7 Jun
2002
7 Jun
'02
7:13 a.m.
It does sound a _little_ high, but it might be OK. Try
loading it (A 100
Ohm wirewound resistor would do, for example). See if that brings the
voltage down to nearer 9V.
I tried a 100k Ohm since I didn't have a 100 Ohm, no effect. I don't have
a bulb or anything handy to try with that.
I think I am going to dig out the modem it goes to, and test the voltage
with it hooked to that. That will at least give it real use load (the
modem powers up just fine, it just fails to respond to all AT commands
except for getting the firmware, and it won't attempt to go off hook or
dial... I'm not 100% convinced the modem is really "dead" so much as it
just doesn't want to co-operate).
Thanks to all those that responded... I knew I could get some good
answers here.
-chris
<http://www.mythtech.net>
7 Jun
7 Jun
9:10 a.m.
On Fri, 7 Jun 2002, Chris wrote:
I think I am going to dig out the modem it goes to,
and test the voltage
with it hooked to that. That will at least give it real use load (the
modem powers up just fine, it just fails to respond to all AT commands
except for getting the firmware, and it won't attempt to go off hook or
dial... I'm not 100% convinced the modem is really "dead" so much as it
just doesn't want to co-operate).
Hey, you know what? I missed the beginning of this thread, but this is
exactly the sort of behavior you get from a modem which requires a 9V AC
power supply, when you run it with a 9V _DC_ supply.
ok
r.
9:27 a.m.
Most of the Hayes modem platforms required AC because the negative voltage
for the RS-232 was generated from it, along with the negative voltage rail
for some op-amps. In the case of a Hayes modem, it shouldn't have affected
basic off-hook and command functions, but audio would be pretty bad, DTMF
synthesis may or may not work, and RS-232 would be out of spec (but probably
workable, since most RS-232 will work at GND and +5)
--John
-----Original Message-----
From: cctalk-admin(a)classiccmp.org [mailto:cctalk-admin@classiccmp.org]On
Behalf Of r. 'bear' stricklin
Sent: Friday, June 07, 2002 17:10
To: Classic Computer
Subject: Re: [OT] measuring DC wall wart
On Fri, 7 Jun 2002, Chris wrote:
I think I am going to dig out the modem it goes to,
and test the voltage
with it hooked to that. That will at least give it real use load (the
modem powers up just fine, it just fails to respond to all AT commands
except for getting the firmware, and it won't attempt to go off hook or
dial... I'm not 100% convinced the modem is really "dead" so much as it
just doesn't want to co-operate).
Hey, you know what? I missed the beginning of this thread, but this is
exactly the sort of behavior you get from a modem which requires a 9V AC
power supply, when you run it with a 9V _DC_ supply.
ok
r.
9:31 a.m.
On Fri, 7 Jun 2002, r. 'bear' stricklin wrote:
On Fri, 7 Jun 2002, Chris wrote:
I can certainly agree with that. I've seen it happen several times. Most
all the modems I have around here use 9V AC wall-warts too. Some use DC or
exotic voltages, but 9V AC seems to be the norm.
-Toth
I think I am going to dig out the modem it goes
to, and test the voltage
with it hooked to that. That will at least give it real use load (the
modem powers up just fine, it just fails to respond to all AT commands
except for getting the firmware, and it won't attempt to go off hook or
dial... I'm not 100% convinced the modem is really "dead" so much as it
just doesn't want to co-operate).
Hey, you know what? I missed the beginning of this thread, but this is
exactly the sort of behavior you get from a modem which requires a 9V
AC power supply, when you run it with a 9V _DC_ supply. 
11:02 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 518 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020608/8885a5d9/at…
12:44 p.m.
It seems to me that you can take a common TO-3 NPN transistor, e.g. the
2N3055, or something of that sort, drive its base from a pot between the
positive and negative voltage, and put a low-valued resistor, e.g. 0.5 ohms or
so in the emitter-to-ground path, you have a programmable current sink. If
you then tweak the pot until the voltage on the collector is the rated
voltage, the voltage on the emitter will tell you what the current is. Else,
you can tweak the pot until the voltage on the emitter is such that it
reflects the rated current, based on the resistor value. The emitter will
follow the base voltage, hence the term "emitter follower." By doing this,
you eliminate the need for a high-power resistor, and, you have a wider range
of voltages and currents you can monitor. If you're concerned about the
transistor's gain, you can use a darlington pair.
