I'm trying to build a pulsed voltage controlled current source using an op-amp. It has a tendency to oscillate (although sometimes to works okay, just often enough to give me hope that I might be close).

I'm looking to drive currents from 100 mA - 1A through a load which ranges from 12 Ω - 35 Ω. I'm using an LM675 (datasheet) because I have it at hand, and it's rated for the voltage and current. Vin is provided from a Labview-controlled DAQ. The schematic is below.


simulate this circuit – Schematic created using CircuitLab

The 1 Ω resistor and 0.22 uF cap from the examples on the LM675 data sheet. I confess I don't understand stability all that well.

I've tried similar circuits with other op-amps and the past, and sometimes been sucessful, and other times encountered the same problem I have now. I'd appreciate any advice.

  • \$\begingroup\$ 1. LM675 is not unity-gain stable. Depending on the load, this could be your problem. 2. Notwithstanding all the example circuits in the datasheet show the 1 ohm + 220 nF hanging from the output, capacitive loads can muck things up for most op-amps. Try removing the cap and see what happens. \$\endgroup\$ – The Photon Jul 31 '16 at 2:15
  • \$\begingroup\$ LM675 is stable for gains 10+. With R_Load 12-35 ohm noise gain is in 3.5-8 range, so no way. Just try adding compensation network as Fig.12 in LM675 datasheet. But some extra info is needed: 1)Is load purely resistive or has any reactive component? 2)What about power supply decoupling? High output currents demand stiff supply. \$\endgroup\$ – carloc Jul 31 '16 at 6:49
  • \$\begingroup\$ The photon: Without the 1 ohm +220nF off the output the circuit will always oscillate and worse than with them there. \$\endgroup\$ – Dan L Jul 31 '16 at 12:53
  • \$\begingroup\$ carloc: How do I calculate noise gain? I suspected this was related to my problem, but I wasn't sure how to calculate gain for since it wasn't an amplifier. I will try to network from Fig 12. To answer your other questions: 1) The load is nearly resistive. Its remote from the circuit and fed by about 2 ft of BNC cable, so there is a slight capacitance in parallel with it. The load does have a high TCR, and its resistance does increase (but still stay within the 12-35 ohm range). 2) I've got 20 uF between each supply and ground near the op amp. \$\endgroup\$ – Dan L Jul 31 '16 at 12:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.