Prius Gen2 + EM57 FOC no workie.

[ZooKeeper]

Trolling for suggestions I guess. I verified and re-verified all my wiring (resolver etc) and I can run the blinky test and motor in SINE firmware just fine, but 100% no joy with FOC. This leads me to think there is something in the resolver circuit or inverter pinswap (5 for Gen2 Prius), or.... an unholy combination of the two that is causing it not to tune, but after stepping back for a few days and re-tracing my steps, I am no further along.

Current state (pun intended), is that any time voltage >~5v is applied in Run, the motor turns with speed proportional to voltage. Throttle application causes the motor to slow, change direction and resume previous speed, but run less smoothly. AGAIN - suggesting a resolver wiring issue

Maybe I just need convincing that the third time will be the charm???

TIA.

EDIT: FOC v5.24.R

[ZooKeeper]

Update, I *think* I found at least part of the problem. When I checked jitter again, I found it to be nearly 360*

I then power cycled the VCU and inverter and rechecked, jitter now <1* and was like WTF!?!?!?! After another Run, jitter back to well over 300* and that ain't gonna work.

Any ideas on why jitter would change like that? All metal bits are earthed together, resolver wires are twisted (untwisted resulted in jitter over 5* which I fixed way back in testing).

[manny]

Is this on Sine firmware?

If so the slip angle is added to the angle variable.

[ZooKeeper]

I did not think of that :doh:

But both plots are from 0 motor volts, the only difference is what throttle applied fist or not.

Current jitter testing plans is to re-run tests w/o any HV:

• If jitter is excessive
  - Disconnect one resolver pair at a time & retest
  - Disconnect progressively from motor to VCU while testing each change

• if jitter is not excessive
  - add HV & retest
  - further separate cables
  - remove Toyota to Nissan jumper & wire VCU directly
  - install a ferrite (?) at the VCU end of cable

My 3ph cables are twisted (8ga 300v rated) extension cord, I would *think* they would not be a large emi source at low V and Hz, but I ain't to radio engineer so could be completely wrong. I do not pick up interference on any radio I own (AM/FM/CB) with antenna in the same room.

I think it bears reminding readers that I am absolutely stretching my knowledge with this project, in terms of programming, motor control and safe battery implementation at HV. I know perfectly well how not to die, but combining that with making it work.... a challenge

[ZooKeeper]

Update: A self-inflicted situation is the short answer.

How/why? Well, see when you move from one procedure to another and do not completely restart from "0", things get missed. I am betting there are a few here who have done this, damned sure I have a time or two.

So what happened? In step 4 (based on my notes) of the "FOC Tuning" video, Damien admonishes everyone to use the plot function to check for 360* of rotation and measure the resolver signal for jitter "< 5* is ok, < 2* is better). Turns out this was at least my THIRD round of attempting the tune procedure and I HAD LEFT the FWD pin PULLED HIGH! The FWD wire was hiding behind the VCU and connected to a 12v source. So when I went troubleshooting, I found the grossly excessive jitter and lost two days due to wheel spinning.

MAKE SURE none of your direction wiring is pulled high when checking your resolver signals!

To be perfectly clear, I had carefully checked the resolver signal and rotation TWICE before ever attempting to power the motor and had not changed anything in the wiring since then.

The question still remains: Why will this motor not spin while performing the FOC tuning procedure? I have tried voltages from 50-100vdc and all four potential resolver connection arrangements, resulting in no joy

Second question: Which is the accepted SINE test procedure? 3Hz and add amps until it turns, or Ampnom = 70% and add Hz until it turns?

[MattsAwesomeStuff]

Duplicate. Delete pls.

[MattsAwesomeStuff]

The question still remains: Why will this motor not spin while performing the FOC tuning procedure? I have tried voltages from 50-100vdc and all four potential resolver connection arrangements, resulting in no joy

Is it jittering and pulling current, or, just nothing happening?

Second question: Which is the accepted SINE test procedure? 3Hz and add amps until it turns, or Ampnom = 70% and add Hz until it turns?

I'm the wrong person to answer what's proper, but I was told a mixture of both. You start with both low, and then just see what can make it spin. I was told to increase ampnom.

Are you confused because you can observe 2 separate and conflicting sets of instructions, or, are you confused because you've come up with these two procedures on your own and aren't sure which is valid?

[ZooKeeper]

Thanks Matt, two different procedures is the confusion inducing thing. I am not sure, to your point above if it actually matters, but as I cannot get the syncofs to work correctly with my combination it does present a potential factor.