Larry wrote:
"Eisboch" wrote in
news

the impedance of the switching power supply appears as a
dead short to ground, tripping the GFI type breaker. There's

nothing

wrong with the circuit, it's just a characteristic of the

supply type.




The input of any switching power supply is merely a full wave
bridge rectifier and some LARGE electrolytic capacitors, which
are what causes the huge surge when you plug them in...charging
those caps.

They use large caps for a reason.....to reduce the effects of
powerline pulses, especially OFF pulses that blink your lights.
The large caps can hold up the output DC for several hundred
milliseconds during those brief power "blinks".

Unfortunately, the idiots are trying to see how few parts they
can make them out of so leave out any surge-reducing varistors or
even low value surge resistors from the primary circuit that
would stop the huge pulse and plug-in-arcing.

Larry


Wouldn't adding something like a starting capacitor fix that?

cavelamb himself wrote in
:

Wouldn't adding something like a starting capacitor fix that?



No, the circuit is just too simple...4 diodes in a bridge, caps
across the DC output of the rectifiers about 160VDC to power the
power MOSFET switching transistors running at 100 Khz or so. The
custom IC made to run all this measures the output DC volts and
varies the pulse width fed to the power transistors to regulate
the voltage. There's nothing analog about it. The MOSFETS are
either OFF, no heat generated or SATURATED, very little heat
generated because they are nearly a short at saturation...full
on. The DC output comes from another set of higher frequency
rectifiers hooked to a high frequency toroid transformer,
sometimes with more than one secondary winding like the DC power
supply in your desktop computer to get different voltages and
polarities....

When the load increases and output voltage TRIES to drop, the IC
senses this very fast and widens the pulse with to the switching
transistors, pulling the voltage back up with more power to the
toroid transformer. If the line voltage changes, the output also
tries to change, causing the sensing of the IC to vary the pulse
width in the appropriate direction to compensate. Some switching
power supplies, like those very light wall bricks that run and
charge your cellphones, pocket PCs, and laptop computers will
tolerate a voltage change so huge you don't even have to worry
about what line voltage you're plugging it into from 80VAC to
280VAC! If you plug it into 115 in USA, the pulse width is wider
than if you plug it into 240 in Europe. It doesn't care what
frequency because we're just going to directly rectify AC into DC
and feed it to the big input caps to store for the switchers.

What's hilarious in all this is someone with a very wide input
Switching powersupply plugging it into some kind of "SURGE
PROTECTOR" to keep any power line surges from "damaging" it. The
switcher could care less! If a big pulse of voltage comes into
it, the huge input caps just absorb it, turning the pulse into
more power for the switchers to use. Hell, if you surge it at
300V that's well within "normal" peak voltage on the 240VAC
system it was also designed for! The surge protector IS the
power supply itself! It's output is steady as a rock until the
big input filter caps drop below about 70VDC...on power down.

Switchers are neat.

Larry
--
Merry Christmas!
http://youtube.com/watch?v=Qi_NhFS4xEE

On Sun, 9 Dec 2007 22:26:58 -0500, "Ernest Scribbler"
wrote:

"Richard Casady" wrote
Mine is a digital with a ten amp fuse on the Amp function. Fluke 87 as
a matter of fact. All the professional heating and cooling guys, and
the appliance repairman, seem to have one like it.

That's well and good, but an ammeter with a 10A fuse may not be very helpful
in a circuit that's tripping a 15A breaker. I think I'd start by removing
all the power and using the 87's ohm meter function.

A good digital you can plug it into a wall socket with it on Ohms
without damage.

Casady

On Tue, 11 Dec 2007 14:59:12 -0800 (PST), terry
wrote:

On Dec 9, 8:16 pm, (Richard Casady) wrote:
On Sat, 08 Dec 2007 17:19:29 -0800, Peter Bennett

wrote:
More details would be useful. As it stands, your description doesn't
make sense to me, as the common North American 30 amp shore power
service _is_ 110 volts. Do you mean that the boat has a 30 amp power
inlet and shore power cord, but the dock only has a household-style 15
amp/110V outlet?

There is also the twenty amp outlet. The plug's blades are at ninety
degrees. Only such plugs I have ever seen were for window type air
conditioners. They make the outlets with the 'T' shaped holes to take
both those and the fifteen amp ones with the parallel blades.

Casady

Here in Canada we often use those 'horizontal pin' plugs for 230
volts.
In our garage, for example, we have 115 volt vertical pin outlets and
a 230 volt so we can plug in both types of tools. Both are breakered
at 15 amps.
On a workshop workbench we also have both.

Come to think of it, the 20 Amp ones have one blade each way, and the
two blade horizontal are 220 in the US as well. I think. Been a while
since I last wired any outlets, but think that I now have it right.

Casady

"Richard Casady" wrote
an ammeter with a 10A fuse may not be very helpful
in a circuit that's tripping a 15A breaker

A good digital you can plug it into a wall socket with it on Ohms
without damage.

I'll keep that in mind in case I ever want to try it.

My point is that, at best, a current measurement in this case is only going
to tell you what you already know.

On Wed, 19 Dec 2007 11:47:59 -0500, "Ernest Scribbler"
wrote:

"Richard Casady" wrote
an ammeter with a 10A fuse may not be very helpful
in a circuit that's tripping a 15A breaker

A good digital you can plug it into a wall socket with it on Ohms
without damage.

I'll keep that in mind in case I ever want to try it.

It isn't something you try. It's always carelessness. It's nice that
it survives a dumb move

My point is that, at best, a current measurement in this case is only going
to tell you what you already know.

That is true. At one time you could buy a shunt to extend the range on
a meter. I bet they still sell them.

Casady

On Wed, 19 Dec 2007 17:02:24 GMT, (Richard
Casady) wrote:

That is true. At one time you could buy a shunt to extend the range on
a meter. I bet they still sell them.

Current shunts are widely available from a number of sources:

http://www.google.com/search?hl=en&q=ac+current+shunts

Or you can make your own:

http://www.uoguelph.ca/~antoon/gadge...ts/shunts.html