0

Im making a battery tester / discharger. For this i'm using a constant current circuit made with an opamp and a mosfet. I built and tested the circuit by itself with around 3-4A and it was stable. I needed higher currents so i made 2 PCB's each with 4 parallel constant current circuits. Problem is it oscillates when the current gets to high. The diode is just for polarity protection.

I am about to go crazy over this behaviour as i don't really understand what is causing the oscillations and therefore i don't know how to get rid of it.

schematic

simulate this circuit – Schematic created using CircuitLab

If you have any ideas as to how i can get rid of the oscillations, please tell me. When i tested the circuit by it self i found C1 and R2 necessary to keep it from oscillating. If you need further information, just say so and i will update the question.

In advance, thank you!

Swagministeren
  • 241
  • 2
  • 7
  • Can you elaborate on the purpose of this circuit? maybe someone can suggest something better. – Norm Apr 24 '18 at 12:57
  • 1
    To stabilize it you can try to add a resistor (e.g. 1kOhm) from top of R1 to the inverting input and a capacitor (e.g. 50 pF) from the opamp output to the inverting input. The best values for the components are hard to guess at. – HKOB Apr 24 '18 at 12:57
  • what do you mean by parallel circuits exactly? – PlasmaHH Apr 24 '18 at 12:59
  • https://electronics.stackexchange.com/questions/246321/stability-of-constant-current-circuit – τεκ Apr 24 '18 at 13:02
  • https://electronics.stackexchange.com/questions/187563/designing-a-linear-mosfet-driver-stage – τεκ Apr 24 '18 at 13:03
  • I think that this behaviour is due to the mismatching values of the same components on the parallel circuits. This is a very simple response but i think is that the cause. – thoraz Apr 24 '18 at 13:07
  • Can you explain a bit more about what mismatch it is you think is happening thoraz? – Swagministeren Apr 24 '18 at 13:08
  • If you change simultaneously the value of the DAC, if the circuits are not perfectly matched, maybe one them sink more current than others creating some kind of instability (I think that this condition creates a positive loop in the others circuit's opamp, but I need to elaborate this concept, if i'm not wrong). I suggest you to abandon the parallel choice. – thoraz Apr 24 '18 at 13:16
  • The reason why i wanted to parallel the FET's in the first place was to be able to distribute the heat more evenly on a big heatsink im using. Because of this i im not too fond of abandoning the design just yet. My thinking was that each circuit would be "locally" regulated and there would be no problem scaling it up. – Swagministeren Apr 24 '18 at 14:07
  • What i think you mean, thoraz, is that the current draw from each CC circuit affects the voltage on the battery and thereby changes the current in the other circuits forcing them to regulate their own currents and then affecting the other circuits again, untill the whole thing starts singing? Can i simulate this behavior somehow or is there anyway to predict it? I have a very limited knowledge of phase margins etc. I will try HKOB's suggestion the day after tommorow when im back at work. – Swagministeren Apr 24 '18 at 14:07

1 Answers1

2

Decouple the heavily capacitive loading on the op-amp (gate capacitance and Miller capacitance of the MOSFET) with a resistor of 100 ohms or so, and add an AC feedback path around the op-amp to stabilize it.

Op-amps will tend to oscillate with capacitive loads, anything more than tens of pF can cause problems, depending on the gain (unity gain is worst) and the op-amp design.

I don't see any point to the 10K in your circuit.

schematic

simulate this circuit – Schematic created using CircuitLab

Spehro Pefhany
  • 376,485
  • 21
  • 320
  • 842
  • 1
    Tried your suggestion. When in my test setup it works! Also with 8 in parallel. Thank you! Got it running stable with over 30 amps. – Swagministeren Apr 26 '18 at 06:12