OPA551: Dual OPA511 optimisation for audio path

Hey David, 

Thanks for putting this together, I see many good things I like to see in an audio amp circuit. 

Stability is a very important characteristic, especially for parallel amps.

Sometimes I do a sanity check to make sure my AC stability analysis makes sense. I often do this by injecting a small signal AC source without breaking the loop, and monitoring the output to make sure things look reasonable on the output. 

When I run a basic sanity check, I see that things are not happy with the parallel operation: 

I see that U2 is actually configured as an inverting amp connected after the U3 amp. This is not desired as we will have two 180 degree out of phase signals joining together at the Zobel network. 

When I alter the connection to U2 (run output of U3 to the non-inverting input of U2), I get the correct response:

I have now setup your design for stability analysis on U3:

I altered the value for C5 to 10nF as the Zobel network appeared to be causing some phase degradation in the frequency response.

Here is the phase margin for U3 after slight alteration to Zobel (nice and stable at 65 degrees)

I also checked U2 to make sure we are stable (stable as well)

I am including my simulation file for my phase margin simulations. Feel free to let me know if you have any questions on the design. 

OPA551 Phase margin JN.TSC

Thanks,

Jacob