...We've seen the 0dB upstream line attenuation issue before and it is a Wi-Fi Hub reporting problem. Quite possible that the Wi-Fi Hub isn't performing as expected on your line.
But if you test with the Wi-Fi Hub plugged into a microfilter and that plugged into the Test Socket then the first act of carefully withdrawing the faceplate of the Master Socket will disconnect all extensions. With the Hub connected to the line you'll have a comparison test with and without home telephone extensions, 'phones and accessories.
Hi again @Gondola...
Okay, disconnected everything and left the router turned off for 30 minutes so it didn't look like a disconnection. Removed all telephones and anything that might interfere, tried a new filter and plugged into the test socket.
Here are the stats:
The speed has gone up but the actual rate is still higher than the maximum rate and the attenuation is still the same.
...I'm not happy with the attenuation figure as it definitely doesn't match the estimated line length to the street cabinet. And there's obviously something not right with the reported max rate versus actual access line speed.
I'd feel happier if we got you an alternate router to test with.
@Gondola I did try using my own Netgear router a month or so ago and it only had a DHCP lease time of 15 minutes. I was told that this was probably because it was not a TalkTalk approved router and I should stick with one provided by TalkTalk.
How would you suggest moving forwards?
...If the Netgear router wasn't configured to request the lease renewal then this would cause a connection drop.
What normally happens is that the Openreach DHCP server is requested to renew the lease at the half-way point in the lease being granted. So, the Openreach lease is 15 minutes but is automatically renewed at seven and a half minutes and the IP address is maintained. There isn't a problem with this method of operation and it's very efficient.
If you don't have an alternative to test with I'll ask the TalkTalk team to consider letting you have an alternative router for tests on a temporary basis if that's ok with you.
...The 'new' line attenuation figure indicates an access line speed of around 69Mbps. But it also indicates a metallic line length of around 500m which still seems more than anticipated.
The S/N ratio is often set high when the attainable line rate or maximum speed is a lot higher than the DSLAM's capped product speed. For example, if the line could theoretically support a speed of 90Mbps but the DSLAM limits this to 80Mbps.
If you can, leave the setup as is for a few days and see if the errors diminish, the S/N ratio goes down and the speed goes up.
The error count hasn't increased at all which means the line is stable and good. The attenuation figure for the line is consistent with my calculated 69Mbps speed and that's pretty much what's being delivered.
If I hadn't expected the line length to be shorter / the attenuation figure to be lower I'd say the broadband service was performing as expected.
It's clear that the Wi-Fi Hub is showing a different set of line statistics that doesn't tie in at all with the modem stats.
Disregarding that, if the 68-69Mbps speed is within the expected range for the line then it'll be difficult to get Openreach to investigate further. So the key questions are:
Hi again @Gondola,
Thanks again for all of your insight. I have switched back to using the TalkTalk WiFi Hub as the modem and connected the Orbi back to it in DMZ mode. This seems to be a more stable way of accessing the internet. With the BT Openreach Modem during some DHCP lease sessions I could not access the internet and had to manually refresh the DHCP lease in order to access the internet.
Anyway, here are the stats after I connected it back to the TalkTalk WiFi Hub:
It appears that the download speed has increased but again, and as with the Openreach modem, the maximum rate is lower than the actual rate. The attenuation has gone back to the higher value. I've also noticed quite a bit of fluctuation with the upload speeds during this process. While the download speeds seem to slowly be increasing the upload seems to yo-yo a bit.
I checked my order confirmation and it shows the following information:
Minimum guaranteed speed: 40.5Mb
Typical download speed: 47.8Mb to 71.6Mb (my current connection speed is on the higher end of that spectrum)
Typical upload speed: 12.2Mb to 18.9Mb (my current connection speed is on the lower end of that spectrum
I was definitely getting higher speeds than this with Sky, so I thought I would run my details through their website and this is what I goat:
Minimum guaranteed speed: 54Mb
Typical download speed: 60Mb to 73Mb
Typical upload speed: 18Mb to 19Mb
The TalkTalk stats seem to fit with the service I am getting but the Sky stats seem to fit with the service I was getting from them. I assumed that all Fibre connections were provided via BT Openreach so not sure how they can differ?
On the 3.1dB noise margin figure. The last few days without any errors is a good indicator that the metallic line is intrinsically interference free and as such would qualify for the Openreach Strategic Dynamic Line Management to use the xdB process allowing good (unimpaired) lines to work with a signal to noise margin as low as 3dB.
On the estimates of line speed. These are based on Openreach survey data. My belief is that both Sky and TalkTalk order the standard profile versus the speed or stable profiles. Your line would certainly take the speed profile and maybe these are the figures that Sky are quoting if the speed profile has been used for the line before. I cannot think of any other reason for the disparity in the quoted figures.
Openreach do apply charges for profile changes so ISP's don't usually offer profile changes as a service.
As for the line attenuation disparity. There's some misreporting somehow. It certainly doesn't help me in the analysis of line performance.
I think the OCE's could step in at this point and see if the reported line stats are replicated in tests from their end.
I think the speeds quoted by SKY and the speeds quoted by ourselves may be different because they are potentially referring to sync speed while we are referring to throughput speeds. Do you know if the speeds quoted by SKY are the estimated sync speeds or throughput speeds?
I assumed it meant throughput. When I was with Sky the sync speed was about 78Mbps.
The sync speed has now increased to 70 Mbps, however the “maximum rate” is showing as 68Mbps. Is there a reason why the actual rate always appears higher than the actual rate?