Power MOSFET as a voltage controlled resistor

atlbraves49

We need to have a load resistance to match an internal resistance for peak power.. so we've decided to use a Power MOSFET as the varying load resistance, but my question is.. when I'm looking for which power MOSFET to buy, to use, what should I be looking for in terms of properties?

This is what we need:

The Power MOSFET....
1) Has to be able to handle the ~25W of power that will be fed into it
2) The resistance has to vary between a couple ohms and 25 ohms.

Any ideas?

Pumblechook

BUZ and IRF series.


BUZ11A 90 Watt device. (with suitable heat sink).

IRF540 100 Watts 22 Amp.



IRF640 125 W 200 v.

All quite cheap.

atlbraves49

and they would all be able to vary in resistance from 1ohm to 25 ohm, depending on the "voltage control" ?

Pumblechook

I think all the BUZ and IRF types go below 1 ohm and the max resistance is pretty high. All enhancement mode I think. Need a postive voltage between gate and source to turn them on (low R).

To prevent the resistance from ever going too low you could put a low valve high power resistor in series.. 0.5 Ohm say.. In the source supply and apply the control voltage between the power supply side of the resistor and the gate (this will stabilise the current through the Mosfet to some extent).


http://www.datasheetcatalog.org/data...IRF640_S_1.pdf

atlbraves49

thanks for the help; one more question, what should i be looking for on datasheets that will tell me it can reach 1-25 ohms? On the datasheet you posted, the only thing I see related to resistance is "Drain-Source(ON) resistance" of hundreds of mOhms (which of course is too small), so what am I missing.. in terms of knowing whether a power mosfet can reach 25 ohms?

Pumblechook

With no bias voltage on the gate the MOSFET will be more or less an open circuit falling to near zero with a certain positive voltage on the gate so will cover the range you want. You will have to experiment and see what bias voltage is required. As I say if you put a fixed resistor in series of 0.5 or 0.75 Ohm it will ensure you can never go below those values even if the MOSFET is switched hard on.

I will try and draw a diagram.

Pumblechook

Select R2 to give the max require resistance. Select on test. I would guess 50K or more.

DIAGRAM is WRONG. R2 to be below the 10 K potentiometer. In fact you could have resistors top and bottom to get your required range.

I cant guarantee this is all correct. You need to experiment.

Attached Thumbnails

 

Pumblechook

Select R2 to give the max require resistance. Select on test. I would guess 50K or more.

DIAGRAM is wrong. R2 to be below the 10 K potentiometer. In fact you could have resistors top and bottom to get your required range.

I cant guarantee this is all correct. You need to experiment.

uart

Hi atlbraves. Just wondering, are you trying to use this power for any purpose othe than creating heat in the mosfet?

atlbraves49

Yep, it's for impedance matching (for peak power).

We have a thermoelectric module, generating power, with some internal resistance, and we want a load that will automatically match that impedance to get peak power.

uart

So you've got a thermo-electric module converting heat to electric power and then you want to convert the electric power back into heat in the mosfet. Is this just a measurement exercise or are you trying to do something else. Are you sure you're not really looking for a DC-DC converter to match the thermocouples output characteistics and produce electric power rather than heat?

atlbraves49

yes this is completely a testing / measurement exercise, we are NOT interested at all in utilizing the power that is being created by the thermoelectrics. The setup is for testing various conditions related to thermoelectrics, so what happens to the power is irrelevant. Right now it is being dumped across resistors that are manually matched (with rheostats) to the internal impedance. We are looking to automate that "impedance matching" so we dont have to sit there and tune rheostats.

Corneo

I was thinking of using an op amp to drive the gate of the MOSFET. Use the voltage across the internal resistance as the inputs. This way, the op amp will automatically adjust the gate of the FET to such a level that you have the best matching. Just an idea.

Gordona

I know this was a long time ago but I have a similar need for a variable resistance (0-240 Ohms) at 12 volts, which can be controlled with a varying voltage from a micro. I looked at the spec sheet for the above devices and it's not clear how to read the salient specs. Any help would be appreciated.

Mike_In_Plano

You can purchase a dynamic load, but that costs money...

Or, you can try this one :)

Most any op amp will do including 741 or TL061, T071, TL081.

R1 needs to be BIG, i.e 10 watt and M1 needs to be fastened to a fairly big heat sink. The little heat sinks that hang off the part are about 1/20th what you'll need. A sheet of 1/8" 6" square aluminum should work nice. Or, even better, you can get a big computer heat sink with integral fan.

Best Luck,

Mike

Attached Thumbnails

 

Gordona

Mike;

Thanks for the input.