Broadband Related > Broadband Hardware

Graph Interpretation / FTTC Line Issue

(1/2) > >>

Magsy:
About a month in on 80/20, ECI /r modem and ECI cab, 450-550m distance.

Tried to unlock on day 1 but my usb > ttl cable was bust however no speedtest was less than 60/20 with 15ms ping. Day 4 interleaving came on at a shallow depth, pushed ping to 21ms. Still had 59.9mbps downloads for next few weeks.

Cam home yesterday to find I have 30ms ping and can't break 49.9mbps. Unlocked the modem and found I'm synced at 50/15.

So I don't have original stats and I'm willing to accept a drop off but the start of this graph looks broken...Discovered had no dial tone and BT have just text to say that is resolved, however the connection is still the same. I've resynced it but the hole is still there.

Anyone care to suggest if this is normal / what has happened?

Thanks
 

Black Sheep:
I'm no expert on graphing, but there's plenty of folk on here who are. What I can say, is that most circuits require a manual reset of the FTTC port once an issue has been resolved. Rebooting the Modem/Hub will not be enough. :)

Bald_Eagle1:
The attachment is from a Huawei HG612 connected to an ECI DSLAM, so if by 'hole' you mean the 'V' shape at the left of the graphs, then yes, it's quite normal.

Magsy:
Thanks, I should have looked around first but in my horror I saw the v' notch and just thought surely that is broken?! :D

I can now see that is a normal exclusion zone for VDSL.

BT fixed the no dial tone issue and I resynced a few times, had 6db margins but 50/15 sync same as during the line fault. Had a DLM forced resync at 4am and I'm back to 60/17, the big suprise is not just a return to lower (8ms) interleaving but I'm back on fast path woo! Graph looks the same lol..


asbokid:
Glad you got it sorted, Magsy  :)

The  original author of eDMT,, that is ben1066 of this very forum, kindly released the source code for his tool.

See:  [1]

Do we have the sourcecode for DrSox's patches for the newer /r ("Arcadyan") model of the ECI modem?

Now that the method for obtaining Hlog and QLN are understood [2], eDMT could be updated some more.

cheers, a

[1] http://forum.kitz.co.uk/index.php/topic,10635.msg225981.html#msg225981
[2] http://forum.kitz.co.uk/index.php/topic,11818.msg231617.html#msg231617

Navigation

[0] Message Index

[#] Next page

Go to full version