Using a higher rated DC transformer

[Mike Monett]

So--inquiring minds would like to know: just what's inside that
particular wall-wart, to give results like that?

Cheers,
Tom

Hee Hee. Now we get to examine different ways of opening wall-warts

I just whack it with a hammer then glue the pieces back together with
PVC/CPVC Transition Cement. That glues just about every plastic except
polyethylene. Just dab a bit on first. If it softens the plastic, away you
go!

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm

 

[Tom Bruhns]

Hee Hee. Now we get to examine different ways of opening wall-warts

I just whack it with a hammer then glue the pieces back together with
PVC/CPVC Transition Cement. That glues just about every plastic except
polyethylene. Just dab a bit on first. If it softens the plastic, away you
go!

Well, we need to get Tam to do the autopsy (or perhaps just exploratory
surgery), or to ship one of us the wall wart in question. I was going
to measure one that I have that I've noticed in the past has a large
drop from OC to full load, but I haven't found that Round Tuit yet.

Cheers,
Tom

 

[Jim Thompson]

Hee Hee. Now we get to examine different ways of opening wall-warts

I just whack it with a hammer then glue the pieces back together with
PVC/CPVC Transition Cement. That glues just about every plastic except
polyethylene. Just dab a bit on first. If it softens the plastic, away you
go!

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm

Got any neat trick to open a TV remote? I've got a JVC TV that's
outlived three remotes... the ON/OFF always fails :-(

...Jim Thompson

 

[[email protected]]

Yes.

The only caveat is that the 12V rating may be 'nominal' and vary a bit with the
current drawn but there's precious little you can do about that and these things
are hardly precicion power supplies in the first place, so get the one the Radio
Shack guy says will do the job.

Graham

Um, Graham, unfortunately, you have entirely too much confidence in
Radio Shack's sales reps.

;-)

I've been to a local Radio Shack, looked in the battery holder bin,
asked the sales girl if they had any AAA battery holders, was told No,
went back to the bin, looked for a few more minutes, then found them.

Michael

 

[Spehro Pefhany]

Hee Hee. Now we get to examine different ways of opening wall-warts

I just whack it with a hammer then glue the pieces back together with
PVC/CPVC Transition Cement. That glues just about every plastic except
polyethylene. Just dab a bit on first. If it softens the plastic, away you
go!

Gently squeeze it in a vice/vise, preferably with plastic jaws. The
plane described by the joint should be parallel to the jaw gripping
surfaces. You may have to turn it a bit and squeeze again. It should
crack apart almost entirely at the line, and it should be possible to
glue it back together again so that you can barely tell it was opened.

Don't blame me if you electrocute yourself or someone else or start a
fire due to improper modifications.


Best regards,
Spehro Pefhany

 

[[email protected]]

No. It just means that you can draw up to 2.5 A from it without having the
thing catch fire.

Here is what a typical non regulated supplyu might do. I have a 12V, 300ma
AT&T telephone transformer. At no load it puts out 16V. At 10 ma it puts out
14V. At 300 ma it puts out 12V.

Tam

<newbie>
Would putting a 12V voltage regulator on the wall wart's outputs help
any?
</newbie>

 

[Mike Monett]

Well, we need to get Tam to do the autopsy (or perhaps just exploratory
surgery), or to ship one of us the wall wart in question. I was going
to measure one that I have that I've noticed in the past has a large
drop from OC to full load, but I haven't found that Round Tuit yet.

Cheers,
Tom

OTOH, we could do a semi engineering approach and ask Tam to measure the
output voltage with a bunch of different resistors.

Then we can plot the internal resistance as a function of output current.
Maybe that would tell us something.

Regarding your suggestion on flattening the top of the waveform, you'd
think that would improve the regulation and reduce the internal resistance.

To try it, I added a clamp to flatten the top of the waveform and give it a
ramp just like the normal line voltage with lots of TV's and hi-fi's turned
on in the neighbourhood.

The output voltage behaved very strange. The voltage remains constant with
a 300 mA load no matter what the source resistance is. For example, 10
megohm and 10 ohm give the same output voltage of about 14.125V.

I played with it for a while, but it keeps doing the same thing. I think
it's time to let Helmut have a look.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm

 

[Tom Bruhns]

<newbie>
Would putting a 12V voltage regulator on the wall wart's outputs help
any?
</newbie>

Well, yes, sure. But that's a wonderful leadin to some related topics
to ponder...

The reason (presumably) that they didn't put a regulator in the wall
wart in the first place is that the equipment to which it's supposed to
attach isn't bothered by voltages over some moderate range, and the
expense of adding a regulator (taking more room, resulting in a larger
case, ...dissipating more power, resulting in a larger case, ...) was
not worth it. Remember, too, that we're talking an extreme range of
loads here, and the equipment with which it is supposed to work could
also provide a fairly constant load that results in a much more stable
output voltage than indicated by the range from no load to full load.
Also, very often the equipment connected to the supply has a regulator
inside it. In fact, think of the reasons for using a wall wart in the
first place: a key one is that you keep line voltage out of your
equipment, meaning you don't have to worry about a passle of
regulations concerning the safety of equipment connected to the line.
You buy your wall warts from someone who has already taken care of that
part of the equation for you. Doing the regulation is easy for you;
taking care of world-wide power safety issues is not. You also avoid
the hassles of dealing with the plethora of international power plugs
and switching power voltages in your equipment; you use a cheap
connector on your equipment and stock a range of wall wart types.

