1
\$\begingroup\$

I am attempting to trouble shoot a circuit and I am a bit stumped. The circuit is high current source (10-20A) based on a master-slave configuration using OPA549 high power op-amps. There is one master and 4 slaves. A set point voltage controls the current depending on the load resistance. In this case the load is wire wound coil with 1 ohm resistance.

enter image description here

When the set point voltage is positive the circuit works as expected. However, when the set point is negative the circuit exhibits oscillation about the set point. From reading, I realize such configurations suffer from stability issues when driving capacitive loads, but I imagine the load in this case is predominately resistive and inductive in nature. Maybe a variations in the op-amps lead to them fighting to regulate the output and I should increase the isolation resistance at each output. However, I am puzzled why I only see oscillations for negative voltages. Both supply voltages also show oscillations when the set point is negative as well. I suspect the oscillations in the supply voltages are caused from the variable current draw of the op-amps and not the other way around. The output voltage (labelled N1) and the AC coupled supply voltage are shown below. Any thoughts would be greatly appreciated. Thanks in advance.

enter image description here

\$\endgroup\$
  • 1
    \$\begingroup\$ Hmm, maybe smaller feedback resistors? Should the feed back come from the output? Does it oscillate without the slaves? With just one slave? What are the TVS diodes for... they may look a little capacitive. edit.. I assume you just haven't shown the power supply by pass caps. \$\endgroup\$ – George Herold Jan 20 '15 at 19:10
  • 1
    \$\begingroup\$ Big support of George's comment about decoupling. Your oscillation frequency (~3 MHz) suggests one or more of the op amps is really unhappy. I also suspect you may have layout problems - current this high can be a challenge. You do have a ground plane, and you have taken current flow paths into account, right? As to why the problem occurs for negative voltages, that's because the internal construction of the negative side is different from the positive - look at the pulse response figures - rising and falling edges are not dealt with identically. \$\endgroup\$ – WhatRoughBeast Jan 20 '15 at 22:26
  • \$\begingroup\$ @GeorgeHerold I looked into your suggestions. The circuit is stable with just the master, but oscillates once I add a single slave. The circuit with all four slaves is stable when the load is open circuit. I tried doubling the isolation resistance, but to no avail. Do you think having the master sample the load voltage rather than the output voltage would make a difference? I presumed the only difference would be compensating for the voltage drop output resistance. \$\endgroup\$ – user2966694 Jan 22 '15 at 15:45
  • \$\begingroup\$ @WhatRoughBeast Thank you for your comments. I did not show the bypass capacitors for simplicity. The TVS diodes are used to suppress transient voltages spikes when the current is switched off when driving inductive loads. In terms of pcb layout, I keep high current traces pretty thick (1cm) and I have ground plane for the return current. – \$\endgroup\$ – user2966694 Jan 22 '15 at 15:46
  • \$\begingroup\$ @user2966694 - Another thing you might try. Instead of driving the slaves directly, try isolating each slave with about 100 - 1000 ohms for the signal input. That is, use a 100 - 1000 ohms between the master output and each slave input. \$\endgroup\$ – WhatRoughBeast Jan 22 '15 at 16:27

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.