I have a little circuit board, salvaged from a washing machine, that is
a nice multi-output switcher. I'd like to use it in a project, but
there's one problem: all the output voltages are negative.

I measured the voltages and traced the circuit. The transformer has four
secondary pins; three have diode rectifiers, and one does not; it goes
straight to what I think is the "ground" output. All the filter caps for
all three outputs have their positive pins connected to that lead too.

The three diodes point in the direction that makes all the outputs
negative, too (cathodes to the transformer).

The outputs are -5, -12, and -22 volts; very reasonable for a small
supply, except that they are *backwards*.

So my question is, can I just reverse the diodes and capacitors, and
swap the two leads that send one of the output voltages to the reference
used to control the optically isolated feedback, and have a nice supply
with *positive* outputs?

Or is there something about the waveform from the transformer -- or
something else -- that would prevent this from working?

Isaac

On 7/15/2013 11:19 PM, isw wrote:
I have a little circuit board, salvaged from a washing machine, that is
a nice multi-output switcher. I'd like to use it in a project, but
there's one problem: all the output voltages are negative.

I measured the voltages and traced the circuit. The transformer has four
secondary pins; three have diode rectifiers, and one does not; it goes
straight to what I think is the "ground" output. All the filter caps for
all three outputs have their positive pins connected to that lead too.

The three diodes point in the direction that makes all the outputs
negative, too (cathodes to the transformer).

The outputs are -5, -12, and -22 volts; very reasonable for a small
supply, except that they are *backwards*.

So my question is, can I just reverse the diodes and capacitors, and
swap the two leads that send one of the output voltages to the reference
used to control the optically isolated feedback, and have a nice supply
with *positive* outputs?

Or is there something about the waveform from the transformer -- or
something else -- that would prevent this from working?

Isaac


Too many variables.
Publish the schematic somewhere.

"isw" schreef in bericht
]...
I have a little circuit board, salvaged from a washing machine, that is
a nice multi-output switcher. I'd like to use it in a project, but
there's one problem: all the output voltages are negative.

I measured the voltages and traced the circuit. The transformer has four
secondary pins; three have diode rectifiers, and one does not; it goes
straight to what I think is the "ground" output. All the filter caps for
all three outputs have their positive pins connected to that lead too.

The three diodes point in the direction that makes all the outputs
negative, too (cathodes to the transformer).

The outputs are -5, -12, and -22 volts; very reasonable for a small
supply, except that they are *backwards*.

So my question is, can I just reverse the diodes and capacitors, and
swap the two leads that send one of the output voltages to the reference
used to control the optically isolated feedback, and have a nice supply
with *positive* outputs?

Or is there something about the waveform from the transformer -- or
something else -- that would prevent this from working?

Isaac

Most likely not. Fastest way to make sure is looking at the waveforms with
an o'scope. But even if the waveforms look like good enough and you want to
give it a try, you'll have to reverse elco's and maybe other components as
well. It's almost sure that there is a kind of feedback from the secondary
to the primary that also has to be modified too.

petrus bitbyter

On 07/16/2013 07:33 AM, petrus bitbyter wrote:
"isw" schreef in bericht
]...
I have a little circuit board, salvaged from a washing machine, that is
a nice multi-output switcher. I'd like to use it in a project, but
there's one problem: all the output voltages are negative.

I measured the voltages and traced the circuit. The transformer has four
secondary pins; three have diode rectifiers, and one does not; it goes
straight to what I think is the "ground" output. All the filter caps for
all three outputs have their positive pins connected to that lead too.

The three diodes point in the direction that makes all the outputs
negative, too (cathodes to the transformer).

The outputs are -5, -12, and -22 volts; very reasonable for a small
supply, except that they are *backwards*.

So my question is, can I just reverse the diodes and capacitors, and
swap the two leads that send one of the output voltages to the reference
used to control the optically isolated feedback, and have a nice supply
with *positive* outputs?

Or is there something about the waveform from the transformer -- or
something else -- that would prevent this from working?

Isaac

Most likely not. Fastest way to make sure is looking at the waveforms with
an o'scope. But even if the waveforms look like good enough and you want to
give it a try, you'll have to reverse elco's and maybe other components as
well. It's almost sure that there is a kind of feedback from the secondary
to the primary that also has to be modified too.

petrus bitbyter



Just call "common" Plus [+]. The difference of potential is 12 Vdc, etc.
no matter what the label says. If you are worried, use big silicon
diodes to make sure nothing flows ass backwards.

