Electronic Load

Thread Starter

paulktreg

Joined Jun 2, 2008
833
Hello

Can anyone point me in the direction of a circuit diagram for an electronic load? I have downloaded one from Elecktor but it is a bit more involved than I would like. (But if it's all I can find then it will suffice).

I want to load computer ATX power supply 12V lines mainly up to 20A. (If it works on the 3V3 and 5V0 lines then thats a bonus). I am currently doing this with a monster heatsink and aluminium clad power resistors but I would like to have more control. Something using MOSFETS or high power darlingtons would be ideal controlled by a potentiometer.

All input appreciated.

Regards

Paul
 

beenthere

Joined Apr 20, 2004
15,819
When you expect to dissipate power as waste heat, the heat sinks and even fans come along for the ride. Using resistors makes more sense. A FET or BJT used in the ohmic region will get just as hot as a resistor. 240 watts is always going to be 240 watts.
 

Thread Starter

paulktreg

Joined Jun 2, 2008
833
When you expect to dissipate power as waste heat, the heat sinks and even fans come along for the ride. Using resistors makes more sense. A FET or BJT used in the ohmic region will get just as hot as a resistor. 240 watts is always going to be 240 watts.
Yes I appreciate that but as I said in my posting I need more control.
 

studiot

Joined Nov 9, 2007
4,998
Which Elektor load are you referring to and what control do you need?

There have been several such articles. One, at least, was switched (pulsed) to lower the total power drawn and also to test under AC conditions.
 

thingmaker3

Joined May 16, 2005
5,083
If control is needed, how about PWM through the power resistor? Much easier to impliment than switches in a linear region, and much less rough on the switches.
 

Thread Starter

paulktreg

Joined Jun 2, 2008
833
Which Elektor load are you referring to and what control do you need?

There have been several such articles. One, at least, was switched (pulsed) to lower the total power drawn and also to test under AC conditions.
The one I have is dated 2/2002 and is titled "High Power Adjustable Load" by P. Hirschbrich. I would post a copy of the circuit diagram but as Elektor charge for these articles it doesn't seem quite right.

This particular load is 0 to 20A but 100W maximum which is a little low (12V at 8.3A) and means I will have to add further 2 parallel FETs to increase the load nearer 20A.

As I mentioned previously I want to load up the 12V line (might need four of theses to load 4 lines) on ATX power supplies from approximately 1A to 20A and anywhere inbetween. A simple potentiometer control would be fine with a future option of controlling it via a PC I/O card in the future.
 

studiot

Joined Nov 9, 2007
4,998
I haven't seen any articles at quite that level, however here is a short bibliography

Constant current load for psu and amp testing - Practical Eelectronics Oct 75 p825
Electronic load - Elektor june 90 p16
Electronic load for psu testing - Elektor Nov93 p5
Active load for testing battery chargers - Elektor march 97 p 66
Adjustable constant current source - Practical Electronics may 05 p322

I would suggest the nov93 article as far superior to the 2002 one.


Meanwhile attached is a photo of a pair of power resistors I knocked up many years ago as an impoverished student.
I took an old electric fire bar and unravelled the wire. After measuring the length of 10 ohms I set a line of small nails in the baseboard at the 1 ohm distance and wound the wire back and fore to produce the two stepped resistors you see.
It still works fine today - you can see the burning from running at red heat. The non linearity doesn't matter so long as you measure current and voltage at each step. Connections are made with flying leads and croc clips.

Croc clips v Alligator clips - now here's a subject to go at in that English thread.
 

Attachments

Wendy

Joined Mar 24, 2008
23,415


I've thought of building one of these myself, so I drew it. Haven't figured out the Darlington transistor though, it needs to be able to handle what you are talking about feeding it.

Note: Polarity is critical. I added a diode on the collector of the darlington, but remember it has to be able to handle the current too.

Heatsinking is a must, and a fan on the heatsink might not be a bad idea. The 0.1Ω resistor must be rated for the watts you will feed it, and can be substituted for higher values if it will help. If you intend to use it for measurement you can get 4 wire resistors for accuracy.
 
Last edited:

Thread Starter

paulktreg

Joined Jun 2, 2008
833
Bill_Marsden

I have seen the MJ11016 darlington suggested in other articles. It is still available and at 30A would probably do the job.

I have given the heatsink some thought, and being a PC enthusiast I have one or two intel stock core 2 duo processor heatsinks/fans doing nothing. If i can modify one of these and mount it to the transistor I think it would be the ideal way of getting rid of all them watts!

I may well just knockup the circuit bird nest style and see how it performs.

Many thanks for the help.

Regards

Paul
 

Wendy

Joined Mar 24, 2008
23,415
You might elimate one of the 1N4001's to make sure the circuit turns off entirely. It could draw a couple of amps on the minimum setting otherwise. Let me know how it goes, this could wind up as a projects entry, and ultimately in the cookbook we are thinking of implementing on the site.
 

John Luciani

Joined Apr 3, 2007
475
Hello

Can anyone point me in the direction of a circuit diagram for an electronic load? I have downloaded one from Elecktor but it is a bit more involved than I would like. (But if it's all I can find then it will suffice).

I want to load computer ATX power supply 12V lines mainly up to 20A. (If it works on the 3V3 and 5V0 lines then thats a bonus). I am currently doing this with a monster heatsink and aluminium clad power resistors but I would like to have more control. Something using MOSFETS or high power darlingtons would be ideal controlled by a potentiometer.

All input appreciated.

Regards

Paul
I have built many electronic loads using a MOSFET in the feedback loop of
an opamp. For loads that need operate below 100V I would a MOSFET
rather than a BJT. The MOSFET will have a lower on-resistance which will
allow you to get to higher currents and possibly simulate a short-circuit.

I have a schematic and an example PCB layout at
http://www.luciani.org/geda/util/matrix/index.html

(* jcl *)
 
Top