There are a lot more switching-type "wall wart" supplies out there now
than there were twenty years ago or even ten, because the technology
and volume has made them much cheaper and smaller than they used to be,
because of a demand for supplies that run over a much broader range of
input voltages, because of demand for or regulations for supplies that
are more efficient (especially when lightly loaded), ... and the
switching supplies pretty much always have much better output
regulation than the transformer-driven unregulated ones. That's
inherent in almost all the switchers.

Cheers,
Tom

 

[Tom Bruhns]

OTOH, we could do a semi engineering approach and ask Tam to measure the
output voltage with a bunch of different resistors.

Then we can plot the internal resistance as a function of output current.
Maybe that would tell us something.

Regarding your suggestion on flattening the top of the waveform, you'd
think that would improve the regulation and reduce the internal resistance.

To try it, I added a clamp to flatten the top of the waveform and give it a
ramp just like the normal line voltage with lots of TV's and hi-fi's turned
on in the neighbourhood.

The output voltage behaved very strange. The voltage remains constant with
a 300 mA load no matter what the source resistance is. For example, 10
megohm and 10 ohm give the same output voltage of about 14.125V.

I played with it for a while, but it keeps doing the same thing. I think
it's time to let Helmut have a look.

Well, yes, if you have any series resistance in the circuit, you'd
expect at the very least that you'd have some variation from that; and
certainly rectifier diodes have more drop at 300mA than at 1uA. When I
was SPICing around with the choke-input filter circuit with a light
load, I did notice that when you first fire it up, it can overshoot
quite a bit, and charge the output cap to a voltage higher than the
peak of the AC input to the rectifier. I suppose the limit on that,
assuming a sinusoidal input and ideal diodes, is 4/pi times the sine
peak.

Actually, some of the waveforms I've seen surprised me by not being
flattened on the tops as I would have expected, but having a little
peak there (and it was these strange ones I was thinking of, not
flat-topped ones that would of course have better output regulation),
slightly more "triangular" than a sine instead of being more "square".
Don't ask me why. I just thought for a moment and realized that there
was no way I was going to be able to figure out the cause easily,
shrugged, and moved on to other things. Sometimes life's just too
short to worry about such things.

Cheers,
Tom

 

[Mike Monett]

Actually, some of the waveforms I've seen surprised me by not being
flattened on the tops as I would have expected, but having a little
peak there (and it was these strange ones I was thinking of, not
flat-topped ones that would of course have better output regulation),
slightly more "triangular" than a sine instead of being more "square".
Don't ask me why. I just thought for a moment and realized that there
was no way I was going to be able to figure out the cause easily,
shrugged, and moved on to other things. Sometimes life's just too
short to worry about such things.

Cheers,
Tom

I look at it as more of a sawtooth at the peak while all the caps in the
neighborhood get charged. Then the diodes cut off and the line voltage
waveform continues on as a normal sine wave. So that might give the
triangular peak you mentioned.

You are lucky. I find these little anomalies fascinating. When one catches
my attention, I simply have to find out what's going on, and I'm not
satisfied until then. Not knowing the answer bugs me.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm

 

[Tam/WB2TT]

...

I agree that for a common wall-wart type supply, the 10mA voltage
sounds too low, given the open-circuit and 300mA load output voltages.
16V OC, ~15.5V @ 10mA, and 12V @ 300mA seems more like what I'd expect.
I can think of a couple possible ways that you could see such
voltages, but they aren't all that plausible, given that it's a simple
wall-wart. One way is that the supply uses a choke input filter. Then
you get an open-circuit output voltage nearly as high as the sine input
peak voltage, and the output at full load is considerably less: diode
drops and I*R drops lower than the average of the absolute value of a
sine, which is 2/pi times the peak voltage, or sqrt(8)/pi = 0.9*Vrms.
The output voltage at 1/30 full load could well be mid-way between the
full load and open circuit values in that case. A second way is to put
an NTC thermistor in the output path: it's a moderately high
resistance at 10mA, but at 300mA it heats up and drops to a low
resistance. -- Hey, I did say it's not very plausible, didn't I??
;-) Another thing to consider: the line voltage may not be very
sinusoidal. I've seen some pretty ugly ones, but probably not ugly
enough to give results like that.

So--inquiring minds would like to know: just what's inside that
particular wall-wart, to give results like that?

Cheers,
Tom

I had responded to your other posting, but it never showed up. What I
mentioned is that this is a real Bell System transformer. I believe it had
to withstand a short circuit without self destructing. It was called
reluctance protection, or something like that. My guess is that there is a
lot of inductance involved, but it is all in the way the transformer is
wound. This was probably not a good example for the OP.

I also measured a Nokia cellphone charger. Claims it is 3.6V, but measures
7V at no load. Could well be the same technology. These people have to meet
all kinds of safety and fire code requirements.

