Print

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

[Weaver]

I have ADSL1 20CN on an ultra-long line.

I used to have an ADSL Nation faceplate micro-filter, then changed to the BT SSFP / Pressac (?) faceplate.

More recently still, as I have no phones at all (and have never had any extensions), I’m now using Andrews & Arnold 'blank' faceplates with no filtering in them, just an RJ-45 socket which is straight through to the test socket internally.

Someone told me that if there are no phones, then there's nothing to filter, so I might as well go the straight-thru blank plate route.

Is that actually true? Do the Pressac / Adslnation / BT Mk2 / Mk3 have any merits for someone in my situation?

I read about the filters in BT Mk2 and Mk3 devices. Would they give an ADSL1 user any benefit at all, or are they VDSL2-only?

Is a zero-component-count unit better because there's no possibility of inline loss bearing in mind I need ever mV I can get.

I'm not sure where I might grt a genuine new BT Mk2/Mk3 (which?) faceplate to try and answer my own question?

[burakkucat]

. . . as I have no phones at all (and have never had any extensions), I’m now using Andrews & Arnold 'blank' faceplates with no filtering in them, just an RJ-45 socket which is straight through to the test socket internally.

Someone told me that if there are no phones, then there's nothing to filter, so I might as well go the straight-thru blank plate route.

Is that actually true?

Yes, absolutely.

Do the Pressac / Adslnation / BT Mk2 / Mk3 have any merits for someone in my situation?

None whatsoever.

I read about the filters in BT Mk2 and Mk3 devices. Would they give an ADSL1 user any benefit at all, or are they VDSL2-only?

When filtering is required, the Mk 3 SSFP would be the best choice for any xDSL circuit.

Is a zero-component-count unit better because there's no possibility of inline loss bearing in mind I need ever mV I can get.

That is, again, absolutely correct. You need to minimise the insertion loss and the best way to achieve that is not to have any components inserted into the circuit! 

At the basic level -- a dangley micro-filter -- there is just a low-pass filter between the incoming line at the telephone. There is nothing at all between the incoming line and the modem/router. It is straight through. All modem/routers will have a high-pass filter within, just "past" the "far side" of the socket.

[Weaver]

@burakkucat  - Many many thanks for saving me a lots of wasted time faffing about. Your wisdom much appreciated. If I could send you a packet of Dreamies, then I would, but my Siamese girl, Caoimhe, would glare at me for a month.

[GigabitEthernet]

The MK2 or MK3 actually might perform better even on a line with no extensions as they have a REIN filter built in...

[Weaver]

The MK2 or MK3 actually might perform better even on a line with no extensions as they have a REIN filter built in...

Could you clarify/expand for me AlecR? Am half asleep this morning after a rough night.

Pondering on the possible merits of the MK2 or MK3 is what set me wondering about this. Also there was I believe a positive report about a (Zyxel?) VDSL2 modem that had an input filter stage that was suggested as a reason for the device's strong performance.

[GigabitEthernet]

I was told by an Openreach engineer when I had a REIN issue, that the REIN team recommend a MK2/MK3 SSFP is installed as it has an RF3 filter built into it (I believe a coil of wire).

[Weaver]

> that the REIN team recommend a MK2/MK3 SSFP is installed as it has an RF3 filter built into it (I believe a coil of wire)

Perhaps I had better start digging around for one.

Set me thinking now... :-)

[4candles]

Probably worth a punt for the price - I certainly would in your place.

If your ultra-long line is nevertheless 'clean', you may, as you hint in your OP, be better off without the additional insertion loss.

IMHO a small price to pay to 'suck it and see'.

Whatever happens, I'm sure I won't be alone in being interested in any result.

[Weaver]

Anyone have any shopping tips?

[roseway]

You can get them on eBay: http://www.ebay.co.uk/bhp/vdsl-faceplate

The first three items (at present) appear to be genuine, but I would avoid the apparently cheap one, because it's an auction and there's postage to be added.

[Weaver]

I notice a good few people complaining that they got an BT Mk3 and their sync rate was lower. Unless of course this is just the wrath of DLM and the testing is not sufficiently scientific, it may be that the Mk3 filtering does nothing good for them as they don't have a real noise problem, but it does add insertion loss at presumably the highest bins.

However, what is too tight for them might be better for me, still on ADSL1, than a Mk2.

The alternatives are that insertion loss means either a straight-through plate or a MK1 NTE2000 or NTE2005 would be the winner.

Unfortunately, my hands are not up to all that screwing involved in a lot of swapouts, even with an electric screwdriver. So I'll have to make a plan.

[GigabitEthernet]

I personally notice no difference between the MK2 and MK3 faceplates in terms of sync speed or errors.

My kind suffers from REIN too so I don't know if that helps?

[Weaver]

@AlecR apol, could you remind me what kind of line characteristics and xDSL protocol etc you have? Many thanks.

[GigabitEthernet]

I am on an FTTC connection, I think 800m from the cabinet.

I get a ridiculous number of FECs (millions) on either faceplate and the CRCs are also very similar too. Like I say, I believe REIN is the cause of most of these issues, which neither the MK2 nor MK3 seem to affect in a very big way.