4A Active Load

[janengelbrecht]

This 4A (max 30V) active load is made with Daves in mind.
I have shown a 2A version here before - but it really didnt live up to my needs.
So here is my 4A version:  

Question is : 2 Logic ICs....would you use a cheap Microcontroller instead ? 
(Then you could use something else than RV1 and RV2 to make the control voltage : PWM and simpel RC filter. And the user could use a cheap Rotating encoder to dial it in. Instead of 2 voltmeters a cheap 2x16 character LCD display could be used!)

The Active Load has only been simulated in Labcenter Proteus ISIS 7.....so maybe it will blow up in my face when i build it ?  

[void_error]

Question is : 2 Logic ICs....would you use a cheap Microcontroller instead ? 
(Then you could use something else than RV1 and RV2 to make the control voltage : PWM and simpel RC filter. And the user could use a cheap Rotating encoder to dial it in. Instead of 2 voltmeters a cheap 2x16 character LCD display could be used!)

I'd use a cheap micro with an LCD. That way it could also calculate and display power and handle the over temperature protection and save a lot of board space. You could also program it to test load regulation of power supplies by varying the output current between two (or more) preset values although for that you might want to use a DAC instead of PWM because of the delay the RC filter will add.
I'd also recommend adding at least two more MOSFETs for higher reliability and use TO-247 devices for their lower junction-case thermal resistance.

I'm planning to design/build a DC load myself after I finish the analog part of my lab power supply.

[Seekonk]

Still think your RV1 RV2 arrangement is weird.  I would add a half ohm source resistor to each FET to better balance.   4A X 30V is 120W.  While it is likely you wouldn't be in that situation often, it could happen.  Anytime you dissipate 25W in a FET you are asking for trouble due to package limitations.  I would at least add one more FET.

[janengelbrecht]

True that....better be safe than sorry.... Thx
RV1 and RV2 arrangement weird ? Well it works I will not label it COARSE / FINE though....one just has to fiddle around and look at the meter to see when the current needed is present
Think i will build two versions: analogue and uC version too But yes...4 MOSFETS it is...with small source resistors too......thx guys

(But i will never bee able to test it further than 3A....hasnt got anything that could deliver more amps )

[void_error]

But yes...4 MOSFETS it is...with small source resistors too......thx guys

I was thinking into splitting it into 4 (or as many as you need) separate current sinks, each with its opamp and mosfet since opamps are rather cheap, that way the current balancing problem caused by the difference in threshold voltages is eliminated, the only causes of error left being resistor tolerance and opamp input offset voltage.
You can even hook up several loads to the same control circuitry that way, all of them using the same reference.

[janengelbrecht]

Point taken - will give it a try And spend some time figuring out someway to test it at 4 amps

[ManCave]

I was thinking into splitting it into 4 (or as many as you need) separate current sinks, each with its opamp and mosfet since opamps are rather cheap, that way the current balancing problem caused by the difference in threshold voltages is eliminated, the only causes of error left being resistor tolerance and opamp input offset voltage.
You can even hook up several loads to the same control circuitry that way, all of them using the same reference.

+1!

otherwise you might find (as I did) that you'll have one or two  mosfets hogging the current and hence overheating. No two mosfets are the same. And if they are not the same, they will not pass the same current through them with the same voltage on the gate. The effect will then snowball with increase in temperature there will be other contributors taking effect. I would definitely use separate opamp + source resistor per MOSFET!

[janengelbrecht]

I will surely try it out... Big thx from here to you guys for your inputs

[void_error]

Point taken - will give it a try And spend some time figuring out someway to test it at 4 amps

Lash up a step-down converter maybe? For example something that puts out 5V @ 4-5A - that's just 20-25W. MC34063 + external PNP + inductor + diode + caps and a bunch of resistors - the simplest and chepest way to get 4A out of something that puts out like 12-36V at less than 4A. Or a good PC PSU with independent regulation for each output voltage or a beefy laptop power pack. A car lead-acid battery charger with a cap on the output might also work, they can put out over 10A although some of them have more than just a transformer + rectifier. I've got a BOSCH one and it only has a 12V centre-tapped transformer with a full-wave rectifier followed by an ammeter. It can put out 10 amps of rectified sinewave.

[janengelbrecht]

Point taken - will give it a try And spend some time figuring out someway to test it at 4 amps

Lash up a step-down converter maybe? For example something that puts out 5V @ 4-5A - that's just 20-25W. MC34063 + external PNP + inductor + diode + caps and a bunch of resistors - the simplest and chepest way to get 4A out of something that puts out like 12-36V at less than 4A. Or a good PC PSU with independent regulation for each output voltage or a beefy laptop power pack. A car lead-acid battery charger with a cap on the output might also work, they can put out over 10A although some of them have more than just a transformer + rectifier. I've got a BOSCH one and it only has a 12V centre-tapped transformer with a full-wave rectifier followed by an ammeter. It can put out 10 amps of rectified sinewave.

It seems i have  laptop SMPSU that delivers  19V / 4.7A  Im a happy dude now

[Seekonk]

C

RV1 and RV2 arrangement weird ? Well it works I will not label it COARSE / FINE though....one just has to fiddle around and look at the meter to see when the current needed is present

Clearly with RV1 being 10% of RV1 the intent was to have a fine.  That 1K is a heavy load to place on the 10K pot.  That arrangment doesn't work well at all.  Change it and you won't need ten turn pots to adjust it.