intraspin library with resolver interface

evs,
I'm not sure if anyone has interfaced a resolver with InstaSPIN-MOTION. As you said, the ADC sampling is tricky with a fast control loop and just a single ADC converter. I've forwarded to our resolver expert to see if they can comment.

The resolver library from TI takes samples over the entire carrier cycle, and hence would, ideally, need a dedicated S/H if not an ADC.To avoid ADC (or S/H) timing contention, the resolver exc carrier and phase current sampling moments would need to be synchronized. This also assumes that the resolver feedback signals are clean of power switching disturbances.

Secondly, if the inverter GND and resolver fbk GND are not isolated, then the power switching instances would contaminate the resolver feedbacks, in which case, the resolver fbk can be sampled only synchronous to phase current sampling when  there is no switching. Care should be given to ensure that the resolver sine or cosine feedbacks' have max value corresponding to this instant for max ADC utilization. 

rgds,

ramesh

I think, I have 3 options to solve this:
1. Add an extra ADC SOC to sample at the middle of the positive Rotor signal. So I can under sample and have maximal S/N. The problem is that sampling the motor and sampling the Resolver at one point are going to conflict, resorting in delays in the measurements.
2. Modify the ADC that it is free running at app 160khz and collect and filter with the CLA. add an extra timer to manage MainISR. basicly split ADC and MainISR function in 2 separate loops.
3. make new hardware with 2 controllers one for motor control and 1 for feedback

1 has the least impact and 3 the most. I have my doubt if option 1 is accurate enough. at the moment I searching if option 2 is possible with the hardware.

Option 1 is viable, start with sampling the resolver feedback as this is varies at high frequency. The phase current will be relatively flat during the time the resolver signal is being converted.