1
\$\begingroup\$

I have an AC signal that is isolated after a transformer and is an input to inverting summing op amp, and a DC circuit which supplies voltage as an offset to the same summing op amp.

The AC is 1 VAC, 400 hz, while the DC is 1V.

To create a common potential point, is it possible to connect the AC neutral after the transformer to the DC COM or should we keep them separate?

Would combining these two compromise the isolation of the AC and DC circuits?

I personally thought AC neutral was also hot and would damage the DC com port if connected, but I am not entirely sure.

Thanks for the help!

\$\endgroup\$
3
  • \$\begingroup\$ Are you in fact trying to lift the AC voltage so that it always remains a positive signal? Consider deleting your previous question because you talked about grounding in that question when you really meant connecting the AC output to a suitable DC voltage on your target electronic circuit. \$\endgroup\$
    – Andy aka
    Mar 17, 2020 at 13:57
  • \$\begingroup\$ @Andyaka I am trying to lift the AC voltage from 1,-1 to 0-2 with offset of 1 V. Our microprocessor can only process positive signals. I was lifting the AC voltage with an inverting summing op amp. There is a inverting unity gain buffer after as well to get the signal positive. \$\endgroup\$
    – Rickyfn2
    Mar 17, 2020 at 14:02
  • \$\begingroup\$ If you connect the secondary AC neutral to the DC return (ground), you now have a common reference point for both the AC and the DC. The AC will measure 1 VAC relative to that common return, and the DC will measure 1 VDC relative to that common return. I assume that you just have two terminals on the secondary side of the transformer. If so, then it really doesn't matter which one gets connected to the common return, unless you care about the phase of the secondary AC signal. \$\endgroup\$
    – SteveSh
    Mar 17, 2020 at 14:20

3 Answers 3

1
\$\begingroup\$

Only if you have an isolating transformer.

There are two kinds of transformers relevant here: Autotransformers, which do it all in multiple taps of 1 winding, and isolating transformers, where the primary and secondary windings are fully separated.

Nothing should ever connect to AC mains hot or neutral. The reason is simple: plugs can be flipped over or sockets can be miswired, and that AC mains neutral may actually be hot, and have 120V or 230V on it. That means do not use an autotransformer!

enter image description here

Note lack of polarization on the CEE 7/4 socket, and many 2-prong USA style plugs can insert either way into a NEMA 1-15).

So as long as your transformer is an isolating type, and secondary "neutral" is floating compared to AC mains hot or neutral, then yes, it's fine to tie your DC "GND" or Vss to it. (not to be confused with AC mains safety earthing).

You might do that in a center-tap full-wave rectifier, for instance -- a sensible choice at such a low voltage, since it needs only 1 diode instead of 2.

\$\endgroup\$
0
\$\begingroup\$

The AC is 1 VAC, 400 hz, while the DC is 1V.

If you have a 1 volt DC reference voltage then apply that to one end of your transformer secondary (and use the other as input to your ADC) and forget about the op-amp because it just isn't needed. However, you still need to do the numbers game.

You say your AC voltage is 1 VAC and that, without further context means 1 volt RMS and, if we assume a sinewave, that becomes a peak-to-peak voltage of 2.818 volts so, in order to lift it to a wholly positive voltage, the DC offset would need to be 1.414 volts.

I am trying to lift the AC voltage from 1,-1 to 0-2 with offset of 1 V.

Now that contradicts 1 VAC so, you have to make this decision but, either way you can apply an offset DC voltage to one of the secondary wires and use the other as signal input to your ADC.

Would combining these two compromise the isolation of the AC and DC circuits?

You can't make an omelette without breaking eggs - with both connected to each other there is no isolation between them (and neither would there be isolation if using an op-amp summer) so, once again, you have to decide if this is important to you or not much of a big deal.

One further bit of advice - use an RC filter just before your ADC to give a little bit of anti-aliasing and the bonus is that the resistor will also act as a current limiter should the actual AC p-p level develop into a larger signal under recognized fault or overload conditions.

\$\endgroup\$
7
  • \$\begingroup\$ We have 0.707106 V RMS so the actual peak to peak of the signal on the scope was from 1 to -1. I should have clarified on that a little more, my apologies. I was thinking of an RC filter as well for anti-aliasing, but at the same time we want to be in phase to the input signal and the RC would shift the phase of the input signal. \$\endgroup\$
    – Rickyfn2
    Mar 17, 2020 at 14:41
  • \$\begingroup\$ How much phase shift will your ADC sampling produce at 400 Hz? \$\endgroup\$
    – Andy aka
    Mar 17, 2020 at 15:14
  • \$\begingroup\$ I do not fully know the answer to that question, I can look into the datasheet for the ADC but i believe probably not enough to even be able to measure. \$\endgroup\$
    – Rickyfn2
    Mar 17, 2020 at 16:06
  • \$\begingroup\$ What sampling rate are you using? \$\endgroup\$
    – Andy aka
    Mar 17, 2020 at 16:40
  • \$\begingroup\$ I do not have it off the top of my head, but we are oversampling the signal and we are well above nyquist as well. \$\endgroup\$
    – Rickyfn2
    Mar 17, 2020 at 18:35
-1
\$\begingroup\$

Can't you use a diode clamper to increase the dc offset?enter image description here

\$\endgroup\$
0

This site is temporarily in read-only mode and not accepting new answers.

Not the answer you're looking for? Browse other questions tagged .