I have a phase circuit that I have designed that uses four Op Amps to take an AC Input in the range of 50-60hz and then produce copies of that signal in but each phase shifted by 120 degrees lagging the next. IE. Output 1 is 0 degrees, Output 2 120 degrees lagging Output 1, and Output 3 is 120 degrees lagging Output 2.
The power rails are set to
+12v on pin 4
-12v on pin 11
OpAmp 1 is connected as an inverting amp with a gain of ~2 (using a 10k and 20k set of resistors. )
Op Amp 2 and 3 are connected as an inverting unity gain of 1 using 10k set of resistors.
Op amp 4 is connected as a non-inverting amp with a gain of 2.
Using an OPA4228:
When I run a signal of say 1vrms into the configuration I get what I expect except that there is massive noise all over what started as a very clean 60hz signal from a lab signal generator.
If I remove hte OPA4228 and drop in a more generic quad op amp (supposedly more noisy) like an LM324 all of the outputs clean right up and the circuit produces the desired results.
My question is why is the OPA 4228 not dropping in and simply working?
Reading through the specifications it should operate better than the LM324 or other garden variety quad op amp. (I am trying to achieve a fairly tight SNR, and although the LM324 works, I can see noise artifacts on the 2nd and 3rd outputs that I would like to reduce. Output one is clean to about -80-90dB with mostly the same artifacts as the signal generator.)
Any thoughts or suggestions. Seem like I am missing something simple here.
Thank you.
Russell