on the board (in case it was picking up any interference)
If you look at the last few posts I have attached charts (router stats) of a strange occurrence where my BB seemed to stop working for 12-15 minutes, it didn't drop the connection, didn't resync, its odd.
Also on the error side of things

CRC Errors (Up/Down): 24 / 6,192

HEC Errors (Up/Down): 14 / 20,567

These errors normal/abnormal? They don't seem to impact our usage in any way. Otherwise, I've been connected for 5+ days  Any suggestions? I
@pmarti22
I'm hoping there is a solution for this missing bandwidth. As you can see I'm trying everything! (even changing to shielded cabling) keep monitoring this thread, hopefully we'll get troubleshooting tips.
cheers
PJ

[#4]

Quote from: pj2014

...strange occurrence where my BB seemed to stop working for 12-15 minutes, it didn't drop the connection, didn't resync, its odd. Also on the error side of things CRC Errors (Up/Down): 24 / 6,192 HEC Errors (Up/Down): 14 / 20,567 These errors normal/abnormal?

PJ,
The gap might not be anything overly strange.  RouterStats can do that if the router is unresponsive.  I have seen this occur when the router (possibly a dying one) was overly busy handing FEC errors.
Cannot advise if the error count is hight without knowledge of the uptime over which these were 'acquired'.

[#5]

Hi Townman,
Doh, forgot to attach the stats. Here you go! No FEC errors so router is probably ok (its only 3 or so weeks old anyway, standard plusnet one)
cheers
PJ
Uptime: 5 days, 2:59:08

DSL Type: ITU-T G.992.5

Bandwidth (Up/Down) [kbps/kbps]: 1,155 / 13,547

Data Transferred (Sent/Received) [MB/GB]: 830.86 / 25.84

Output Power (Up/Down) [dBm]: 12.4 / 0.0

Line Attenuation (Up/Down) [dB]: 15.4 / 29.0

SN Margin (Up/Down) [dB]: 6.6 / 6.5

System Vendor ID (Local/Remote): TMMB / ----
Chipset Vendor ID (Local/Remote): BDCM / IFTN
Loss of Framing (Local/Remote): 0 / 0
Loss of Signal (Local/Remote): 0 / 0
Loss of Power (Local/Remote): 0 / 0
Loss of Link (Remote): -
Error Seconds (Local/Remote): 4,447 / 0

FEC Errors (Up/Down): 0 / 0
CRC Errors (Up/Down): 25 / 6,226
HEC Errors (Up/Down): 15 / 20,725

[#6]

Quote from: pj2014

Hi Chris, Thanks for getting back to me.  So in a nutshell, I've been reading other threads where PlusNet customers with 29db of downstream att. are syncing at around 16000 Mbps and d/loading at around 88% of that. I also have a 29db attn, but cannot understand why I only sync @ 13500 Mbps. So far - quiet line test is, quiet! - changed micro filter - changed adsl cable to a filtered one (resulted in an increase of approx. 1 Mbps) - tried another router (BT2700 HGV, original not hacked with any firmware) - Only have a BT master socket  (BT engineer came over to reactivate the line a month or so ago and fitted a BT openreach printed faceplate. Its not an I-plate, I think its a basic NTE5) - No extensions at all so no need to try the master socket.(right?) - Wired BT handset which I have even left unplugged for a day on the advice of others on the board (in case it was picking up any interference) If you look at the last few posts I have attached charts (router stats) of a strange occurrence where my BB seemed to stop working for 12-15 minutes, it didn't drop the connection, didn't resync, its odd. Also on the error side of things CRC Errors (Up/Down): 24 / 6,192 HEC Errors (Up/Down): 14 / 20,567 These errors normal/abnormal? They don't seem to impact our usage in any way. Otherwise, I've been connected for 5+ days  Any suggestions? I @pmarti22 I'm hoping there is a solution for this missing bandwidth. As you can see I'm trying everything! (even changing to shielded cabling) keep monitoring this thread, hopefully we'll get troubleshooting tips. cheers PJ

Thanks @pj2014.  I too have ordered the Cat5e modem lead to see if it makes any difference.  Will keep you posted and reading your log for any extra info too.

