According to Don Foreman :
On Thu, 16 Mar 2006 11:50:29 GMT, Ignoramus2206
wrote:
[ ... ]
The numbers cited in ebay auctions for new mastech supplies are very
impressive. Check this auction: 7599501997:
#
Two level of control for both current and voltage outputs: coarse and
fine for ease of use
# Adjustable outputs: 0-30V and 0-5A
# Input voltage: 110V AC and 220V AC switchable
# Line regulation: CV = 0.01% + 1 mV, CC = 0.2% + 1 mA
# Load Regulation: CV = 0.01% + 3mV, CC = 0.2% + 3 mA
# Ripple noise: CV = 0.5 mV RMS, CC = 3 mA RMS
# Protection: constant current and short-circuit protection
# LCD reading accuracy: +/-1% for voltage and +/-2% for current
# Environment: 0-40C, relative humidity 90%
# Size: 11.5" x 6.5" x 5.5"
# Weight: 13 lbs
I am not sure if I should believe these numbers, but as you said,
plating does not require that sort of precision.
It's quite easy to design a linearly-regulated supply with those
specs. You won't see that level of performance on the bench, though.
Example: just 300 micro-ohms of contact resistance at each connection
will produce more than 3 mV of drop at 5 amps. Banana jacks aren't
nearly that good. (Not that it matters.....)
Agreed ... not that it matters for *this* application.
However, there is another feature which could be a part of this
which would allow this king of regulation at the *load*, instead of at
the output terminals. That is -- if the supply has remote sense
terminals along with duplicate output terminals on the back panel. If
so, run heavy wires from the outputs to the load, and run a shielded
twisted pair from the remote sense terminals to the load, and bingo, the
supply adjusts for voltage losses in the output connections.
I've got some supplies by Power Designs, 0-20V, 0-500mA, which
have four switches in two concentric pairs which let me dial in voltage
in 1V steps, 0.1V steps, 0.01V steps, and 0.001V steps, plus a pot for
trimming between 0 mv and 1mv added. Also, there is a toggle switch to
add 10 V to the dialed in output voltage.
These things have an oven-stabiLized zener reference in them, so
they don't give the specified accuracy until they have been powered on
for a half hour or so. :-)
But really nice to be able to dial that accurate an output
voltage. And a lot easier to reach a selected voltage than twiddling
with a 10-turn pot. :-)
BTW I got two of them mounted to a shared rack mount panel at a
hamfest. The price was really cheap, because they "didn't work
right".
A glance at the back panel told me why. They had been set up to
use the remote sense and the rear panel outputs, and the wires
had simply been cut at the terminal strip. Without the remote
sense wires, there should have been jumpers connecting the sense
and the output terminals, so they were sensing locally.
They worked perfectly, once that minor factor had been dealt
with. :-)
Here is a slightly later design. Mine had standard face type
panel meters, not the edge-on style used in this, and the
location of the binding posts was different. Also, it seems to
have lost the convenient output short button for the current
limit setting. The light just above and to the left of the
current adjust pot comes on when it switches to constant current
mode.
http://www.sphere.bc.ca/test/power-s...s/pd-2005a.jpg
Of course -- I knew these from work, where I had used them, and
the larger 0-36V 0-5A ones with no such switches. That one had
a Variac inside connected to the same shaft as the voltage
selection dial, so there was not as much power being wasted in
the regulator circuitry when it was set for low voltage and high
current.
Both styles also had constant current output mode available.
And the smaller ones even had a button to short the output from
the front panel, so you could more easily set the current
limits.
Enjoy,
DoN.
--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. |
http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---