Hi folks,
I've sent out info to half a dozen folks, and the idea of having a load
test has come up more than once. Chris, W7JPG, was the most
inquisitive, I've cc'd his note below, and my thoughts on the whole thing.
-------------------
my reply:
Hi Chris,
You're right in everything you say. I had stated originally that
this unit did not test under load, and that I was thinking of adding
some load features. I did not build the unit, it came from a "Collins
Guy". I've been thinking about using fixed resistors (large wattage, as
you say), probably with a momentary connection, probably thru relays for
the HV, just push button switch directly for the bias (low load, maybe
20-30 ma.) and filaments (maybe 3-4 amps). Using, for example, 10 kohms
for the HV/LV load, you'd not have to measure current, it'd be the same,
in ma., as the indicated voltage divided by 10. Would not be full load,
but enough to indicate bad diodes in a doubler ckt, for example.
I'd probably use a 50 watt 10k resistor, and for intermittent short
duration use that should be OK. I did a little "surfing" ystdy, that
can be found in the $5-15 range (Mouser). A small relay could be used
to keep the HV off of a switch. Since the volt meter is switched
between HV & LV, one resistor could be used for both.
I just noticed, the orig. schematic doesn't show a fil. voltage mtr.
-------------------
his query:
Now, for my question. It looks like your test box measures all voltages
under "no load" conditions. Did I miss something?
My interest in the thread on "load bank" tests was the practical matter
of how someone would set up the proper circuit and components to
simulate the load presented to a 516F2 by, let's say, a KWM-2A. The
info I found on such a "load bank" for low voltage - high current
equipment required a lot of high wattage resistors to dissipate the heat
generated for each of the main circuits that would be under a load test:
heater, bias, B+ draw.
After looking at your fixture, I realized (correct me if I am off base
here) that to test a specific power supply (516F2, AC-4, etc.), one
would be able to make a comparison check of "no load" voltages against a
properly functioning example of these PSU's or a specification that
stated the "normal" "no load" voltage levels.
So, if a PSU was tested without a load and all the no load voltages were
nomimal, what advantage, if any, would a load test provide over a "no
load" test? :-0
I would appreciate any comments.
73,
Chris
W7JPG
-------------------------------
Any other thoughts from list members would be appreciated by all.
73,
Al, W8UT
www.boatanchors.org
www.hammarlund.info
"There is nothing -- absolutely nothing -- half so much
worth doing as simply messing about in boats"
Ratty, to Mole
====================================================
On 11/27/2012 1:47 PM, Al Parker wrote:
Hi folks,
I have scanned the article, one small page, written by K0ZQD in V2
Iss. 16 of The Collins Coll's. MMag. It gives the schematic, leaving
construction details to the builder. I also have scanned a skem of the
p.s. pinouts for the 516F-2, from the same issue. Make your own
adapter(s) for various supplies, if wanted. I'll send the scans to
those who have/do request it.
Mike brings up a good point, as written, the voltages are unloaded,
and will be high. I'd like to switch in the load, so a relay for the HV
would be advisable for that. As Mike says, use ohms law for resistor
sizing, and you'll know, close enuf, what the currents will be.
I've used my test rig on a lot of Collins and Drake supplies, it's
a quick test, doesn't show ripple, but ckg ESR on the filter caps will
tell you what you need to do there (do it before you power up the
supply, so the caps don't do any reforming).
Al, W8UT
On 11/27/2012 1:06 PM, Mike Bryce wrote:
There wouldn't be much too it.
all you need would be a hand full of resistors, using ohm's law to
calculate the values. The values would be calculated at the max
current the TRx series would draw.
On would need to load the 6.3 v filament down to the same current as
the tubes, if you don't the unloaded transformer would produce more
voltage than normal since it's not powering the tubes.
a couple of cheap LED digital panel mount meter could be used to
monitor the voltages, plate, screen, and bias
likewise a meter could be used to monitor current on the plate supply.
I don't feel it would be necessary to monitor bias current or screen
voltage current either.
it would be an interesting project, that would make checking the
supply quick. A few test points for an 'oscope to monitor ripple would
be icing on the cake
later
mike, wb8vge
On Mon, 26 Nov 2012 11:50:11 -0500
Al Parker <anc...@ec.rr.com> wrote:
hi folks,
Some yrs ago there was an article on building a test
setup for Collins p.s.'s. I think it was in one of the
Collins collector magazines, I can search it out. I came
into possession of one of the units, and have made cable
adapters for the AC-3/4's, and I think the HP-23's also.
It is for voltage only, and did not include a load test,
but I have been thinking of adding that to mine.
73,
Al, W8UT
_______________________________________________
Drakelist mailing list
Drakelist@zerobeat.net
http://mailman.zerobeat.net/mailman/listinfo/drakelist
