Probable Overheating Street Cabinet

[TightNewCustomer]

Originally Posted by Rimmer100

Guess I'll need to get an engineer out. It's been cold for a bit so I'll wait until the weather heats up and it stops working again. That is if the weather ever does heat up again!

Many thanks for your suggestions.

If the weather is too cool, try setting fire to the cabinet and then testing you services, if they don't work, then you know the cabinet was at fault*.

*comment for entertainment only

[Feenix]

Originally Posted by on in an hour!

mmm rubbish eh?? so how come on a daily basis i have sacm's and set tops locking on in the high 50's,removing attenuation (which is just the same as upping the drop in the cab),allows them to lock on lower?

If you've been a Service Tech for years then you should know the importance of correct signal levels at the CPE.
It's not the same as upping the drop in the cab as you are only increasing the forward path levels. If you move the drop up a tap you will get more forward path signal as well as a reduction in return path as the 4dB-ish that is lost when going through a tap block is gone. Removing attenuation/equalisation/simulation without testing the signal levels will just cause issues in the future. Yeah, for a short term fix then great, get the sub back online but it is probably delaying the inevitable.

The only cases I've encountered whereby removing attenuation reduces the return path power level is old skool attenuators (not the forward path type), faulty attenuators or where the SNR is so abysmal that you need to blast 15dB's into the SACM to get it to lock.

[on in an hour!]

as i said all along feenix,it was a suggestion to the OP to get his svcs up and running until he can get a tech out,as i also said the tech may want to get you guys to look at the levels from the node.

[Rimmer100]

So I didn't keep the original appointment to have my cable re-run. The previous engineer must have managed to tweak something as the internet or ondemand hadn't dropped out at all even though there was some fairly hot weather. As I was pretty busy at that time I cancelled the appointment as it seemed a waste of Virgin's time.

Time rolled on and some idiots tried to steal some copper from so electrical cables in Dartford. I was one of the lucky ones that was without power for 4 days. During this time whatever 'fix' that had applied (or maybe it was coincidence) stopped working. When it got hot again recently, bye bye internet and ondemand.

Today they were meant to be redoing the cable like in the call that I cancelled. I had an 8:00 - 13:00 appointment but no-one has shown up.

If they do eventually sort out doing this I will definitely update this thread with new readings from the modem etc.

[Rimmer100]

So I think the cable got re-run today. Interestingly, the first thing I knew that the engineers were on site was that my internet stopped working. When I looked at the modem I saw that there was no sync and had a look out of the window. There they were busily carving up the cement. I was on my mobile before hand, so they could have tried to call and not got through. The didn't come and knock on the door as I had it wide open, just to make it fairly obvious that I was in. I expected that they would come and see me once they had finished to test the levels but by chance I saw the online light on the V+ box. Looked out of the window and the van had gone. Very strange.

Anyway this is a recap of levels for ease of display:

Cool Levels 3rd June 2009:

Cable Modem Upstream
Upstream Lock : Locked
Upstream Channel ID : 2
Upstream Frequency : 29200000 Hz
Upstream Modulation : QPSK
Upstream Symbol Rate : 2560 Ksym/sec
Upstream transmit Power Level : 51.0 dBmV
Upstream Mini-Slot Size : 2

Cable Modem Downstream
Downstream Lock : Locked
Downstream Channel Id : 1
Downstream Frequency : 330750000 Hz
Downstream Modulation : QAM64
Downstream Symbol Rate : 6952 Ksym/sec
Downstream Interleave Depth : taps12Increment17
Downstream Receive Power Level : -7.2 dBmV
Downstream SNR : 27.4 dB

Levels now (19th September 2009)

Cable Modem Upstream
Upstream Lock : Locked
Upstream Channel ID : 1
Upstream Frequency : 25800000 Hz
Upstream Modulation : QPSK
Upstream Symbol Rate : 2560 Ksym/sec
Upstream transmit Power Level : 46.0 dBmV
Upstream Mini-Slot Size : 2

Cable Modem Downstream
Downstream Lock : Locked
Downstream Channel Id : 1
Downstream Frequency : 330750000 Hz
Downstream Modulation : QAM64
Downstream Symbol Rate : 6952 Ksym/sec
Downstream Interleave Depth : taps12Increment17
Downstream Receive Power Level : 0.1 dBmV
Downstream SNR : 33.1 dB

Obviously the time of year is going to make a difference. Massive change on the downstream receive power.

Until it gets hot again next year I'm not really going to know if a difference has been made.

What do you guys think of the new levels?

[on in an hour!]

spot on

[Welshchris]

I heard that VM are gonna use Shakin Stephens song 'Green Door' as a new marketing ploy for their broadband.

[Ignitionnet]

Originally Posted by alex250

Dual Docsis area has a mcns at 330.75 aswell as 586.75, mcns frequencies do vary from area to area tho

331 is primary downstream for ex-Telewest, 339 dual DOCSIS. They use the IRC plan not the HRC ex-ntl use, I guess that's where the chap's confusion kicked in.

586.75 is the magic number for ex-CWC areas, 402.75 for original ntl.

---------- Post added at 09:02 ---------- Previous post was at 08:59 ----------

Originally Posted by alex250

just get network to set the return path levels on a hot and sunny day, then you'll be in spec when its hot and even more so when its cold, you would need a service lad to refer it to them

Return attenuation drops during the summer usually from passives expanding due to the heat causing better contacts, when it goes back up again modems will increase transmit power. Return levels are set to ensure that network SNR is good and modems are transmitting sufficiently over the noise floor and diddling with it purely to drop transmit power runs the risk of modems dropping their transmit power and getting too close to the floor. Especially an issue with modems that are close to the nodes and actives and would be transmitting low anyways.

That said I know the opposite can happen, some of the actives are a tad beyond what should be their operational life span but that's an issue that should be dealt with on the actives rather than messing with the line up of the RP