On Tue, 16 Jul 2013, dave wrote:

On 07/16/2013 07:33 AM, petrus bitbyter wrote:
"isw" schreef in bericht
]...
I have a little circuit board, salvaged from a washing machine, that is
a nice multi-output switcher. I'd like to use it in a project, but
there's one problem: all the output voltages are negative.

I measured the voltages and traced the circuit. The transformer has four
secondary pins; three have diode rectifiers, and one does not; it goes
straight to what I think is the "ground" output. All the filter caps for
all three outputs have their positive pins connected to that lead too.

The three diodes point in the direction that makes all the outputs
negative, too (cathodes to the transformer).

The outputs are -5, -12, and -22 volts; very reasonable for a small
supply, except that they are *backwards*.

So my question is, can I just reverse the diodes and capacitors, and
swap the two leads that send one of the output voltages to the reference
used to control the optically isolated feedback, and have a nice supply
with *positive* outputs?

Or is there something about the waveform from the transformer -- or
something else -- that would prevent this from working?

Isaac

Most likely not. Fastest way to make sure is looking at the waveforms with
an o'scope. But even if the waveforms look like good enough and you want to
give it a try, you'll have to reverse elco's and maybe other components as
well. It's almost sure that there is a kind of feedback from the secondary
to the primary that also has to be modified too.

petrus bitbyter



Just call "common" Plus [+]. The difference of potential is 12 Vdc, etc. no
matter what the label says. If you are worried, use big silicon diodes to
make sure nothing flows ass backwards.

It's odd that it's all negative voltage out. I'd just poke around until I
found another power supply, they are endless. Not just as "ac adapters"
but in inkjet printers (well if the supply is internal) though they tend
to have higher voltages, sometimes a good thing. DVD players (I'm
starting to see those in the garbage), DVD recorders (I only found one of
those) even that blu-ray player I found 2 weeks ago (it works, no remote
though, so I'm not stealing the power supply from it). VCRs for a period
had the switching supply (and often odd sets of voltages) separate from
the main board, though the more
recent and cheaper ones have it all on one board (which migh be extracted
by careful cutting but since there are lots of sources of power supplies,
I'd skip VCRs).

If this supply was only needed for one voltage, yes, I'd just treat ground
as positive, making sure the ground of the power supply doesn't get
grounded to the whatever being powered.

Reversing polarity by reversing diodes and electrolytics has the problem
that one of those outputs (and maybe more, though not likely) is being
esnored to provide feedback, so trying to change polarity means having to
rework that feedback circuit. Likely the main output, the one with the
most current, has the feedback, the other ones with no feedback probably
could be repolarized by flipping the diode and electrolytic, but they may
not supply that much current.

Michael

On 07/16/2013 10:15 AM, Michael Black wrote:
On Tue, 16 Jul 2013, dave wrote:

On 07/16/2013 07:33 AM, petrus bitbyter wrote:
"isw" schreef in bericht
]...
I have a little circuit board, salvaged from a washing machine, that is
a nice multi-output switcher. I'd like to use it in a project, but
there's one problem: all the output voltages are negative.

I measured the voltages and traced the circuit. The transformer has
four
secondary pins; three have diode rectifiers, and one does not; it goes
straight to what I think is the "ground" output. All the filter caps
for
all three outputs have their positive pins connected to that lead too.

The three diodes point in the direction that makes all the outputs
negative, too (cathodes to the transformer).

The outputs are -5, -12, and -22 volts; very reasonable for a small
supply, except that they are *backwards*.

So my question is, can I just reverse the diodes and capacitors, and
swap the two leads that send one of the output voltages to the
reference
used to control the optically isolated feedback, and have a nice supply
with *positive* outputs?

Or is there something about the waveform from the transformer -- or
something else -- that would prevent this from working?

Isaac

Most likely not. Fastest way to make sure is looking at the waveforms
with
an o'scope. But even if the waveforms look like good enough and you
want to
give it a try, you'll have to reverse elco's and maybe other
components as
well. It's almost sure that there is a kind of feedback from the
secondary
to the primary that also has to be modified too.

petrus bitbyter



Just call "common" Plus [+]. The difference of potential is 12 Vdc,
etc. no matter what the label says. If you are worried, use big
silicon diodes to make sure nothing flows ass backwards.

