Print

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

[Weaver]

Some day, when DSL has become a seriously endangered species, we will need to collect DSL CPE modems and DSLAMs and park them in a museum installation. I don’t mean just in a glass case, but connected to a seriously long piece of copper cable. I have a lovely moor but OR will want all their copper here back some day so I won’t be that much help. I think that to see DSL alive and coping with the real problems of length and interference, never mind crosstalk, copper will be needed in a huge length that goes out and comes back to where it started so that the ATU-R modem (xTU-R ?) and ATU-C modem can be seen parked next to one another. Surely BT must do this at Ipswich (or wherever?) when they test systems, but maybe they use more artificial means.

Anyway, I have water, foul weather, ice on occasion and plenty of wind. All these museum facilities can be supplied at very low cost. Burakkucat even has an ATU-C if memory serves, so has all the hardware, no ?

[burakkucat]

xTU-C and xTU-R where x is either A or V (i.e. ADSL or VDSL), respectively, and TU is transceiver unit.

The -C device is the central office transceiver unit (i.e. the exchange or cabinet located unit) and the -R device is the remote transceiver unit (i.e. the CPE).

[Alex Atkin UK]

Could a potentiometer be used to simulate attenuation or would it introduce too much interference?

[Weaver]

A potentiometer would only do part of the job - need capacitance and inductance really. But what do I know. It would actually be good to pick up some real interference as all DSL did in life.

@Burakkucat - I don’t know what I was thinking. I knew that but my brain had turned to mush. Thanks for sorting that out!

[burakkucat]

Burakkucat even has an ATU-C if memory serves, so has all the hardware, no ?

Close. 

I have two devices (both made by Planet Networking & Communications) that can be configured as VTU-C. A VC-231 and a VC-230N.

[burakkucat]

Could a potentiometer be used to simulate attenuation or would it introduce too much interference?

One really needs to remember that the metallic pathway is acting as a radio-frequency transmission line in differential mode. So any inserted lump attenuation needs to be a four element pi constructed out of non-inductive resistors.

[Weaver]

Alex, things are different at high frequencies, where electric and magnetic fields surrounding the line play an important role, something that is not a significant factor at low frequencies. As Burakkucat said, see transmission line. I certainly would have potentiometers to tweak things, but it’s not enough for realism. In a more realistic model of a wire or pair at high frequencies, you have both series and parallel resistors, and inductors in series plus capacitors in parallel, as described in that article.

Burakkucat wrote:
> Close

I remembered the moment after I wrote it that Burakkucat’s kit was VDSL2-capable so the term should be a VTU.

[Alex Atkin UK]

Close. 

I have two devices (both made by Planet Networking & Communications) that can be configured as VTU-C. A VC-231 and a VC-230N.

I remember looking at those ages ago, was tempting to get one to see how different modems react with them or even if they would work as a modem.  A bit expensive for a toy though.

[burakkucat]

I remember looking at those ages ago, was tempting to get one to see how different modems react with them or even if they would work as a modem.  A bit expensive for a toy though.

My VC-231 was a new, sealed & unopened, boxed European version which cost £24. I just threw out the enclosed PSU and sourced a UK replacement.

The VC-230N was gifted by Asbokid. I remember reimbursing him for the delivery and he added that sum to his then on-going collection for the "Dogs Trust".

[Alex Atkin UK]

My VC-231 was a new, sealed & unopened, boxed European version which cost £24. I just threw out the enclosed PSU and sourced a UK replacement.

The VC-230N was gifted by Asbokid. I remember reimbursing him for the delivery and he added that sum to his then on-going collection for the "Dog's Trust".

Nice, all the ones I saw were at least four times that price.

[burakkucat]

If you do ever get one, they are simplicity to set up. Just four DIP switches --

• CO or CPE mode
• Fast-path or Interleaving
• Profile: 17a or 35a
• Target SNRM: 6 dB or 9 dB

Numbers 2 - 4 only have relevance when CO mode is selected.

If you were going to use one in CPE mode, you would have to use your pfSense system to be the VLAN end-point.

[kitz]

There are iirc several of telephone museums.    BT do or did maintain a couple of their own and there is also a family/group of enthusiasts who have also set up a private one.   

I remember posting a link to some of their videos posting a couple of years ago where they had created a mini working telephone exchange with old equipment.  I also seem to recall that one of the 'engineers' was a woman who I think had learnt from her father.

[Weaver]

Kitz, I’ll do some digging.

In the meantime, don’t throw away your old modems or any CPE. Never mind if your spouse accuses you oof hoarding, tell him/her it’s for a good cause. We can find somewhere to send it all to. I would volunteer but I don’t have the storage space any more.


I think g.fast modems will be highly valued, because they represent the summit of the DSL art, and because they will be relatively rare. So hang on to yours when you go FTTP.

At the other extreme, I think some early DSL modems will be valuable because of their rarity and because they show that particular point in evolution. Say a USB modem from 20 years ago such as the Thomson Speedtouch ‘frog’ USB modem. Is that correct? The non-USB Speedtouch modems of 20 years ago were really superb IMO, good DSL and amazing software in them, really sophisticated. I doubt that I still have mine, unfortunately. I can’t remember the best model numbers so I’d like some advice.

I think the procedure for storing hardware should be to reset it to factory settings and standard admin IP/username/password, then download and save a copy of the latest manufacturer’s firmware and flash it. Save it somewhere with a readme.txt text file containing full identification info and version number plus the URL the firmware came from. Even if the manufacturer deletes the web pages archive.org / way back machine should archive it all, but best to be on the safe side.

Do you think it would be a good idea to have such readme files in XML and make up our simple standard format for them? That way code could query them all, using XSLT or hack soup code. Or combine them all perhaps.

I’m wondering if eBay can still provide working kit from ten or twenty years ago. I’d like to buy some such kit from eBay but don’t have enough storage and don’t have the facilities to test such kit.

I wonder if in certain cases, where kit to be archived has a PSU brick that is nasty, ancient or sickly we should replace the brick with a low-cost modern good quality equivalent brick. What do you think? That would improve reliability a lot given the age of some desireable h/w.

And the question of what to acquire if absent among our number, the must-haves?

[Alex Atkin UK]

If you do ever get one, they are simplicity to set up. Just four DIP switches --

• CO or CPE mode
• Fast-path or Interleaving
• Profile: 17a or 35a
• Target SNRM: 6 dB or 9 dB

Numbers 2 - 4 only have relevance when CO mode is selected.

If you were going to use one in CPE mode, you would have to use your pfSense system to be the VLAN end-point.

The main reason I didn't get one was looking at the HG612 interfaces and wondering what handles PTM.  If I had known I only needed to handle the VLAN I probably would have, but with FTTP around the corner it makes no sense now.