Prius Gen2 adapter board development thread

[ZooKeeper]

I reverted to FOC 4.85.R for testing, no uac and no change, am not getting any output and no errors. If there were some errors, I would be inclined to tear the inverter apart, being there are none, I will try and simplify the wiring and re-try.

[imwoody36]

you may reach out to me if you like,
we are on the same hardware and the same software
ive got motor movement, but we seem to have parameters issues.
812-820-5050

[imwoody36]

zookeeper, what do you make of this gen 2 setup?
https://openinverter.org/wiki/Configura ... 2_Inverter

I noticed these numbers

Toyota Prius Gen2 Inverter
Only inverter parameters

{
"pwmfrq": 2,
"pwmpol": 0,
"deadtime": 0,
"ocurlim": 100,
"minpulse": 1000,
"il1gain": 3.18,
"il2gain": 3.18,
"udcgain": 3.78,
"udcofs": 327,
"udclim": 540,
"snshs": 0
}

so... we seem to have 3 choices for pwmfrg on the webinterface drop down menu. does this mean that selection 2 is the one?

[johu]

Yes, 4.4kHz.
You can see the translation from numbers to string here or by typing the "list" command in custom command

[imwoody36]

thank you Johu!
is there a link for the revised motor parameters with the version 4.85foc or is there anyone who has it working with a gen 2 prius transmission ( either mg1 or mg2)?
my resolver seems to show the best waveform when plotting angle on the web interface when I select 2 pole pars.
but my motor just jumps and hangs.
it seemed to get worse with the inverter parameters listed a few posts earlier but I left them as they are published by johu
I have tried different current and swapping phases in software.
am using 220 vdc



also..zookeeper I saw your youtube , I suggest something more robust than toyota prismatic cells for the 12volt supply like I have here.

[johu]

You need to be systematic about pole pairs, otherwise you are left with too many variables. Set both polepairs and respolepairs=1 and try to do a full motor revolution while plotting angle. How many times does angle go from 0 to 360?

[RetroZero]

has to be someone who does it first.
Tried to clean up wiring before rechecking resolver and throttle wiring.
First, preharge wouldn't work anymore, and whilst testing the problem, I touched preharge signal from board to 12v.
Poof, no more Led. Very pissed off with myself. Hope that's 'all' I blew up .

[johu]

Might want to invest in a lab power supply instead of powering logic from battery while testing. Wonder how I know?

[SciroccoEV]

Might want to invest in a lab power supply instead of powering logic from battery while testing. Wonder how I know?

Or at least put an incandescent light bulb in series as a current limiter/warning indicator.

[RetroZero]

yup, online order here we go.....

[RetroZero]

i'm assuming it's the same ref as V3 boards - LED 0603 GN ?

[imwoody36]

okay set pole pairs to 1 and it appears to me that there are 6 zero crossings.see the photo
does that mean.....what?

[ascii-art]

This reverse engineering report might be useful. It says 8 poles for both MG1 and MG2 on pg. 10.

Evaluation of 2004 Toyota Prius Hybrid Electric Drive System. Oak Ridge National Lab
https://info.ornl.gov/sites/publication ... b57399.pdf

[johu]

That would mean 6 pole pairs for both polepairs and respolepairs

This reverse engineering report might be useful. It says 8 poles for both MG1 and MG2 on pg. 10.

And that would mean 4. Now what

[imwoody36]

I tried 4 and 6 , motor does try to spin, does hum, and does not jump off the table.
im using under 10 and usually under 3 for the manualID.

guess i can try 8...

[ascii-art]

Here's a detailed disassembly report that includes pictures of the rotor - Figures 4 &9. Looks like 8 magnets / 4 pole pairs.

REPORT ON TOYOTA/PRIUS MOTOR DESIGN AND MANUFACTURING ASSESSMENT. Oak Ridge National Lab 2004
https://info.ornl.gov/sites/publication ... b57348.pdf

[Jack Bauer]

So this idiot on youtube made a video that might help :

[RetroZero]

At least he didn't fry his control unit
Off to electronics shop in next town to see what I can find tomorrow ... Idiot

[imwoody36]

I notice that my angle plot is a modulated sine with 6 zero crossings in one revolution, but in Damiens youtube video at time 17:23 his is a nice sawtooth.

most information including the oak ridge report linked above and the youtube comments seem to agree there are 4 pole pairs for the prius mg2.
but I believe I saw damiens pole pairs was set to 2 and resolver set to 1 at time 21:37
these photos show that mg1 and mg2 resolvers are the same. I have just taken apart a gen2 transmission.
and you can also see there are 48 sections to the motor and 14 sections ( slots) to the resolver.
im wondering how 6 goes into 14....but both are divisible by 2 ( his pole pair setting.)

[ZooKeeper]

Is there any way to validate PWM output from the VCU using typical tools like resistors, LED, lamps, multi-meters, etc? After my melting insulation event, my motor has not turned or even tried. While I see and smell no evidence of damage to any component of the inverter or VCU, there is no measurable output on the 3ph wires as I had before.

Need to figure out which side is not working, or both. With the inverter and VCU connected as designed; a LED connected between 12v and MVU thru a resistor results in it being on, but dimly with no change in brightness regardless of VCU settings.

[RetroZero]

If anyone has nothing better to do than play with their new boards, I'm trying to narrow down what I fried. Since Johannes is on holiday, I am hoping to localise the problem and repair it myself....
I have PWR led on permanently and was hoping to test 'though hole points' in the first instance. Maybe it would narrow down my fault....
The 2 through holes next to 'stm' led have 0volts when powered up.
I have 0,3v on led point closest to R32 and zero the other side.
I really hope I haven't damaged STM

[imwoody36]

does it produce a web interface and does any" live" data show up?

do you have an oscilloscope?

[RetroZero]

Yes 'the web interface appears, but no data.
I don't have an oscilloscope however.

[SciroccoEV]

Yes 'the web interface appears, but no data.
I don't have an oscilloscope however.

Have you selected auto update and refeshed?

[RetroZero]

No, I'll try. But Led next to smd doesn't light up...