0
\$\begingroup\$

I have a circuit like this:

CircuitLab Schematic 8rtgbw

The part on the left generates AC signal. The part on the right detects whether AC is running through the branch when S1 is closed. I use a resonating LC circuit which is tuned to the frequency generated on the left. This induces AC which is then amplified by the op amp and read by the scope.

There are some limitations to what I can do, hence such a complicated detection of a switch. Don't try to improve that, everything works fine and I'm getting what I need. But only when the left and right parts are on a different battery! When I connect them together like you see here, I'm getting a constant AC sine signal of about 20mV even when S1 is not closed. This is about 80% of the signal I'd ideally receive so it dramatically decreases my resolution. When I connect my scope to the + and - in red section though, I don't see any ripples or fluctuations (or perhaps they are too small, my scope has only 20mV sensitivity).

The problem IMHO is that a small AC is somehow spreading through the circuit and making the detector coil resonate when it shouldn't. When I split the generator and detector parts and put them on a separate battery, everything works fine, hence I assume this is not happening due to some wireless transmission.

When I remove the C2 cap, the coil stop resonating and the parasitic signal disappears (along with what I want to see, of course). When I remove the coil itself, the signal disappears too. That's why I'm assuming this is caused by an unwanted resonation.

Do you have any ideas how to get rid of this? I really can't have it on two batteries, it must be just one. (I tried putting caps – large and small – across the power source, across the generator branch, and to various other places of the circuit to smooth the DC, no luck.)

\$\endgroup\$
  • \$\begingroup\$ You don't show a decoupling capacitor on the op-amp supply or on the R4/R3 bias junction. I'd start there. \$\endgroup\$ – Transistor Feb 10 '16 at 9:34
1
\$\begingroup\$

I had an issue once due to the ground reference being at a slightly different potential at the input to what it was at the opamp. This was due to extremely poor routing of the pcb which introduced an unwanted resistance (Back when we tried to do things 2 layer and ground was more of a weaving spaghetti than what is now just a solid plane). It was a long time ago but I think it was a similar situation where we had a CT that was producing a waveform when there was no current flowing through it. Perhaps have a think about that sort of stuff also.

\$\endgroup\$

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.