Dick
----- Original Message -----
From: "Chris" <mythtech(a)mac.com>
To: "Classic Computer" <cctalk(a)classiccmp.org>
Sent: Friday, June 07, 2002 1:13 PM
Subject: Re: [OT] measuring DC wall wart
It does sound a
_little_ high, but it might be OK. Try loading it (A 100
Ohm wirewound resistor would do, for example). See if that brings the
voltage down to nearer 9V.
I tried a 100k Ohm since I didn't have a 100 Ohm, no effect. I don't have
a bulb or anything handy to try with that.
I think I am going to dig out the modem it goes to, and test the voltage
with it hooked to that. That will at least give it real use load (the
modem powers up just fine, it just fails to respond to all AT commands
except for getting the firmware, and it won't attempt to go off hook or
dial... I'm not 100% convinced the modem is really "dead" so much as it
just doesn't want to co-operate).
Thanks to all those that responded... I knew I could get some good
answers here.
-chris
<http://www.mythtech.net>
8 Jun
8 Jun
12:12 p.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1133 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020609/4e805ccb/at…
3:27 p.m.
Well, perhaps you can find out whether or not a given wall-wart can do what
the label says, in general, I've found they don't, even within 10%. With the
fixture I suggested, which takes about 10 minutes to build, maybe 30 if you do
it right, and I agree that a heatsink, in this case a big one, is indicated,
not because there are wall-warts that might damage the 3055, but because a
nice 5"x8" heatsink makes a handy platform to which to attach things,
(TO-3's
are rated, typically for 80W, and wall warts aren't.) so you can use a fairly
large 10-turn pot with a vernier or the like, and a big P-clamp to tie it
down. You also get a place to which to screw the barrier strip you will need.
If you want to be really fancy you can even include individual binding posts.
As for replacing one component, a resistor, with three or more, I'd point out
that the resistor approach will, at best, provide an approximation with the
one and only wall wart in the universe, if they've built it yet, that actually
does what the sticker says. The three components I mentioned give you a tool
that will tell you what the voltage is at a given current, and what the
current is at a given voltage, either of which you can measure. If you have
lots of AC wall-warts ( of the 40 or so that I have in the box downstairs,
only about three are AC types.) you need simply put a pridge rectifier at the
collector side of your circuit and you're on your way. When you're done you
have a tool that will measure many different wall-warts, not just one. What's
more, you don't have to go looking for that one resistor that almost works.
Dick
----- Original Message -----
From: "Tony Duell" <ard(a)p850ug1.demon.co.uk>
To: <cctalk(a)classiccmp.org>
Sent: Saturday, June 08, 2002 6:12 PM
Subject: Re: [OT] measuring DC wall wart
>
> It seems to me that you can take a common TO-3 NPN transistor, e.g. the
> 2N3055, or something of that sort, drive its base from a pot between the
> positive and negative voltage, and put a low-valued resistor, e.g. 0.5
ohms
or
> so in the emitter-to-ground path, you have a
programmable current sink.
If
True, but why do you insist on over-complicating everything. You want to
take a circuit consisting of one component (the resistor) that will
work for AC and DC warts, and which is relatively robust agianst
overloads and eeplace it with a circuit containing 3 components (at
least), that only works for DC (and is polarity sensitive), and which
can burn out easily if overloaded. That 2N3055 is going to need a
heatsink as well...
You can make power resistors from stuff you find lying about the house
(old heating elements, etc -- heck you can even use a jar of salt water
with 2 electrodes in it at a pinch -- electrolysis is a problem if you
use it for any length of time on DC, but it will act as a power resistor
for long enough to test a PSU). Not everyone has 2N3055s lying about. I
do, but...
-tony
8186
days inactive
8187
days old
7 comments
6 participants
participants (6)
-
ard＠p850ug1.demon.co.uk -
edick＠idcomm.com -
jcwren＠jcwren.com -
mythtech＠mac.com -
red＠bears.org -
tothwolf＠concentric.net