Dave's Constant Current Load; traps for young players...

[T4P]

... DOH!

[electroguy]

Looking at page 6 of the LMC662 data sheet (http://www.ti.com/lit/ds/symlink/lmc662.pdf,
the device is unstable at unity gain for capacitive loads greater than about 100pF.

sorry if this is a silly question, can you use a LPF on the output of the opamp going into the gate of the mosfet. ie: a 100k resistor, then a cap to gnd.
Obviously this would limit the response time? but would it stop the oscillations?

[jimmc]

Yes that is possible, if you introduce a dominant pole (I'm old enough to call it a dominant lag) into the control loop then the system should be stable.
Rule of thumb is the time constant of the dominant lag should be 4 times that of any other in the loop.

(On a practical note 100k is a bit high, 1 to 10k would be more reasonable.)

Jim

[free_electron]

The self balancing of parallel mosfets applies to switching mode, but not necessarily linear mode. It only works if you choose mosfets with a high enough channel resistance

ah yes, that part is of course key. use mosfets made for linear operation. throwing in a mosfet that goes from off to on in 1 microvolt is always a bad idea for these kind of things.. damned i forgot what the y call this.. i know the symbol is S and in dutch it is called 'steilheid' ( steepness ). its basically the relation between vgs and rds. aargh .. can't remember the english word .. transconductance ! eh no ..  that's  gm , but that is Ids vs Vgs ..

drat drat drat ..

-2 minutes later- Transfer Admittance ! 

My favourite way of building a a constant current load with multiple mosfets is to have a separate regulator circuit for each mosfet. The opamps cost nothing, and you get an exact equal share of power..

and that is how it SHOULD be done !

[LEECH666]

My favourite way of building a a constant current load with multiple mosfets is to have a separate regulator circuit for each mosfet. The opamps cost nothing, and you get an exact equal share of power..

and that is how it SHOULD be done !

That's what I am plan to do here (link below) with my dummy load implementation.

https://www.eevblog.com/forum/beginners/derpy-load-is-this-dummy-load-design-any-good/msg103974/#msg103974

I have too many projects running at the same time ... 

[icon]

I am puzzled by what you are trying to do. Your amplifier has a gain of 10K/13.6K = 0.735, and so when the input is 3.6V, your current will be 4.89A.

Your puzzlement is understandable. I don't know what I was trying to do either. Of course it doesn't need that resistor network - I was misunderstanding what was going on.

Built on a vero-board with a capacitor between -input and output, it seems to work OK, leading me to wonder whether the breadboard is a bit dicky. The only thing is a residual wobble at 2Xmains frequency.



It turns out that the MOSFET doesn't start turning on until about 3.5V (seems to vary - perhaps with temperature?) and it maxes out at about 0.85A at the maximum op-amp output of 5V. I could up the supply, but then I'd need a regulator for the panel meter - I'm not sure the heatsink would like either.

Anyway, thanks for all the help; I learned a lot more about op-amps than I would have if it had worked!

Regards
John

[electroguy]

you are most likely correct saying the 100hz is coming from your mains. Have you got capacitors as close as you can to the power supply pins of your op amp? (try 100nf and 10uf in parallel, put the 100nf as close as you can to the chip, then the 10uf as close as you can to the 100nf).
What input power supply are you using? switchmode or linear? have you tried a regulator such as 78L05 (make sure you have enough output capacitance and also the 100nf near the op amp IC) to see if the 100hz ripple is reduced?

[icon]

Hi

It has a 3.3uF cap soldered directly to the power supply pins on the op amp, but no 100nF - might root through the box and see what I can find.

It's a linear supply, with a 7805 and all the usual output caps.

Thanks
John