Kitz Forum
Broadband Related => Telephony Wiring + Equipment => Topic started by: cypher007 on September 01, 2017, 11:28:37 PM
-
I have a normal master socket which has a filter plugged in the front feeding a cordless phone.
inside the faceplate I have connected 2 runs of cat5e from terminals 5 - 2.
one goes to a filtered connection to my alarm.
the other goes upstairs to the router.
now I was getting some fluctuation on TalkTalk FTTC 80/20 service. it would always be around 71-76000 kbps sync, and the TPlink vr900 reckoned it could manage 85000 down 22000 up.
all was well for some months until we went on holiday. when I came back I happened to check, as I do every so often, the sync speed. it had dropped to 66999 down 19999 up. and no amount of rebooting would budge it. thought the router still says it can run 85000.
so I thought id check the wiring and this is where things get weird.
I removed an old extension cable, twisted pair, from the alarm circuit and replaced it with the cat5e mentioned above. as I found removing this from the master socket 2-5 made the router think it could do 33000 up and 89000 down.
so after wiring it in the new cat5e I thought yay. then checked the router and [censored]. it was actually worse. the up had improved the router thought it could do 27000. but the down had dropped to 76000 and the snr had dropped for down.
the sync is still stuck at 19999 up and 66999 down.
is the cat5e causing problems as I have 2 extensions going to 2 separate sockets, one which is the router the other the alarm?
would fitting a filtered faceplate and connecting one cat5e to the unfiltered and the other to the filtered fix things?
if so whats the best faceplate?
has my upstream dslam got stuck on 66999 or is TalkTalk doing some sneaky bandwidth limiting?
-
If you have the classic NTE5/A, not the newer NTE5/C, I would recommend that you fit a Mk 3 SSFP. Then --
- From the IDCs on the back of the lower front face-plate, connect the CAT5e cable that runs to the alarm system.
- From the IDCs inside top left front of the SSFP, connect the CAT5e cable that runs to your modem/router.
-
I was looking at the MK4 and back NTE5C.
it seems cheap enough on ebay, but are they the real deal?
on run-it they seem expensive at nearly £30.
whats the best option? the mk3 still?
-
If you have an existing NTE5/A then I would advise that you fit a Mk 3 SSFP. Here's one example (http://www.ebay.co.uk/itm/GENUINE-BT-Openreach-MK3-FILTER-NTE5a-MASTER-FACEPLATE-SOCKET-VDSL-ADSL-BRANDNEW-/272682664213) and here's a second example (http://www.ebay.co.uk/itm/BT-Open-Reach-MK3-VDSL-ADSL-BROADBAND-SOCKET-with-screws-/152629618227).
Knowing how the NTE5/C, etc, are constructed I wouldn't use one of them. :no:
-
I'd agree with burakkucat on the way to go with this.
With your current setup the connection to the alarm is acting as a bridge tap and you don't want that.
Using an MK3 SSFP as advised will make that bridge invisible to the VDSL side as it will be filtered-off at the faceplate.
Stick "bt mk3 openreach" into an ebay search and you can get a Mk3 filter for £8.85 delivered.
One seller even has a full NTE5/A with filter for £9.50 right now - this one (http://www.ebay.co.uk/itm/Block-for-BT-Openreach-MK3-mk3-nte5a-backbox-screws-See-Instructions-/152444114465?epid=806956957&hash=item237e609621:g:Yj0AAOSwJhNXUAb5)
:)
-
Cheers guys will get a mk3. One thought though some people said it increased there attenuation. Mine without the alarm is about 10db.
-
Realistically any device inserted into the circuit will increase the attenuation. ;)
You will often see the phrase "insertion loss" used and that can be regarded as the absolute value of the difference between the attenuation with the device in the circuit and the attenuation without the device in the circuit.
insertion loss = | attnwith - attnwithout | dB
-
fitted it and all seems good now. well apart from the fact the dslam is still stuck.
up down
Current Rate(kbps) 19999 66997
Max Rate(kbps) 33093 90081
SNR Margin(dB) 14.8 11.8
Line Attenuation(dB) 17.3 10
Errors(pkts) 3 0
still getting some errors on upstream, but always had that.
-
At a glance, it does seem as if the DLM has banded your circuit. However if there are very few errors recorded per day, the DLM may well relent. So let it be and just use the circuit.
-
after a few hours of light use, mainly youtube, the errors upstream are 16 pckts.
any ideas whats causing these, as they are always on the upstream.
the other problem I have is because I use the router for file sharing I have to reboot it maybe once or twice a week. as the media server wont update its DB with the new files otherwise. I asked TPLink and apparently this is normal. I'm wondering if this is causing the Dslam to get stuck.
-
after a few hours of light use, mainly youtube, the errors upstream are 16 pckts.
any ideas whats causing these, as they are always on the upstream.
No, sorry. :no:
the other problem I have is because I use the router for file sharing I have to reboot it maybe once or twice a week. as the media server wont update its DB with the new files otherwise. I asked TPLink and apparently this is normal.
<Cough><Splutter> That is definitely not normal for a well behaved device! Perhaps something better than the TP-Link device would be appropriate. ::)
I'm wondering if this is causing the Dslam to get stuck.
It's not the DSLAM (digital subscriber line access multiplexer) "getting stuck" but the DLM (digital line management) process that has probably banded the circuit.