I need to supply the galvo driver with a signal "Voltage Range ±3.0V p-p or ±10.0V p-p (Differential Input)" - that's what it says in the datasheet. I have a 0-5V single ended DAC and differential opAmp AD8132. So, I am planning to create the differential signal by the following schematic:

enter image description here

The output of the DAC will give me a signal ranging from 0 to 5V and I can scale it down using R26. The other pot (R25) will generate the proper voltage to offset DAC's signal to 0V DC. This way, a 0-5V wave on one input and +2.5 on the other input will give me ±2.5V between the outputs of the dual buffer U4. These signals then fed into differential opAmp U5 with G=1.

Question 1: is this the proper way to generate a differential signal? I believe that I could place a 0.1uF cap between R26 and +in of U4 and it would automatically offset it to 0V, but since a part of my signal is DC, it would compromise the signal, right?

Question 2: What should I measure on each of the differential outputs of U5 relative to ground?

  • \$\begingroup\$ Your circuit is not going to work without connecting U5 and U6 to a power source. \$\endgroup\$
    – The Photon
    Jan 27, 2015 at 4:45

1 Answer 1

  • Gain:

I believe the gain in this design is too low to get a +/-10V output signal.

$$V_{out diff} = \frac{390Ω}{390Ω}.(V_{i+} -V_{i-})$$ $$V_{out diff} = 1.(5 - 2.5) = 2.5V_{diff-mode}$$
or $$+/-1.25V_{comm-mode}$$

Alternatively you could make use of the Vocm, instead of the V- input to offset the ADC value.

  • Question 1:

Correct, this would destroy the DC component and is only possible if you could guarantee that the ADC generates a continuous AC signal, which I presume is impossible in your application. Even if that would be the case, you still have to take the lower cut-off frequency into account. Signals below 1Hz would require considerable capacitor values.

  • Question 2:

You would measure an in-phase signal with half the amplitude of the differential mode signal on the +Vout pin and an opposite-phase signal on -Vout.

You may want to read this application note as well.

PS: You may want to consider a fixed voltage divider for R25/R26 in series with a smaller value potentiometer if calibration is needed.Like this for instance.

Like this for instance

This allows for more accurate calibration. The current single potentiometer design is very course.

  • \$\begingroup\$ Thanks for the app note - good one. The current schematic has problems because if I change DAC's output range using R26 (say 0-3V), then I have to adjust R25 to make 1.5V offset. Are you saying that I should connect DAC's output to the +in, 0V to the -in, and 0V to the Vocm and I do not need to generate the offset voltage on the right side of the schematic? \$\endgroup\$
    – Nazar
    Jan 27, 2015 at 15:26
  • \$\begingroup\$ Correct. Wire the AD8132 as a single-ended input and use Vocm as offset, like fig 4. in the AN. \$\endgroup\$
    – Ambiorix
    Jan 27, 2015 at 22:50
  • \$\begingroup\$ Not as simple. I want my signal to be offset to 0V, so I connect Vocm to ground. However, because my input signal is DC offset, I get this same offset on the output. \$\endgroup\$
    – Nazar
    Jan 28, 2015 at 13:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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