Tam

 

[Mike Monett]

I had responded to your other posting, but it never showed up. What I
mentioned is that this is a real Bell System transformer. I believe it
had to withstand a short circuit without self destructing. It was
called reluctance protection, or something like that. My guess is that
there is a lot of inductance involved, but it is all in the way the
transformer is wound. This was probably not a good example for the OP.

I also measured a Nokia cellphone charger. Claims it is 3.6V, but
measures 7V at no load. Could well be the same technology. These
people have to meet all kinds of safety and fire code requirements.

Tam

Hi Tam,

We are starting to get interested to find out how your transformer can
behave in this fashion. Could you do us a favor and measure the output
voltage with different resistors? For example, values from, say, 33 ohms to
1k would be great.

The smaller values would dissipate some power. For example, 33 ohms might
give (12^2) / 33 = 4.36 watts. You could probably get away with a 2 Watt if
you measured the voltage quickly then disconnected the resistor.

Another way to do it is to get a bunch of resistors of the same value and
add them in parallel one at a time. For example, 560 ohms might dissipate a
little over 1/4 watt. Ten in parallel would give 56 ohms.

This would give a nice data set to work with, and it might be interesting
for you to see if we can figure out what's inside.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm

 

[Tom Bruhns]

Mike Monett wrote:
....

You are lucky. I find these little anomalies fascinating. When one catches
my attention, I simply have to find out what's going on, and I'm not
satisfied until then. Not knowing the answer bugs me.

Well, I'm generally that way, too. As I recall, it was more that at
the time I had too many other things on my mind that were more
important, and I never got around to looking at it again. I should
perhaps give it a try, though we're not in the same place we were then,
and things may be entirely different.

Saw your posting to Tam asking for more info, and I'll second that. It
will be interesting to try to figure it out. I'm not seeing how the
transformer construction, probably with a magnetic shunt like in a neon
tube transformer, would cause the 10mA voltage to be midway between the
300mA and the no load voltages.

Cheers,
Tom

 

[Eeyore]

The reason (presumably) that they didn't put a regulator in the wall
wart in the first place is that the equipment to which it's supposed to
attach isn't bothered by voltages over some moderate range

It's *highly* likely that equipment that needs a stable voltage wil have a voltage
regulator inside the unit. It makes more sense to do it this way.

Graham

 

[Mike Monett]

Saw your posting to Tam asking for more info, and I'll second that. It
will be interesting to try to figure it out. I'm not seeing how the
transformer construction, probably with a magnetic shunt like in a neon
tube transformer, would cause the 10mA voltage to be midway between the
300mA and the no load voltages.

Cheers,
Tom

That's what caught my eye in the first place. I figure we might get a lot
more success asking Tam to do some measurements instead of whacking his
transformer with a hammer

I posted a query to Helmut regarding the strange behavior of LTspice. Here
is the problem description from the note in the file:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The purpose of this circuit is to examine the voltage regulation of
a standard bridge rectifier. Diodes D1..D4 form a normal bridge
circuit. The current source, ILoad, is a convenient way to change
the load current. This circuit operates as expected when the second
bridge is not present. Diodes D5..D8 form a second bridge to
simulate the peak clamping effect when many TV's and HI-Fi's are
turned on in the neighbourhood. C3 sets the slope of the peak ramp.

When the second bridge is added, resistor R4 no longer has any
effect on the output voltage of the first bridge rectifier. It
remains at about 14V as R4 is changed from 10 ohms to 10 megohms.

The first bridge operates normally when the current source is
replaced by a resistor.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The file is

http://tech.groups.yahoo.com/group/LTspice/files/ Temp/35695454.zip

This might be of interest for people working with dc power supplies. I'll
post the results of Helmut's investigation.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm

 

[Tom Bruhns]

That's what caught my eye in the first place. I figure we might get a lot
more success asking Tam to do some measurements instead of whacking his
transformer with a hammer

I posted a query to Helmut regarding the strange behavior of LTspice. Here
is the problem description from the note in the file:

Mike, I hate to break it to you, but notice that you have the second
bridge output referred to ground...

Cheers,
Tom

 

[Homer J Simpson]

Hee Hee. Now we get to examine different ways of opening wall-warts

I just whack it with a hammer then glue the pieces back together with
PVC/CPVC Transition Cement. That glues just about every plastic except
polyethylene. Just dab a bit on first. If it softens the plastic, away you
go!

The absolute best way is to break them open like this - or by dropping them
on cement. Never saw them open - you lose plastic and may do more damage to
the guts than a broken case would do.

 

[Jeff L]

Got any neat trick to open a TV remote? I've got a JVC TV that's
outlived three remotes... the ON/OFF always fails :-(

Drop it on a concrete / hard floor, if it's a snap together one. The buttons
ooze oil, and gum up / coat the carbon contacts with oil, causing them to
stop working. A good washing with dish soap, cleaning the silicone rubber
buttons and PCB usually fixes them. Make sure you rinse them well after the
soap treatment, and putting them somewhere like on top of a radiator for
few hours should insure they are nice and dry (but I doubt you will have
much of a problem with that in Arizona!)

Jeff