modem holds line..

On 17 Oct 2005 12:56:12 -0700, snipped-for-privacy@yahoo.com put finger to keyboard and composed:

IME the most common cause is stuck contacts in the hook relay. Does the dialtone return if you disconnect and reconnect the modem to the phone line?

-- Franc Zabkar

Please remove one 'i' from my address when replying by email.

Hi...

You'll see a little (probably black) box very near to where the telephone jacks are. Give itaa gentle tap or two with the handle of a pocket screwdriver.

If this causes it to release the line, then the relay is sticking. It may from then on continue to operate, or it may stick again next time you use it.

Ken

the relay (soldered) has approx 20kohms between the coil and 200k between the output, when dis/reconnecting the dialtone returns but then again goes busy by itself!

On 18 Oct 2005 04:27:28 -0700, j snipped-for-privacy@hotmail.com put finger to keyboard and composed:

That's the same behaviour you get when you take your phone offhook and allow the dialtone to time out. It means that something is placing a load across the line, but not a dead short. If the relay contacts are open, then the only other culprit that comes to mind is a MOV, but IME these usually fail OC. I'd measure the voltage between the tip and ring inputs. In my case it is 53V when on-hook and about 10V when off-hook. Disconnect various components until the voltage is back to normal.

-- Franc Zabkar

Please remove one 'i' from my address when replying by email.

Many phones, and some modems, have a diode bridge across the line to power part of the electronics (in a non-DECT phone, all of it).

These tend to go leaky when a surge, e.g. lightning somewhere near, causes damage short of total destruction ...

--
Peter Duck

lighting protection is still working and never took a break ?

--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5

That first figure sounds way too high, 200 ohms would be the low side of believable can you check again?

seing as it doesn't look like a short

also can you measure the current through the modem when it's connected to the phone line (but on hook/powered off) a ohm meter running at 1.5-9V soetimes sees things differently to the phone line (48V on-hook)

Is it possible you've measured the ring detector and not the relay?

Bye. Jasen

On Wed, 19 Oct 2005 00:31:37 +0100, Peter Duck put finger to keyboard and composed:

AFAICS the diode bridge (in a transformer based modem) would always be on the host side of the relay contacts, not the line side. The only circuitry on the line side would be the ringer, and this would be isolated by a 1uF 250V DC blocking cap.

-- Franc Zabkar

Please remove one 'i' from my address when replying by email.

took the modem out of the pc, relay in:0.5kohms both ways (the flyback diode seems to be out but not sure if it can be tested without desoldering) and out:approx 500k ; when I plug in the line the modem goes offhook but not right away and the relay output closes.There is a diode bridge but cannot really tell if the relay is on the host side though it is rated 0.5A 20V/0.1A 100V; MOV is open at all times, How do I test OC and can I pull it out if I do not expect any calls?

That sounds much better.

ooh... you hear the relay closing? that doesn't sound like a relay fault.

sounds line a fault elsewhere

caa you use a terminal program to "talk to" the modem if so run all the tests. at&t1 at&t2 etc... see how it responds to commands like ath (go on hook) ath1 (go off hook) atxdp9 (dial nine pulses) both with the modem connected to to the phone line and disconnected.

could the bridge be for the modems power supply?

at 20V rating it'll be atleast half switched by the modem. otherwise behing it

huh?

Bye. Jasen

On 19 Oct 2005 09:25:23 -0700, snipped-for-privacy@yahoo.com put finger to keyboard and composed:

So it's an internal modem, and you've taken it out of the PC? If so, then this means that the hook relay's coil would not be getting any power. If the relay contacts are closing under these conditions, then this means that the relay itself is faulty. To confirm this, desolder one of the contacts and monitor the tip and ring voltage.

-- Franc Zabkar

Please remove one 'i' from my address when replying by email.

Ambient Ham chipset Modem ATI diagnostics OK.Disabled 'wait for dial tone' and It dials but it cannot connect and after that it just 'arrests' the line; tried at&t1 in Hyper terminal-modem 'connects' at

115200 and spits some characters, ath/h1 ok; checked line-around 50 volts but when modem plugged in, it goes to 13v, desoldered MOV is ok but low resistance measured at it's modem board contacts;bridge output goes to darlington...?; as mentioned-in or out of the PC when I plug in the line first you can hear the dial tone &then after 1-2 min. modem is holding the line

On 21 Oct 2005 12:55:26 -0700, snipped-for-privacy@yahoo.com put finger to keyboard and composed:

The AT&Tn commands test the data path to the DSP. They are not really helpful when troubleshooting a DAA fault such as yours.

If you are measuring 13V on both sides of the relay contacts, then the relay is stuck.

Yes, that's the loop current stabilisation circuit, or whatever it is called. It would be on the host side of the relay contacts. If you are measuring any voltage there, then the relay is stuck.

Change the relay, or at least desolder it.

-- Franc Zabkar

Please remove one 'i' from my address when replying by email.

Franc Zabkar =D1=98=D0=B5 =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0=BE

Shouldn't I suspect the transistor i.e. the circuit feeding the coil-looks like upon line plug-in voltage buids up gradually to switch the relay. You mentioned OC-if it's only function is ring detection can i remove it, though OC's input diode seems to be connected also to relay coil...

If you add: &C2&D2

to your modem init string, it should hang up.

On 22 Oct 2005 04:54:15 -0700, snipped-for-privacy@yahoo.com put finger to keyboard and composed:

The transistor is driven from the DSP chip which gets its power from the slot. The relay's coil voltage also comes from the slot. No slot = no power. If the transistor or coil are getting power from the phone line, then you have a *serious* isolation problem.

Change the relay, or at least desolder it.

OC = open circuit

-- Franc Zabkar

Please remove one 'i' from my address when replying by email.

Thanks again checked the transistor driving the coil-ok as well as flyback diode Seems defective optocoupler(OC) could be connecting line to the relay even when modem out of PC?

checked relay-desoldered clicks just fine but is it possible that when defective it closes the output without stimulus at the coil? here is daa schematic

Since mov is ok I do not see what else can arrest the line...

Hi...

That relay is so small and flimsy it may well be that you can hear it click even though the contacts may be welded shut.

Ken