"Larry Jaques" wrote in message
...
On Sat, 14 Sep 2013 11:53:48 -0400, "Jim Wilkins"
wrote:

The replacement 10A diode is larger and doesn't fit under the
heatsink
clamp. I hung it off the side of the board.

...which resulted in better cooling. Win/win, wot?

I hope so. The physically larger 10A diode seemed OK at the panels'
max current. I raised the current until the diode got hot enough that
the forward drop crept downward.

I tested the inverter's low voltage behavior with a 30V 5A lab
supply
like this, bought locally:
http://smart-prototyping.com/index.p...product_id=177

Sweet! I've never seemed to have a regular need for a variable
supply
so I've never bought one. I usually rely on wall warts.

Mine have paid for themselves by revivifying dead power tool
batteries.
http://www.instructables.com/id/Brin...-Back-To-Life/
It's like voodoo, you have to know the ritual and make a sacrifice,
and you get at best a short-lived zombie rather than a healthy
subject. The safest way is to make a condemned prisoner do it while
you watch through binoculars. The usual procedure is to assign your
obedient lab tech Jim to do it while you take a long lunch.

When I was learning the intricacies of op-amp measurement circuits OJT
I realized that voltage can just sit there but current HAS to be
accounted for, and began thinking in terms of it, and how to control
and measure it. External meters don't work as well as a power supply
that shows current directly without a voltage drop across the meter
and jumper wires to consider. The big advantage of a lab supply is its
adjustable current limit which protects the load, and sometimes the
user.

These are fantastic if someone else pays for them:
http://www.home.agilent.com/en/pc-85...?&cc=US&lc=eng

and this has the capability of the 4-quadrant V-I supply that is the
heart of an automated analog semiconductor tester:
http://www.keithley.com/products/dca...rpose/?mn=2400
http://en.wikipedia.org/wiki/Automatic_test_equipment

4-quadrant means it can source or sink current in any combination of +
or -, like big op amps. Regulated power supplies may not take well to
a higher external voltage source forcing current back in, like if the
line power fails while charging a battery. I needed the extra
Schottkys to replace Si diodes that I attached to alligator clips and
connect in series with batteries I'm charging.

Cheap import lab supplies are good enough for my home experimenting,
such as measuring the resistance of a motor or antenna by forcing 1
Amp through it and reading milliOhms as milliVolts. Even a $4 HF DVM
will resolve tenths of a milliOhm that way. The power supply's output
voltage can be turned down very low so an open shouldn't damage the
circuit being checked, for example the wiring harness of an old
vehicle.

I use my homebrew Variac power supply (0-40V, 20A)
Big honker, ain't it?

It won't do both at once. I was looking for parts at a flea market to
complete a less ambitious one when I found the salvaged front panel,
with a 10A Variac and 40V and 20A meters, so I used the other
components I had already collected to resurrect it. It's not regulated
(yet) but good enough to quickly charge a dead battery or bench-test
heater and wiper motors that need more than the 5A limit of my lab
supplies. I don't like to connect sickly, sluggish DC motors directly
to a battery. It's safer to turn up the voltage on them slowly while
watching the current.

I've measured the DC current or AC power most of my antique
electronics consume and written it on them for when they need work
later. Yesterday I extended the wiring to my 1950's TV antenna rotator
and then watched the power the controller drew when turned back on to
confirm I hadn't miswired it. Now I can unplug it where it comes into
the basement instead of going out in the rain when lightning is
predicted.
jsw