"Dummy Load" for Power Supply Testing

[N3JBH]

Tom:  I just sent you an email with a bitmap of my schematic.  If it doesn't work, I will drop a hard copy in the mail.  Let me know if its readable for you to use.

Ken WB6MMV

 Yes if you really dont mind may i please have a copy too? (mycall)@yahoo.com   And thank you very kindly in advance jeff.

[PE1HZG]

As to the transistor dummy load, keep in mind that the Hfe (the amplification factor) of a transistor raises when the temperature goes up. That is, assume a transistor with a Hfe of 30. Give a base current of 100 mA and it will draw 3A (transistors are *current* sources, not *voltage* source).

However, as the temperature goes up, Hfe raises to 35. Hence, more collector current and more dissipation, more dissipation is more heat, more heat is more Hfe... thermal runaway.

The easy way is to keep an eye on the meter and adjust the base current. The advanced way is to use a control circuit. Me, being a lazy bum I do keep an eye on the meter. YMMV.

73,
Geert Jan PE1HZG

[WB6DGN]

As to the transistor dummy load, keep in mind that the Hfe (the amplification factor) of a transistor raises when the temperature goes up.

Well, that rained on my parade a bit (as well as embarrassed me for not remembering it myself)!  While the basic control circuit would be pretty simple (a negative feedback loop), getting it to track ten separate modules correctly and fairly accurately would be a horse of a different color, I suspect, and using a separate feedback loop for each module would increase the parts count and construction time prohibitively.  Which leads me back to the original proposal of using resistance wire (of SOME type) as the load element.

As I originally noted, the two types of resistance wire I have considered are Constantan or Manganin.  Problem is, I know very little about them and the help I've found on the internet is very limited.  Can anyone lead me in the right direction as far as information on these materials goes or suggestions for alternative materials?  All help is welcome and appreciated.
Tom DGN

[AD5X]

You might be interested in the "High Current DC test fixture" in the "Articles" section of my website (www.ad5x.com).  It only goes to 24 amps, but it can certainly be extended.

Phil - AD5X

[WB6DGN]

Hi Phil,
Very nicely done!  Especially the packaging.  The approach is very similar to what I had originally planned except you used "real" resistors and I had planned to kluge my own out of resistance wire.  I had just assumed that resistors that size would be too costly, especially in the quantity that I needed, but I was wrong.  After looking over your project, I found nine 2-ohm, 120-watt resistors at a very reasonable price.  I can then use some smaller variable resistors with the slider adjust to "fine tune" the loads to exactly 10-amps at the test cables.  The resistors for the 5-amp load and the four 1-amp loads are also fairly easy to find and not very expensive.  That will give me 1 to 99 amps in 1-amp steps.  As for cooling, I think I am going to try an idea borrowed from when I used to fly.  The engine cowling in most light planes is sealed for leaks and the forward motion of the plane pressurizes the interior of the cowling to increase the amount of heat the air will carry out of the cowling through an exit port of proper size to maintain the pressure.  So, rather than pulling air through the case at 1 atm I am going to try the same approach.  That's going to be a lot of heat to get rid of when all those resistors are energized.  That was a nice touch to include an AC port to monitor ripple and other noise on the power supply.  I've noticed the ripple specs for most switching supplies are several orders of magnitude greater than most linear supplies and could probably bear watching.
Thanks very much for the information; your project is a great inspiration to help me with mine.
By the way, I think I've seen an approach to minimizing the runaway problem with transistor loads and it might be fun to tinker with that at a later time but I really like the stability and simplicity of the passive components.  If anybody's interested, I'll post my progress as I get time to work on it.
Tom DGN