Hello,

I was wondering if someone can help me with my dial up connection. I
have an internal modem, Lucent Win Modem, 56k. My dial up provider is
BlueLight Internet.

My problem is each time I want to connect, I have to unplug the phone
cable from the modem, run the BlueLight dialer, and then plug the
phone cable to the modem again. If I don't unplug the cable after
running the BlueLight dialer program, I will have a "No dial tone
detected" error.
Also, after I close the dialer (disconnect my internet) and if I don't
unplug the phone cable from the modem, I won't be able to use the
phone nor will someone be able to call me on my phone!! When I lift
the phone receiver, I will either hear silence or annoying loud
repeated beeps.
I am not sure what causes this, it started behaving like this about 6
months ago.

Thank you for any replies.

On 8 Jul 2004 23:01:46 -0700, nyoman.kumala@gmail.com (Nyoman Kumala)
put finger to keyboard and composed:

It looks like the hook relay contacts in your modem are permanently
closed. If you are handy with a soldering iron, it should cost you
about US$5 to replace it.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

Franc Zabkar <fzabkar@optussnet.com.au> wrote:

Can I do this in a USR 56K external? (ATI0 returns "5601".)
Can anyone help me identify the hook relay? Pointer to a circuit
board diagram with labels? Relay identifying marks?

I took apart an old USR 1200 and readily identified the (big) hook
relay. Upon opening the 56k, I don't see anything that looks similar
or that looks removable or big enough to contain a relay.

(My USR 56k works fine except that, after a dialup session, it doesn't
go on hook until I physically unplug it from the phone line. Powering
down the modem leaves it still off hook.)


--
Mike Spencer Nova Scotia, Canada

mspenbellscer@tallswhistleships.ca
(Remove bells and whistles)

On 10 Jul 2004 02:32:22 -0300, Mike Spencer
<mds@bogus.nodomain.nowhere> put finger to keyboard and composed:

Can you post a photo or a scan on your webspace?

Or can you post the part numbers of the various chips/components?


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

me> Can I do this in a USR 56K external? (ATI0 returns "5601".)
me> Can anyone help me identify the hook relay?

To which Franc Zabkar <fzabkar@optussnet.com.au> replied:

Ah! I think so. It'll take a day or two. Have to get my wife to
take the photo because my cam (Kodak DC40 :-) doesn't do closeups.

Meantime, I'll see if there are readable numbers on the bits.

Thanks, Frank. More soon.

--
Mike Spencer Nova Scotia, Canada

Franc Zabkar <fzabkar@optussnet.com.au> asked:

Here you go:

http://home.tallships.ca/mspencer/modem-qry.html

Problem is recapped there. (HTML and 218k GIF image)

TIA

--
Mike Spencer Nova Scotia, Canada

On 14 Jul 2004 23:08:07 -0300, Mike Spencer
<mds@bogus.nodomain.nowhere> put finger to keyboard and composed:

Your modem appears to have a silicon DAA, ie it is transformerless.
The host and line side chip pairs are the two 16-pin ICs above and to
the left of the speaker. IIRC, the part numbers probably begin with
Si3xxx or Si2xxx. These chips are made by Silicon Labs. If you can
tell me the exact part numbers, I may have some reference circuits
that could narrow down the fault to a specific component, hopefully a
transistor or passive component.

Otherwise, the component you have identified is a CU4032K230G2
varistor, which is a device used for surge protection:
http://www.epcos.com/inf/70/db/var_01/01430148.pdf

It is rated at 230Vrms, 300VDC, 1200A, 17.0J, 0.25W. This device may
be your culprit if it is "leaky". If so, then you could simply cut it
out of circuit and replace it with a common 275V MOV.

As for the "P31X HO" device, I have no idea what it is. It may help if
you can tell me how it is identified on the circuit board, and how
many pins it has (looks like 2?). For example, I suspect the varistor
(a 2-pin device) is designated as Z1, Z2, etc. A relay would have at
least 4 pins and a PCB reference such as K1, K2, etc.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

WRT:

http://home.tallships.ca/mspencer/modem-qry.html

Franc Zabkar <fzabkar@optussnet.com.au> wrote:

The one above the speaker (11:30) is SI3012-KB (0050DAAA32).

