Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Semmy]

Having finally gotten around to unlocking and flashing a HG612 I can now see my full stats.
On checking for issues which could be affecting my connection I found the following:-

A gap in bit loading on three frequencies 909, 1089 and 1215kHz

After a bit of a search I have identified these as BBC Radio 5 Live, Talk Sport and Absolute Radio.
The nearest source for these signals is Moorside Edge transmitter (approx 15 miles away) transmitting at 200kW!

Do I have any options to reduce or remove this interference as it leaves three significant gaps in my capacity (and indeed bleeds into the surrounding tones too)

[burakkucat]

Hello Semmy and welcome to the Kitz forum.

Congratulations for your successful research in finding the origin to those gaps in your bit loading. 

Unfortunately (for you) those three broadcasters are licensed to make use of those frequencies whereas we, users of xDSL signals carried on a metallic pathway, have no "rights" / "say" / "privilege" to any of the frequencies that we attempt to use. 

The only way that such RFI can be minimised is to ensure that circuit has good AC balance, there are no HR or semi-conducting joints anywhere along its length (which could allow cross-modulation & perturb the differential mode of the CO to CPE circuit), a Mark 2 SSFP is fitted at the NTE5/A and the length of the circuit from the SSFP to the modem/router is kept as short as possible.

Depending upon whether you would be willing to accept the insertion loss, the fitting of a BT80-RF3, just before the NTE5/A, may also help to minimise those signals' ingress.

[NewtronStar]

I hope you don't mind me jumping into this thread as this seems very similar to my RFI issues over the last 11 months and I did the MKII job but to no avail, how would one go about fitting the  BT80-RF3 (DIY)

edit spelling mistake BC would pull me up on this 

[JGO]

Semmy, as burakkucat says the RF3 has loss, but it is S/( REIN + RFI) which matters, not just S.

 IF you are in a suburban area with a lowsish background of REIN it still can pull extra tones out of the background, as well as seeing off RFI and improve overall. As a BC station is 10 kHz wide,  a strong one can see off 3 4kHz tones.

In my situation with  weak RFI (underground phone lines) the result is marginally worse.

[Semmy]

To be fair I am on a moderately long line at ~700m from the cabinet  however the cable remains underground for the majority of the journey only appearing above ground at the pole opposite my house, crossing the road then descending into the house on the ground floor.

Am I correct in thinking that this would be the most likely point on the line for the RFI to enter as the underground cable would be well shielded?
Is it possible to "obtain" shielded overhead cable as I suspect it would be better to prevent the RFI from entering rather than filter it out after the fact?

I have taken most other possible precautions:-
50cm twisted pair cable from the SSFP to the modem
1m shielded cat6 cables to the router
Ferrites on the switch mode power supplies DC leads (though I am pondering using linear PSU's - a cheap option CPC - Farnell do them at £6ish)

[JGO]

Am I correct in thinking that this would be the most likely point on the line for the RFI to enter as the underground cable would be well shielded?

Yes !   (unless you live on dry sand !)

Is it possible to "obtain" shielded overhead cable as I suspect it would be better to prevent the RFI from entering rather than filter it out after the fact?

Yes better But AFAIK this is a matter for BT and so may not be practice, burakkucat may know better.

[Semmy]

Just had a quick squint at the RF3 and it appears to have a common mode choke as the only "component", so I guess the only hassle would be the added impedance of the choke on the line which would attenuate the signal a little.

[sheddyian]

Ferrites on the switch mode power supplies DC leads (though I am pondering using linear PSU's - a cheap option CPC - Farnell do them at £6ish)

Hi,

Having also been here myself (though with intermittent interference, rather than a broadcast station), I tried running the modem/router from a 12 volt battery to entirely eliminate mains-borne interference.

For me, it made no difference.  But before going to the trouble/expense of building or buying a fancy PSU, if you've got a 12 volt battery from a rechargeable drill, or a spare 12 volt car battery, consider charging that up fully then running your equipment from it. 

As I said, for me it made no perceivable difference, but might be worth a try if you've already got something suitable.

Ian

[4candles]

The only way that such RFI can be minimised is to ensure that circuit has good AC balance, there are no HR or semi-conducting joints anywhere along its length (which could allow cross-modulation & perturb the differential mode of the CO to CPE circuit), a Mark 2 SSFP is fitted at the NTE5/A and the length of the circuit from the SSFP to the modem/router is kept as short as possible.

Depending upon whether you would be willing to accept the insertion loss, the fitting of a BT80-RF3, just before the NTE5/A, may also help to minimise those signals' ingress.

Surely the main point of the Mk 2 SSFP is that it obviates the use of an RF3 as it contains the same inductors ?

[burakkucat]

Surely the main point of the Mk 2 SSFP is that it obviates the use of an RF3 as it contains the same inductors ?

Ah but not quite.

As Black Sheep has mentioned, previously, the wizards of Grimbledon Down¹ have determined that the Mk 2 is still somewhat deficient in its action (I fully expect a Mk 3 to eventually appear) and the current instructions, issued to all Openreach engineering staff, specifies that a SSFP plus a BT80-RF3 should be fitted when attempting to minimise the RFI on an xDSL circuit.

[1] A.k.a. Martlesham Heath

[burakkucat]

Just had a quick squint at the RF3 and it appears to have a common mode choke as the only "component", so I guess the only hassle would be the added impedance of the choke on the line which would attenuate the signal a little.

Yes, that is correct. JGO has provided a useful description of how it operates in an earlier forum thread.

[burakkucat]

I hope you don't mind me jumping into this thread as this seems very similar to my RFI issues over the last 11 months and I did the MKII job but to no avail, how would one go about fitting the  BT80-RF3 (DIY)

The BT80-RF3 (block terminal 80, RF3) is identical in size and shape to the standard BT80 (block terminal 80). So if there is a pre-existing BT80 in the circuit, joining the incoming service feed cable to the internal CW1308 cable (for example), then a direct one for one swap would be the simplest way. Or you could fit a BT80-RF3 adjacent to the NTE5/A (Senior Management permitting    ), disconnect the cable from the NTE5/A and connect it to the screw terminals of the BT80-RF3. Then use a short length of CW1308 cable to connect from the NTE5/A to the IDCs of the BT80-RF3.

[Black Sheep]

Just for info on the aesthetics, seeing as all techy advice is more than covered as usual 
The 'Gubbings' within the RF3 can be removed and will actually fit inside the NTE5 (Master Socket). Ergo, out of sound and out of mind should 'Da boss' object to unsightly additions ?? 

[burakkucat]

Just for info on the aesthetics, seeing as all techy advice is more than covered as usual 
The 'Gubbings' within the RF3 can be removed and will actually fit inside the NTE5 (Master Socket). Ergo, out of sound and out of mind should 'Da boss' object to unsightly additions ?? 

What excellent advice!    That will certainly minimise the chance of the War Department taking offensive action. 

[Black Sheep]

Thank you, kind sir.