cancel
Showing results for 
Search instead for 
Did you mean: 

SSFP Mk3 faceplate faulty still

Lines566
Hooked
Posts: 9
Thanks: 1
Registered: 19-04-2016

SSFP Mk3 faceplate faulty still

Had 40/2 enabled on the 18th of this month, with a max d/l of 110mb, attn of 8.9 and SNR of 26.1 was well chuffed so much so I opted for the 80/20 on the 20th, plusnet found a voice fault on my line and same day OR turned up to fix it, this is where it gets interesting, he fixed the fault and replaced my SSFP mk1 to a mk3, when he went my stats when awol with my attn at 28.7 and SNR of 6.3 was on the phone to plusnet, they found a fault in one of the joints to the cab so had OR out today and he rejoined my line to the cab and checked the exchange, he was getting attn of 35 on his equipment, remembering the threads about dodgy mk3 plate I told him to try a new one and that sorted it, so once the DLM has retrained I'm changing it back to the mk1, just thought I'd put this out there just incase anyone gets a mk3 and messes there stats up
12 REPLIES
Community Veteran
Posts: 5,171
Thanks: 479
Fixes: 20
Registered: 10-06-2010

Re: SSFP Mk3 faceplate faulty still

There doesn't appear to be any problem with either of the speeds or the SNR.

The attn figure of 8.9 was reported when the speeds were at the limit of 40/2.

The higher attenuation figures were reported at pretty much the max speeds for 80/20.

Perhaps it's just calculating the attenuation differently depending on how much of the line is being used. Especially considering the difference in the upstream from 2 to 20. At 20Mb upstream, it could be using far more upstream frequencies, which means it will be using higher frequencies, and higher frequencies have greater attenuation.

Also, the DLM is always monitoring the line, there isn't really any training period.

In summary, I think there's no problem with the stats or the Mk3 plate, I'd recommend just leaving it alone.

Lines566
Hooked
Posts: 9
Thanks: 1
Registered: 19-04-2016

Re: SSFP Mk3 faceplate faulty still

This is after the new mk3 plate was put in, attn is calculated through power resistance of the line, nothing to do with the speed profiles
Community Veteran
Posts: 19,101
Thanks: 443
Fixes: 21
Registered: 31-08-2007

Re: SSFP Mk3 faceplate faulty still

So what do the 80/20 stats look like with the Mk1 plate refitted, or have you decided to leave the Mk3 plate in place as the last stats look fine?

Lines566
Hooked
Posts: 9
Thanks: 1
Registered: 19-04-2016

Re: SSFP Mk3 faceplate faulty still

Going to leave the MK3 on for now till DRM stops learning the line, my attn is back but the SNR is -15db so I'll switch to the MK1 and see if it brings it back up tho I think it's due to them putting me on a busy bundle to the cab
Community Veteran
Posts: 19,101
Thanks: 443
Fixes: 21
Registered: 31-08-2007

Re: SSFP Mk3 faceplate faulty still

I can't see the -15dB in that last set of stats which look fine.They are in fact a tiny tiny margin better because your Upstream has 0.1dB more SNR margin. Assuming the connection hasn't dropped at any point, you will always see some changes in the SNR as the background noise varies with the general conditions. Roughly speaking you appear to be seeing an average of about 3 ES per hour which is pretty good.

Lines566
Hooked
Posts: 9
Thanks: 1
Registered: 19-04-2016

Re: SSFP Mk3 faceplate faulty still

Check my first post, the pic with the keyboard was what I got before my line was messed with
Community Veteran
Posts: 5,171
Thanks: 479
Fixes: 20
Registered: 10-06-2010

Re: SSFP Mk3 faceplate faulty still

But the 26 dB SNRM was when the speed was limited to 40Mb, now that the speed is 80Mb, of course the SNRM will be lower.

Also, I'd like to re-iterate that the DLM never stops monitoring the line. It doesn't make sense to do something after the DLM has finished, because the DLM never finishes.

Community Veteran
Posts: 19,101
Thanks: 443
Fixes: 21
Registered: 31-08-2007

Re: SSFP Mk3 faceplate faulty still

As ejs says that was on the 40/2 product, the stats will be different because there are many more tones being used on 80/20 than on 40/2. The higher frequency tones used to get the greater bandwidth get attenuated more to try and put it as simply as possible.

And as ejs has said, there is no 10 day training period on Fibre. If in the first 24hrs of a brand new line there are high errors, lots of noise etc., DLM can be especially aggressive on the 2nd day. After that as with any other DSL connection it's an ongoing thing. But on Fibre, be wary, you cannot get SNR Resets as on ADSL, and if DLM bands the speed or increases Interleaving, it's doesn't always remove them in a hurry even if conditions improve.

Lines566
Hooked
Posts: 9
Thanks: 1
Registered: 19-04-2016

Re: SSFP Mk3 faceplate faulty still

Ahh ok, never thought of the tones been used, can't deny the attn tho, I'll try the MK1 tomorrow and see if there's any difference, would expect none to be significant
Community Veteran
Posts: 19,101
Thanks: 443
Fixes: 21
Registered: 31-08-2007

Re: SSFP Mk3 faceplate faulty still

It's because the attenuation is calculated over the range of tones used. I would leave the Mk3 plate in place if I were you, the primary difference between it and the Mk1, is the Mk1 has no common-mode filter.

Community Veteran
Posts: 5,171
Thanks: 479
Fixes: 20
Registered: 10-06-2010

Re: SSFP Mk3 faceplate faulty still

The attn in the earliest stats with the Mk1 is 8.9, and the attn in the most recent stats is 9.0, which is not a significant difference. The faulty Mk3 may have increased the attenuation substantially, but then, it was faulty (and yet you still managed to get nearly the full speed despite it being faulty).

The interference filter in the Mk3, which the Mk1 does not have, may be the reason for the very slight increase in the attenuation. The overall effect of the interference filter aims to improve the broadband, by reducing the noise/interference that would otherwise affect the broadband signal.

Community Veteran
Posts: 19,101
Thanks: 443
Fixes: 21
Registered: 31-08-2007

Re: SSFP Mk3 faceplate faulty still

I think maybe there might have been a bit of confusion as a result of the 28.7dB attenuation with the faulty Mk3 plate.