The one left of the speaker (10:00) is SI3021-KS (0044CECAAE.)

Cool. Well, those are the numbers.

This fault apparently occurred during a period during which the telco
was unable to locate the cause of 60hz hum on the line. History now,
tho.

I don't see any PCB reference using K. The object you've identified
as a varistor has PCB ref RV1. The "P31X HO" component has PCB ref
TH1.

The only relatively large component with 4 pins is at lower left
in a line between the vol control and the on/off switch. It is
marked:

- LT +
DF04S
~96~

with PCB ref D1.



--
Mike Spencer Nova Scotia, Canada

Mike Spencer wrote:

It's most likely that the DAA was damaged by a lightning surge, and may
have been the source of hum on your phone line. Good luck.


--
Art Jackson W4TOY Owensboro, KY USA
Life is God's open book test. In order to pass,
you must open His book to find the answers.

WRT:

http://home.tallships.ca/mspencer/modem-qry.html

Art Jackson <nospam@ringading.buz> opined:

How do you get that? It's ca. 4 years old and was bought at a junk
store two years ago. USR/3Com has a 4+ year warranty for 2nd (or nth)
owners?

I try never to deal with the corporate/consumer interface. I suppose
I would if it were clear that USR owed me a modem.

You mean that my modem, once damaged, was the source of hum?
Absolutely not. But whatever caused the hum (or some transient event
that initiated the hum-causing condition) may have damaged the modem.
The hum, which lasted for a month and was eliminated only after 4
telco teams [1] had tried to find/fix it, unquestionably originated
outboard of the demarc. Neither of two USR 56K modems would connect
at all during the hum condition and I had to fall back on an ancient
Practical Peripherals 14.4 (which would connect at 9600 bps). After
the hum was eliminated, both USR 56K modems once again worked, save
only that the one in use when the hum commenced would no longer hang
up.

Thanks.

[1] Telco guys are on strike. Service calls are being
made by management "volunteers" -- guys from marketing, accounting
etc. -- instead of techs. I was able to teach one of these fellows
how to strip the end of a wire with a pocket knife. :-)

--
Mike Spencer Nova Scotia, Canada

On 15 Jul 2004 04:33:37 -0300, Mike Spencer
<mds@bogus.nodomain.nowhere> put finger to keyboard and composed:

A brief description can be found here (let me know if you would like
the datasheet):
http://www.silabs.com/products/wireline/si3034.asp

Your modem's hook switch transistor (Q1?) connects to pins 7 & 8 of
Si3012. If it were shorted C-to-E, then it would cause your off-hook
symptom.

Whenever I've experienced hum, the telco has attributed the cause to
water in the cable.

You won't. The Si30xx chips perform the hook relay function.

RV = variable resistor, which is essentially what a varistor is. Other
designers use Z to refer to MOVs (metal oxide varistors), and still
others use Zs for zener diodes.

This suggests that this device is a thermistor. Thermistors are used
for temperature sensing. I don't understand its function in this
circuit. The Silicon Labs application circuit has no such device.

That's a bridge rectifier consisting of four diodes. This component is
part of the power supply and will have no bearing on your problem. In
fact, no component within the host side electronics can cause your
off-hook problem. You need to confine your search to the rectangular
region between the speaker and the phone and line sockets. This area
of the PCB is the DAA. It is electrically isolated from the rest of
the electronics by the three capacitors labelled Cx (?) which connect
the Si3012 to the Si3021.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

Franc Zabkar <fzabkar@optussnet.com.au> wrote:

Using what I've learned from your posts, I found this:

http://www.modemsite.com/56k/lightning.asp

(q.v.) where it is asserted that

...the usual fault is not with the DAA circuit. In modern
modems, there are usually low value resistors (around 100
ohms) right at the tip and ring inputs. These are what
usually fail.

I surmised as much, now that I know what "DAA" is.

So now I have to do my homework.

Thank you very much. More later if I have some success w/ repair.


--
Mike Spencer Nova Scotia, Canada

On 15 Jul 2004 20:08:30 -0300, Mike Spencer
<mds@bogus.nodomain.nowhere> put finger to keyboard and composed:

If these resistors go open, then the connection between the modem and
the phone line will be broken, ie your modem will not draw a dialtone.
Clearly your modem has some other fault.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.