Forum Home
PC World Chat
 
Thread ID: 76153 2007-01-22 22:55:00 Too far from Exchange? _Nah..........Too close pctek (84) PC World Chat
Post ID Timestamp Content User
518455 2007-01-23 21:10:00 Im approx 200meters from our exchange (give or take 50 either way at most), and Im lovin the speeds: 7900kbps up and 1800kbps down :D How far are you?

its a shame your down speed is not as good as your up :D
plod (107)
518456 2007-01-23 23:53:00 You should have heard Woosh when I first joined up. "Your signal is too strong for your phone to work" I was literally left stunned and for the first time I heard that a wireless signal is too strong. Fast forward to 2207 "Your signal is weak and not stable for your BB to work" Thanks guys!!! Make your mind up. beeswax34 (63)
518457 2007-01-24 04:03:00 It's quite reasonable that too strong a signal can cause problems. Try connecting a "line" signal to the "microphone" input of an amplifier. You'll get distortion.

Since telephone lines attenuate high frequencies severely, most of the emphasis in the design of ADSL modems will be on doing as well as possible with small signals, and lots of noise. It would be rare to have "enough" signal. Most ADSL modems will be far from the exchange., by simple geometry.

A (balanced) attenuator might be a good idea.
Graham L (2)
518458 2007-01-24 07:25:00 If the problem is because the signal is too strong, then this attenuator could be knocked up and put in the line to modem to see what effect it may have:

www.imagef1.net.nz

With the values shown in ohms, the attenuation is around 13 dB, which corresponds to putting about a kilometre of telephone wire in circuit, of course you could always get Telecom to lend you a reel of standard telephone cable having about a kilometre of wire on it :)
Terry Porritt (14)
518459 2007-01-24 08:32:00 Remember that your attenuator has to allow for the normal telephone circuit dc on hook/off hook detection:D PaulD (232)
518460 2007-01-24 09:14:00 Remember that your attenuator has to allow for the normal telephone circuit dc on hook/off hook detection:D

Good thinking, Watson.

I'll add some DC blocking and wire it into my modem line tomorrow, and see what difference, if any, in line attenuation the modem reports :)
Terry Porritt (14)
518461 2007-01-24 19:59:00 If the problem is because the signal is too strong,

Well its not.
Its just Telecoms latest line of bull****.
pctek (84)
518462 2007-01-24 20:48:00 I thought you'd previously posted that the problem at Wyndham was a Telecom Network one so why are Xnet chasing an individual line issue? PaulD (232)
518463 2007-01-24 21:36:00 Well its not.
Its just Telecoms latest line of bull****.

I don't believe that story either.

However, to test whether it is BS, you could make up an attenuator.

I just carried out some tests using this modified attenuator with DC blocking:

www.imagef1.net.nz

It was connected between the modem which has a cat5 cable and a home brewed high voltage line protector consisting of a gaseous arrestor and bidirectional transorb diodes. There is cat5 cable via a wall socket to the junction box in the roof. Connections are via RJ11 plugs and sockets.

Without the attenuator, the modem always reports 22dB line attenuation, with the attenuator, it reported 45 dB line attenuation.

So the attenuator is adding 23 dB which would correspond to about an extra 1.7Km from the exchange. (13.8dB per kilometre at 300Khz for 'standard' telephone cable).
Terry Porritt (14)
518464 2007-01-25 01:03:00 Did your ADSL connection still work? ;)

That could be still the rough equivalent of a km or even less, rather than 1.7 km, Terry. The ADSL uses a bit more than 300 kHz. Perhaps you need a dispersive attenuator because the real cable will give a bit of frequency dependent phase shift as well as attenuation. A kilometre length of cable might be an idea. :D

BTW, I have an idea that it's been a long time since telephone circuits were actually 600 ... twisted pairs seem to be somewhere about 100-120 ohms. I doubt if it matters ... SWR doesn't seem to be very important at audio frequencies (and ADSL isn't far off audio), except to the the golden-ear hifi freaks. Unterminated stubs did seem to be a problem for modem signals , but that was a different mechanism.
Graham L (2)
1 2 3