It's odd that it's all negative voltage out. I'd just poke around until
I found another power supply, they are endless. Not just as "ac
adapters" but in inkjet printers (well if the supply is internal) though
they tend to have higher voltages, sometimes a good thing. DVD players
(I'm starting to see those in the garbage), DVD recorders (I only found
one of those) even that blu-ray player I found 2 weeks ago (it works, no
remote though, so I'm not stealing the power supply from it). VCRs for
a period had the switching supply (and often odd sets of voltages)
separate from the main board, though the more recent and cheaper ones
have it all on one board (which migh be extracted by careful cutting but
since there are lots of sources of power supplies, I'd skip VCRs).

If this supply was only needed for one voltage, yes, I'd just treat
ground as positive, making sure the ground of the power supply doesn't
get grounded to the whatever being powered.

Reversing polarity by reversing diodes and electrolytics has the problem
that one of those outputs (and maybe more, though not likely) is being
esnored to provide feedback, so trying to change polarity means having
to rework that feedback circuit. Likely the main output, the one with
the most current, has the feedback, the other ones with no feedback
probably could be repolarized by flipping the diode and electrolytic,
but they may not supply that much current.

Michael


Consumer grade UPS units have 120Vac to 12Vdc outputs to charge the gel
cells + 12V to 120Vac inverters. Never throw those away. You can use
them for poor man's solar.

On 07/16/2013 03:16 PM, dave wrote:
On 07/16/2013 10:15 AM, Michael Black wrote:
On Tue, 16 Jul 2013, dave wrote:

On 07/16/2013 07:33 AM, petrus bitbyter wrote:
"isw" schreef in bericht
]...
I have a little circuit board, salvaged from a washing machine,
that is
a nice multi-output switcher. I'd like to use it in a project, but
there's one problem: all the output voltages are negative.

I measured the voltages and traced the circuit. The transformer has
four
secondary pins; three have diode rectifiers, and one does not; it goes
straight to what I think is the "ground" output. All the filter caps
for
all three outputs have their positive pins connected to that lead too.

The three diodes point in the direction that makes all the outputs
negative, too (cathodes to the transformer).

The outputs are -5, -12, and -22 volts; very reasonable for a small
supply, except that they are *backwards*.

So my question is, can I just reverse the diodes and capacitors, and
swap the two leads that send one of the output voltages to the
reference
used to control the optically isolated feedback, and have a nice
supply
with *positive* outputs?

Or is there something about the waveform from the transformer -- or
something else -- that would prevent this from working?

Isaac

Most likely not. Fastest way to make sure is looking at the waveforms
with
an o'scope. But even if the waveforms look like good enough and you
want to
give it a try, you'll have to reverse elco's and maybe other
components as
well. It's almost sure that there is a kind of feedback from the
secondary
to the primary that also has to be modified too.

petrus bitbyter



Just call "common" Plus [+]. The difference of potential is 12 Vdc,
etc. no matter what the label says. If you are worried, use big
silicon diodes to make sure nothing flows ass backwards.

It's odd that it's all negative voltage out. I'd just poke around until
I found another power supply, they are endless. Not just as "ac
adapters" but in inkjet printers (well if the supply is internal) though
they tend to have higher voltages, sometimes a good thing. DVD players
(I'm starting to see those in the garbage), DVD recorders (I only found
one of those) even that blu-ray player I found 2 weeks ago (it works, no
remote though, so I'm not stealing the power supply from it). VCRs for
a period had the switching supply (and often odd sets of voltages)
separate from the main board, though the more recent and cheaper ones
have it all on one board (which migh be extracted by careful cutting but
since there are lots of sources of power supplies, I'd skip VCRs).

If this supply was only needed for one voltage, yes, I'd just treat
ground as positive, making sure the ground of the power supply doesn't
get grounded to the whatever being powered.

Reversing polarity by reversing diodes and electrolytics has the problem
that one of those outputs (and maybe more, though not likely) is being
esnored to provide feedback, so trying to change polarity means having
to rework that feedback circuit. Likely the main output, the one with
the most current, has the feedback, the other ones with no feedback
probably could be repolarized by flipping the diode and electrolytic,
but they may not supply that much current.

Michael


Consumer grade UPS units have 120Vac to 12Vdc outputs to charge the gel
cells + 12V to 120Vac inverters. Never throw those away. You can use
them for poor man's solar.

Majorly poor. Even with no load, a 750 VA UPS will discharge its
batteries completely in under an hour.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

On Tue, 16 Jul 2013 12:16:46 -0700, dave wrote:


Consumer grade UPS units have 120Vac to 12Vdc outputs to charge the gel
cells + 12V to 120Vac inverters. Never throw those away. You can use
them for poor man's solar.

Just don't tie it to thte grid, that would be illegal.

?-)