diff amp or common grnd

[buzzp]

I am measuring mains voltage with a microcontroller. I use a xformer for control power for my circuit. Currently, the design ties the primary and secondary of the transformer together to create a common neutral point in order to facilitate measurements. Since the design already encompasses a transformer, I am contemplating changing the design slightly to use a diff amp and eliminate the tie between the primary and secondary of the transformer. Ignore the additional accuracy obtained by using a diff amp and please tell me your thoughts on transient immunity, etc, based on the two front ends: xformer primary and secondary tied together versus not tied together and use a diff amp.
It seems to me the diff amp configuration would be more immune to noise and transients on the line since the neutral reference for the electronics is isolated from the neutral on the line voltage. However, the signal it is measuring would still bounce around with noise. Nothing about this circuit is fast acting so no decisions would be made based on transients that may make it through the filtering.
On the other hand, I don't want to make the circuit and PCB changes and redo tests, if the gain in going to a diff amp to isolate the neutrals is minimal. I just thought we have a transformer so why not isolate the circuit from the neutral.

Thanks in advance for your comments.

 

[IRstuff]

There additional risks with an isolated circuit, which is that power-on transients can be coupled into the circuit and fry the inputs. You could grossly exceed the maximum common mode input every time the power is applied.

TTFN

FAQ731-376

 

[buzzp]

I guess I don't understand what you are saying. Are you saying any connected equipment on or near the mains I am measuring from will cause common mode voltages at power-up, on the hot and neutral on the mains?

There is some transient protection across the line in the device for non-common mode noise.

Maybe you would lean towards leaving the circuit alone with a circuit neutral common to the mains neutral?

 

[IRstuff]

I would lean toward coupling with a large resistor, which is often done to ensure that the odd transient common modes get discharged through other paths than through the input emitters or bases of differential circuits.

TTFN

FAQ731-376

 

[buzzp]

The resistor is a given. I am not that new. LOL.

I am more curious to hear about the 'common-mode voltages at power-up'. Do you mean motors and the like near my device? Some elaborating would be appreciated.

 

[IRstuff]

Not sure how related my experience is to your situation. We had one digital system with 400Hz 120VAC to DC, and a data recorder running inside a PC powered with 120VAC. Digital differential inputs between the two systems.

The receivers lasted about 3 power-on cycles before they shorted out. Basically, it appeared that there were common-mode potential differences between the two AC systems, even though they presumably shared the same neutrals at some point, but the potentials equalized through the differential receivers.

Your situation may be totally different...


TTFN

FAQ731-376