2
\$\begingroup\$

I'm currently using the MP111 hybrid op-amp set up as a voltage controlled current source with a gain of 1. But what I'm finding is that at low input voltages the gain is much higher than it should be - up to 8 A/V. As we increase the input voltage the gain decreases sharply and then becomes 1.

Why is the gain so off at low input voltages? Is there a way to fix this?

I'm calculating the gain by dividing the current seen passing through the shunt resistor (Rs) by the input voltage.

MP111 datasheet: https://www.apexanalog.com/resources/products/mp111u.pdf

MP111 configuration as current source (figure 8): https://www.apexanalog.com/resources/appnotes/an07u.pdf

Rs = 0.01 Ohms, Rload = 0.05 Ohms, Rin = 101 kOhm, Rf = 1 kOhm.

The positive and negative rails are at +15V, -15V respectively. Boost functionality is not being used therefore Vboost pins are connected directly to the power rails.

Image of the gain vs. input voltage

enter image description here

enter image description here

\$\endgroup\$
  • 1
    \$\begingroup\$ Probably due to input offset voltage. \$\endgroup\$ – Marko Buršič Oct 24 '18 at 6:36
  • 1
    \$\begingroup\$ Marko's suggestion about offset may be a part. But that opamp's specs and your curve don't convince me that's all there is to it. Why not a discrete design, by the way? (That opamp is boutique and expensive.) \$\endgroup\$ – jonk Oct 24 '18 at 8:08
  • \$\begingroup\$ What are +/-Vb pins connected to - show your entire circuit please. \$\endgroup\$ – Andy aka Oct 24 '18 at 12:02
  • \$\begingroup\$ Could you show your layout? \$\endgroup\$ – John Birckhead Oct 24 '18 at 12:40
  • \$\begingroup\$ I've added a schematic of the circuit. In terms of layout - It's just on a protoboard with discrete components on top of the MP111 \$\endgroup\$ – VanGo Oct 25 '18 at 2:42
1
\$\begingroup\$

Thermal feedback, of unknown magnitude and polarity, will either buck or boost your desired control-loop signal.

To test, abruptly change the Rload by 50% (simply hold a similar Resistor in parallel), and observe the behavior.

\$\endgroup\$
  • \$\begingroup\$ What should/shouln't happen when I do this ? \$\endgroup\$ – VanGo Oct 25 '18 at 2:00
  • \$\begingroup\$ You should get a 50% change in some voltage. But you've already stated "the gain is 8:1, not 1:1"; I expect something changing by 8:1, or at least far more than 1:1. \$\endgroup\$ – analogsystemsrf Oct 25 '18 at 5:53

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.