0
\$\begingroup\$

I have a need to trim the Vos of an op-amp (LT1006) running in a single-supply circuit (12.6V). From what I found so far the proper way to do this would be by using some bias voltage connected to both inputs rather than using ground as an input which would only work correctly in a dual supply circuit.

Here's the constraint: the op-amp will already be soldered to the final PCB at this point and I can't easily lift the inputs off the ground to perform the trim using an offset voltage.

I can however quite easily temporarily provide an external negative supply (-12.6V) to operate the op-amp in bipolar mode during the trimming process (and use grounded inputs).

My question is: would my Vos trimming still be correct if I subsequently operate the opamp with a single supply?

Any suggestions to perform trimming in a different way?

More information: The circuit is simply a low-side current sense amplifier using a single supply. Trimming would be performed by shorting Rs and trimming output to equal GND.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
2
  • \$\begingroup\$ Please provide a schematic of your circuit. \$\endgroup\$ – Transistor Aug 8 '16 at 12:57
  • \$\begingroup\$ It's respectful to other users to post as much information as possible about your question, If people have to ask for more info then your post may not be clear enough. electronics.stackexchange.com/help/how-to-ask \$\endgroup\$ – Voltage Spike Aug 8 '16 at 17:04
1
\$\begingroup\$

You can get some idea of the offset voltage change vs. supply voltage from the below figure taken from the datasheet:

enter image description here

Adding the negative 12.6V is equivalent to changing the common mode voltage by -12.6V. The change looks like it could be some tens of uV, but probably typically < 100uV. Something like -0.3V might be better than -12.6.

Usually it's better to trim with an output voltage enough offset from 0V that the amplifier is working well, yet fairly close to ground so as to minimize iterations if you are doing span/zero calibration.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks for the suggestion, indeed I was wondering about the impact of the supply span on the input stage and using -0.3 to -0.5V seems like a sensible choice. Update: re-checking the datasheet for the opamp output capabilities seems to confirm that it can swing the within a few mV of ground with a single supply so even -0.3V would be enough. \$\endgroup\$ – Rob S Aug 10 '16 at 12:08
  • \$\begingroup\$ @RobS -0.3V has another advantage over using -12V in that the output of an op-amp can go into an ADC, for example, that is powered from 0/5 or 0/3.3 and not risk damage if the op-amp output momentarily slams to the negative rail. \$\endgroup\$ – Spehro Pefhany Aug 10 '16 at 12:14

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.