[#7]

Quote from: pj2014

Uptime: 5 days, 2:59:08

Good, very stable. Also show the connection was established during the daytime, which is good for speed.  Unless the uptime is the PPP session uptime in which case this won't necessarily be true. 

Quote from: pj2014

Bandwidth (Up/Down) [kbps/kbps]: 1,155 / 13,547 Line Attenuation (Up/Down) [dB]: 15.4 / 29.0 SN Margin (Up/Down) [dB]: 6.6 / 6.5

So the fundamental problem is that the downstream bandwidth isn't very good in the context of the downstream attenuation.  Not terrible but not very good.  The fact that it went up a lot in response to use of a shielded cable between the socket and the router might imply that there is something bad nearby that needs the shielding.  The downstream SNRM of 6.5 dB might also imply thing have got slightly better since the connection was established and you might go slightly faster if you forced a reconnection (during daylight hours).

Quote from: pj2014

FEC Errors (Up/Down): 0 / 0 CRC Errors (Up/Down): 25 / 6,226 HEC Errors (Up/Down): 15 / 20,725

There's probably some useful information here but I won't comment because I don't have a Plusnet supplied router myself.  The important thing to do is to keep monitoring with Routerstats and look for any sudden changes in noise margin (or error rate).

[#8]

@ReedRichards
Hahaha, just as we spoke about the stable connection  it looks like it dropped it @ 17:33 and reconnected 100kb/s quicker.
Dec 12 17:34:35 PPP link up (Internet) [146.90.63.106] 
Dec 12 17:34:35 PPP CHAP Receive success (Internet)
Dec 12 17:34:34 PPP CHAP Receive challenge from rhost PTW-AG01 (Internet)
Dec 12 17:34:34 PPP CHAP Receive challenge from rhost bras-red14.l-nws (Internet)
Dec 12 17:34:31 xDSL linestate up (ITU-T G.992.5; downstream: 13679 kbit/s, upstream: 1155 kbit/s; output Power Down: 20.9 dBm, Up: 12.6 dBm; line Attenuation Down: 29.0 dB, Up: 15.4 dB; snr Margin Down: 6.1 dB, Up: 6.6 dB)
Dec 12 17:34:05 xDSL linestate down
Dec 12 17:34:02 PPP link down (Internet) [46.208.153.30] 
Dec 12 17:33:34 xDSL linestate down
Dec 12 17:33:17 xDSL linestate down
I'll look around the immediate area where the router is situated for any source(s) of interference. Yup, that's really the long and short of it, not a big issue but I'm trying to maximise the d/s bandwidth 
Reading between the lines, was there an implied reference to the 'suitability' of the PlusNet router    Hehehe, I'll happily try another should PlusNet wish to send me one to test. I might swap the current one for a BT2700HGV I have when I reboot the router tomorrow during the day. I was kind of hoping a DCT member might comment on the number of errors + other info I have provided. Well, maybe tomorrow 
Yup, I'm gonna try set up Routerstats, I was having some difficulty with the setup with respect to the login address, http://192.168.1.254 didn't seem to pick it up (hence why I was using routerstats lite)
Ah, I just realised, I have an old mk1 white sky router (netgear dg...gt) flashed with DGT(sp?) firmware, its highly tweakable so I've been told. No idea how to use it without some solid instructions though! LOL
cheers
PJ

[#9]

The Plusnet-supplied router is okay but I had my own router when I joined Plusnet and stuck with that one.  But it's always a good idea to see if another router makes any difference.
I suggest that when it is full daylight you ask your router to disconnect, power it down for a few minutes then turn it back on (or switch to a different router)  and see what speed you get.  I would expect it to be better than last night's and this should reset your error counts to zero so you can begin to estimate error rates rather than just have error counts over a very long time period.

