modding a pc power supply

[kellro...@billburg.com]

I have a salvaged pc power supply I use as an adjustable bench supply.
But it won't go above about 14.9 volts. Apparently it has an
overvoltage protection circuit. The board has a TL494 (house-numbered
equivalent), and the only other chip on the board is a dual comparator.
The power supply is a Fortron/Source FS-250S40G.
I have another supply that also hits a ceiling when I adjust the
voltage, a Falcon Technology PE-250X with a 494 chip and a 14-pin chip
labeled DBL 339.

Is there some way I can disable the overvoltage protection on these
power supplies?

[James Sweet]

<kellro...@billburg.com> wrote in message
news:1127686295.8...@g43g2000cwa.googlegroups.com...

Look up a datasheet for the controller chip and see how they do the
protection, it shouldn't be too hard to trace the circuit and disable it.

[kellro...@billburg.com]

> >
> > Is there some way I can disable the overvoltage protection on these
> > power supplies?
> >
>
> Look up a datasheet for the controller chip and see how they do the
> protection, it shouldn't be too hard to trace the circuit and disable it.

If it was that easy, I wouldn't ask. Those chips don't have a shutdown
pin or anything real obvious like that. I've spent a lot of time
looking at the datasheet and the circuit board.
I was looking for a germane response from somebody with hands-on
knowledge of the issue.

[Michael Black]

But what makes you think there is a voltage limiter in there? The
design may simply be such that it won't output more than a bit above
the intended output. After all, it's not a general purpose power supply,
it's intended for a very specific purpose with very specific voltages.

At the very least, again this is a good reason to look at the datasheet,
there may be resistors in there to limit the range of the variable
that you are presumably playing with, so someone fiddling with it cannot
output too much voltage. Or there may be something else in the design
that limits the maximum output voltage. So you have to fiddle
with the design, and maybe find that it can't be done at all.

Michael

[kellro...@billburg.com]

> At the very least, again this is a good reason to look at the datasheet,
> there may be resistors in there

Nope, no resistors in the data sheet.

[kellro...@billburg.com]

kellro...@billburg.com wrote:
> I have a salvaged pc power supply I use as an adjustable bench supply.
> But it won't go above about 14.9 volts.

I didn't make it clear: the power supply shuts down. Turns itself off.

[DaveM]

"Michael Black" <et...@FreeNet.Carleton.CA> wrote in message
news:dh7ab2$du2$1...@theodyn.ncf.ca...

Michael is correct. THese power supplies are designed very tightly in order
to get good efficiency . If they were designed to be capable of generating
significantly higher voltages, the efficiency would drop significantly. The
switching transformer in particular is designed for exact voltages, and
without modification of the secondaries, increased voltage probably isn't
possible.
Look at the article at http://www.pavouk.comp.cz/hw/en_atxps.html for a
typical schematic and theory.

Cheers!!!!
--
Dave M
MasonDG44 at comcast dot net (Just subsitute the appropriate characters in
the address)

Never take a laxative and a sleeping pill at the same time!!

[Do Little2]

<kellro...@billburg.com> wrote in message
news:1127695292.7...@z14g2000cwz.googlegroups.com...

Obviously, a pc power supply is not a linear power supply
that one can adjust to any desired voltage. But there is a
way around it.

Suggestion: To increase the voltage, adjust it below the
14.9 Volts shut-off and then add your own windings in
series with the 14.9 Volts windings to get a separate
output voltage. Then use a switch to pre-select the amount
of windings and that in turn will determine your preset outputs.

[James Sweet]

"DaveM" <mason...@comcast99.net> wrote in message
news:upqdnSV5KeB...@comcast.com...

Silicon Chip had an article a while back where they did just that and were
getting 200V+ out to drive a vacuum tube device. With such high frequencies
you don't need many turns at all on the secondary, an extra turn or two will
probably push it's capability easily past 14v.

[kellro...@billburg.com]

> >>> > Is there some way I can disable the overvoltage protection on these
> >>> > power supplies?
> >>> >

>> > Michael

> >
>> Look at the article at http://www.pavouk.comp.cz/hw/en_atxps.html for a
> typical schematic and theory.
>

Thank you for the link.

[Franc Zabkar]

On 25 Sep 2005 15:11:35 -0700, kellro...@billburg.com put finger to
keyboard and composed:

>I have a salvaged pc power supply I use as an adjustable bench supply.
>But it won't go above about 14.9 volts. Apparently it has an
>overvoltage protection circuit.

IME, the overvoltage detection threshold is determined by a zener
diode.

>The board has a TL494 (house-numbered
>equivalent), and the only other chip on the board is a dual comparator.
> The power supply is a Fortron/Source FS-250S40G.
>I have another supply that also hits a ceiling when I adjust the
>voltage, a Falcon Technology PE-250X with a 494 chip and a 14-pin chip
>labeled DBL 339.

That's probably an LM339 quad comparator.

>Is there some way I can disable the overvoltage protection on these
>power supplies?

This should help you:
http://www.pavouk.comp.cz/hw/en_atxps.html

See the OV circuit at the LHS. Note the typo - one of the +5V inputs
is actually +12V.

-- Franc Zabkar

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

[Chris Jones]

kellro...@billburg.com wrote:

I have done this recently. There was an 8 pin dual comparator chip, and one
side was used to generate the 'power good' signal, the other side was used
for the overvoltage shutdown. I think the chip you have with the 339 in
the part number is probably a clone of the LM339, so download that
datasheet. The overvoltage comparator had a diode from the output pin to
the non-inverting input so that once triggered it would stay latched
shut-off. It also had a diode from the +5V power output terminal to [many
other things and] a resistor to the non-inverting input, which would cause
the thing to trip when the diode turned on. I put a few more 1N4148 diodes
in series with the original diode to raise the threshold as necessary,
whilst still preserving some overvoltage protection.

Trace the circuit diagram of your supply, it is worth it, and it won't take
long if you concentrate on the interesting part of the board. Use a
waterproof overhead marker to draw in the components on the solder side of
the board if this helps. Don't worry about the component values at first.

Good luck,

Chris

[Asimov]

"kellro...@billburg.com" bravely wrote to "All" (25 Sep 05 15:11:35)
--- on the heady topic of "modding a pc power supply"

ke> From: kellro...@billburg.com
ke> Xref: core-easynews sci.electronics.repair:343101

ke> I have a salvaged pc power supply I use as an adjustable bench supply.
ke> But it won't go above about 14.9 volts. Apparently it has an
ke> overvoltage protection circuit. The board has a TL494 (house-numbered
ke> equivalent), and the only other chip on the board is a dual
ke> comparator. The power supply is a Fortron/Source FS-250S40G.
ke> I have another supply that also hits a ceiling when I adjust the
ke> voltage, a Falcon Technology PE-250X with a 494 chip and a 14-pin chip
ke> labeled DBL 339.

ke> Is there some way I can disable the overvoltage protection on these
ke> power supplies?

Sure look for a comparator IC something like a 339 or something and
ground the input from the resistors sensing each supply voltage.
Some better supplies have a crowbar detector sensing the 5 volt line
and are designed to self-destruct if there is an overvoltage.

Beware however NOT to exceed the voltage rating of the filter
capacitors as these will readily explode if used outside their specs.

Accordingly, the 12 volt supply might only be good for 16 volts tops
but remember corresponginly less current as well. If rated 12V @ 8
amps then it can only supply 6 amps at 16 volts. Otherwise fizzzzt!

A*s*i*m*o*v

... The current limits placed are based on resistance