Prius Gen 3 Inverter Logic Board Support Thread

[bobby_come_lately]

OK, another question (sorry).

I'm seeing some weird behaviour on udc. Currently have ~20V connected to the points suggested by Damien to avoid exploding capacitors in this video ( - 31:10) though I have not yet bridged to hook in the DC:DC converter.

udcmax is set at 100V. Yet it keeps tripping out for overvoltage. This is because udc is registering as around 770V. I'm getting nothing like that on the multimeter, obviously. I have tried changing udcgain but I still get very weird results. Even with the power disconnected, udc is floating around the same level.

Know the voltage is low even for testing but same thing at ~30V.

Appreciate any pointers.

[Jack Bauer]

Most likely you have an open pin on the 50way connector. Also verify your -5v supply is working.

[bobby_come_lately]

Damnit. Was hoping it wouldn't be that. Thanks Damien.

[Bigpie]

I only have access to a bldc motor at the moment, hopefully going to try get this spinning over the weekend open loop. Is there a simple circuit to convert the hall effect singles to resolver or encoder ones to run with feedback?

[Bigpie]

@Jack Bauer can you confirm the current sensor connector orientation?
In your video the notches face outwards, but on my original Toyota board they face inwards, I want to be sure the instructions I'm writing on the wiki are bang on.

Screenshot 2020-06-20 at 11.53.27 am.png (255.96 KiB) Viewed 6652 times

[Jack Bauer]

The video is correct

[Bigpie]

Thanks.

Anyone able to give an overview of commands needed to get spinning?

I've connected the motor to the 3 mg2 outputs
Connected 60v battery to the HV bus, controlling pre charging manually
Connected pin 1 on Ampseal to 12v, 11 to - and pin 3 to 12v for forward direction.

In the web interface I've run

start 2
set fslipspnt 10
set ampnom 100

Nothing else is connected.

http://www.eastgem.net/d7500.html is the motor I'm playing with

[Bigpie]

Update, still rotationally challenged.

I've now pulled the board back up and just trying to get to see the the switching singles on the bench.

I've tried flashing "4.85.R-foc" posted earlier in the thread.

Wiring wise

Code: Select all

12_IN is 12v
MG2_FORW_IN is 12v
GND is power supply -

Commands in web UI

Code: Select all

start 2

UI Shows

Code: Select all

opmode	Off
lasterr	TMPHSMAX	
status	WaitStart

[9198]: WARN - HIRESOFS
[9583]: DERATE - TMPHSMAX
[500]: STOP - PRECHARGE
[4020]: STOP - OVERCURRENT

tmphsmax is set to 150
udcsw is set to 0

Any tips?

[Isaac96]

Since you're controlling precharge manually you need to set udcsw to a level below what udc is reading.

TMPHSMAX will also be annoying - try changing the tmphs parameter until 'get tmphs' returns a level below 80(???)

Do you have any current sensors hooked up? FOC requires sensors to run (not sure about FOC in open loop mode...)
Just plug in the Toyota ones I guess? That should at least get rid of the errors.
I'm not sure how to use the FOC code in manual mode, having only dealt with sine.
-Isaac

[Bigpie]

I can put the sin back on. Just trying to work out how to get it generating the pwm signals with the bare minimum for now.

[Bigpie]

I've put the sine firmware back on, connected up just the current sensors and set ocurlim to it's max value 65535 and tried the min -65535 and still getting OVERCURRENT error starting in manual mode.

Anyone else with one of these boards got it up and running yet?

[Jack Bauer]

Overcurrent logic on the V1c prius board is very simple. It just takes the active low fault signal from MG2, inverts it with an npn transistor and feeds it to the stm32 on two pins. Most likely problem is the 50 way connector not seated or dry jointed.

[Bigpie]

Doh, that'll explain things, I've got the board on the desk with just the current sensors connected so that pin will be low. I see that on the schematic now. Just trying to verify I'm getting the pwm outputs when I'm expecting them based on my use of the UI.

[bobby_come_lately]

Just got my board back from a local electronics engineer who has been over the board and sorted out the dry joints (he confirmed one on the 50 pin connector). I'm as confident as I can be (having not done the work myself) that that's no longer the issue. But I'm still struggling.

Most basic symptom is that udc is still swinging around all over the place, as you can see from the chart attached. This is with 20V coming in and udcgain set to 4.65. Really appreciate anyone's thoughts on what might be going wrong and whether I've done something really stupid.

[Jack Bauer]

Nobody should be soldering the 50 pin connector without flux. Its just a waste of time.

[bobby_come_lately]

Sorry - probably terminology error on my part. What I mean is there was at least one bad connection but that has been resolved.

Starting to wonder if my inverter is bad, though that seems unlikely.

[Bigpie]

I fluxed it up, then dragged solder across the pads with a flat tip then went through the pins one by one with a fine tip.

[Jack Bauer]

udc should be very stable. Have you checked your -5v power supply?

[bobby_come_lately]

OK I shall be having a beer now while the embarrassment wears off. Turns out the battery I was using for 12v supply had drained.... a lot.

Checked the -5V on your suggestion and it was a bit off (closer to -4V) so stuck multimeter on the battery and... less than 8V.

Think the issue was the 50 pin before as the battery was freshly charged a few days ago. But this is likely the issue now.

Thanks again JB.

[bobby_come_lately]

Sorry, me again.

Got a solid 12V PSU on it and I'm still having the same issue: udc wild. Checked the -5v and it's still at -4.09. Not sure if that is a clue to the issue.

Inductor coil also seems to be making a lot of noise. Kind of a high pitched hiss.

[bobby_come_lately]

Right, fresh head this morning. I went through the schematics and the layout and traced the path of the DC bus connection from the Toyota board through the 50 pin connectors, to the ADC on the chip. That does not seem to be the issue.

BUT, when I check the voltage coming in to R76, it is all over the place.

Does this suggest there is something wrong with my Toyota board?

Or is this a product of that -5v reading being low?

[bobby_come_lately]

Update: ordered another inverter for testing. Paid a little over the odds but desperate to get this working.

I'd appreciate some feedback from anyone with the board running on what voltage they see at -5V via.

[Bigpie]

I'll get a measurement for you tomorrow.
Edit, attached. This is a board but not connected to an inverter.

[Dylan Witt]

does anyone know if the prius gen three inverter control board will work with the is300h inverter? rewiring and all. i would like to be able to use both inverters. should see the control board from evbmw sent to kentucky very soon

[Dylan Witt]

another question. i want to use the pot-meter for my regen braking, is there a bettery diagram on connecting the wires? like sensor 5 volt referance, signal, ground pins. cruse control, accelerator pedal signal. im confused on the wiring, are the connections marked for the signal from the sensors and all i need to do is connect power and ground to each? im sorry for this question.. ive second guessed my self on this and ive confused my self