[#10]

Ok, router disconnected via the admin login and powered down for a few minutes...
I stuck with the same PlusNet router for now, just to eliminate it.
Speed is exactly the same, for now. All the error counts have reset, so will monitor them over the weekend.
Got routerstats (full version) configured and logging away, so will see what appears over the next 24/48/72 hrs
cheers guys!
PJ

[#11]

hey guys,
ok, no dropped connections, no loss of sync but these graphs occurred in routerstats.
Does anyone have any ideas what it may be.
Pretty sure it was nothing from within the house as everything (boiler etc) if off by then. Phone line is still crystal clear (to my ears anyway!)
cheers
PJ

[#12]

Hi PJ,
Yes, there are problems here with RouterStats.  It is not good when a measuring tool can exhibit the problem you are trying to detect in the object being measured.
Remember that RS is requesting data from the router and recording it in the form of the graph.  Unfortunately there is no clear distinction between the data not being available (no response from the router or other issue^*) and the router's data is ZERO because the link is down.
^* RS has a flaw in it - internal WIndows messages can get lost / out of synch.  As best as I can tell from correspondence with the author, RS expects the messages in a 'known' sequence and does not check the content of the messages before processing.  Therefore is a message is lost, the processing is not looking at the right message and the data is not found - that is plots are zero.  This can be verified by looking at the summary tab.  If it shows something other than the ADSL stats (for example tone bins, router interfaces etc., etc) then RS is in a "message sequence is screwed state".  This can be corrected by clicking stop recording and start recording.
All of that said I think that in the case of the first gap on the last graph, that was a resynch.  The SNRM is a little higher than before and (I cannot be sure on the resolution of the uploaded graph) I think the synch rate is slightly different.
HTH,
Kevin

[#13]

Hi Kevin,
Thanks for that explanation. This has happened a few times now and if it occurs again as it did last night @ 23:30 then it wont capture any logs 
I think for the time being I'll switch back to RS-lite, as that seemed to operate correctly.
Oh yeah, that last graph definitely not a re-sync, it was me pressing stop and start in RS when I realised it had stopped recording data. I checked via the router login page -> Gateway -> Event logs -> Connections and there were no dropped connections or re-sync events (I think I have experienced those last week, logs should be here somewhere, PPP link down then up etc etc..)
So sync actually stayed the same, even though the graph scale is slightly different, SNRM d/s is always between approx. 6 and 6.4 through day/night period.
Key information since yesterday afternoon's reboot
Uptime: 1 day, 4:20:36
Bandwidth (Up/Down) 1,171 / 13,667
SN Margin (Up/Down) [dB]: 6.1 / 6.3
Loss of Framing (Local/Remote): 0 / 0
Loss of Signal (Local/Remote): 0 / 0
Loss of Power (Local/Remote): 0 / 0
Loss of Link (Remote): -
Error Seconds (Local/Remote): 5,339 / 0
FEC Errors (Up/Down): 0 / 0
CRC Errors (Up/Down): 4 / 7,279
HEC Errors (Up/Down): 2 / 21,514
So many CRC and HEC errors  I have noticed that general browsing/tasks online sometimes do take longer (lag? is that the right word?)
cheers
PJ

[#14]

Although Routerstats has thrown-up some anomalies where readings stop for a while, you have not reported any changes in noise margin or perceived any changes in error rate.  So whatever is causing your problem seems to be on all the time.  If you are sure it is not the computer that you are using to take the readings then I think it is time to try a different router. 

[#15]

ReedRichards,
Yup, unfortunately error rates have remained high but SNRM movements on the downstream have been minimal. I'm sure as I can be thats its not the computer, I have used my desktop and notebook (dell mini 10) and both had similar figures on errors and steady SNRM.
I will do a reboot tomorrow, daylight hours + reconnect the BT2700HGV (Broadcom chipset I believe. Same as PlusNet supplied router). and monitor the connection for a few days. I am in the process of trying to acquire a Netgear N150 DGN1000SP off auction sites (on the super cheap!) as I have read they contain an Infineon chipset (as opposed to my current routers which all have Broadcom). My exchange appears to run Infineon tech, so reading some websites suggests a positive outcome in running an Infineon chipset router with a similarly equipped exchange.
Hopefully, I'll be able to source one for a fiver, so if this little experiment fails at least I'll have a backup router (gathering dust on the top shelf!)
cheers!
PJ
ps here are the accumulated stats from the router