openinverter forum

jeffb wrote: ↑Sun Apr 24, 2022 10:03 pm

chrskly wrote: ↑Mon Mar 14, 2022 7:49 pm I'm no expert, but I have two other variants of this board so I might be able to help out a bit...

Thank you, that was helpful! I have the web interface coming up now, but I'm having trouble getting the motor to spin. I suspect a sequencing issue with connecting power, and was hoping someone who has gotten this working may be able to spot my error.

When I power up and press 'Display Error Memory', I see the message "[500]: STOP - PRECHARGE". I'm using a 30V, 10A benchtop power supply to provide "high voltage" temporarily so I can attempt to give the motor a test spin. I have connected the power supply to the screws per the wiki (https://openinverter.org/wiki/Toyota_Pr ... Connection). I have a separate 12v wall wart powering the logic board, via the IDC connector pins.

After some research, I figured this was because the 'udcsw' parameter was set to 330v. I changed it to 25v for now. I'm assuming the parameter takes effect immediately, as I saw an affirmative message in the log ("OK Set Ok"). I measured the voltage at the pins with a multimeter and it was indeed 30v, but the power supply reports that zero current is being drawn. I cleared the error messages and pressed 'display error log' and it showed the pre-charge error again again. If I power cycle the 12v to the board, udcsw is set back to 330v.

Q1: Do I need to do anything special after changing udcsw to get it out of the error state?
Q2: Is the "STOP - PRECHARGE" error that I see the second time a stale message from before, or is it failing repeatedly? Is there any way to know if it has precharged successfully?
Q3: How do I get these parameter changes to "stick" after reboot?

I don't have a contactor attached. I tried both powering up 30v before turning on the 12v supply and after. I have seen people using a precharge resistor, but I'm assuming that isn't necessary, since my power supply is current limited.

Under the assumption that the message could be stale, I moved ahead and set fslipspnt to 10 Hz and ampnom to 100%, then pressed the 'start inverter in manual mode' button. I don't have the resolver hooked up (this is with a LEAF motor), but I'm assuming it will at least try to spin in open loop mode. Pressing the 'display error memory' button, I now see the message "[18099]: STOP - OVERCURRENT" As I understand, this is a catch-all message tied to the pin coming off of the Toyota hardware.

Q4: What power supplies are y'all using for testing?
Q5: What sequence do you usually use to apply high voltage?
Q6: Does anyone have any insights as to why this may be failing?

Here's a picture of my setup for reference:


Thanks!

Having run into what sound like the same issues as you, here are some things that I would advise, which worked for me.

Don't try and ignore an error and move forward. It won't work. You'll need to fix your precharge problem first.

Wire up an ignition and start switch. I used the actual ignition switch from my car. I found I needed to go through a lot of loops making changes, stopping/starting the inverter and seeing what the result was.

As for changing the parameters, any changes you make will be lost on stop/start unless you save them. So, if you set udcsw to 25V, save the params, stop/start the inverter, that should clear your precharge error.

As for the what order to connect the power, 'HV' first, then power on the inverter. When the inverter powers up, it closes the precharge contactor straight away - applying the HV. But since you're not using contactors yet, you'll need to manually apply the 'HV' yourself first. Of course, don't connect a real HV source directly like this or you'll blow the main cap in the inverter.

Good luck

chrskly

jeffb wrote: ↑Sun Apr 24, 2022 10:03 pm

chrskly wrote: ↑Mon Mar 14, 2022 7:49 pm I'm no expert, but I have two other variants of this board so I might be able to help out a bit...

...
Q4: What power supplies are y'all using for testing?
Q5: What sequence do you usually use to apply high voltage?
...

Thanks!

Q4: What power supplies are y'all using for testing?
I'm using a series of Main Well NDR-120-24 power supplies in series (they're isolated, so can be added in series/parallel by design). They're capable of putting out 5A each, and I have 3 wired together for 72V DC (and I have another, for a total of 96V, but don't need that yet).

Q5: What sequence do you usually use to apply high voltage?
Part 1 - Supply. The Arduino in the photo controls each power supply via a relay board - the main power switch turns on the 1st power supply, then there's a 2 second delay then the Arduino turns on the 2nd power supply, then after an additional 2 second delay the Arduino turns on the 3rd power supply. Finally, after another 2 seconds delay the three power supply outputs are connected in series, and the 72V DC supply goes 'live'.
Part 2 - Load. The 12V DC circuit connects first, and rudimentary checks are done. Once 12V DC power-up is successful, pre-charge HV connects, and HV checks occur. Once pre-charge is complete, the final HV circuit connects. Note - I'm running a Curtis controller at this point in time - my Toyota Hybrid testbed is yet to be built.


Regards,

dukbilt

Dylan Witt wrote: ↑Sun May 30, 2021 7:57 am guys im needing some help. can post on my page of the forum on my electric rx8 build. so i have a 100amp hour 47 cell li-ion preius prime plug in battery. i have re-arranged all the cells in this configuration. 47 cells, 100amp hours at 195volts dc full charge. my question, i dont have a charger and i know that with some help i can use the gen three inverter to charge my battery, but im only wanting to use my 115vac connection for testing, just for because of the limited amount of electrical wiring for 220vac charging cable needed, the real question is how would i connect my ac lines to my phase wires on mg1? line hot to phase three, ground to phase 1 and neutral to phase 2? or does the orientation even matter since it is running at 60hz.

Did you ever figure this out? I'm in the same boat

Not yet, drove around today 22 miles and the car pulls about 70 amps while driving. From a 75 amp hour battery

Dylan Witt wrote: ↑Mon May 30, 2022 6:14 am Not yet, drove around today 22 miles and the car pulls about 70 amps while driving. From a 75 amp hour battery

so, could anyone please verify something for me, is it normal for the inverter to draw 77 amps at 60 mph, 235 volt battery. im getting cogging in reverse but can spin the tires going forward. i might be in between the syncofs, im using the p314 hybrid transmission, i set syncofs to 21100 after all of my testing. that was before i added the large prius prime battery.

There is no such thing as a normal or abnormal current for an inverter / drivetrain to pull. It all depends on the conditions of your vehicle, build choices, terrain etc etc.

Dylan Witt wrote: ↑Wed Jun 01, 2022 9:50 amSo, could anyone please verify something for me, is it normal for the inverter to draw 77 amps at 60 mph, 235 volt battery.

Ballpark, yes.

77 amps @ 235v = ~18kw = ~24hp.

I didn't see what your donor vehicle is, but, it takes roughly 20hp for a smaller car to maintain highway speed on flat ground.

And just to do some dead reckoning, here's a calculator: http://www.enginuitysystems.com/EVCalculator.htm
(zero out acceleration to find steady speed)

I set your mass to a chunkier 1500kg.
I set your coefficient of drag to a less optimistic 0.35
I set your frontal area to 2 m^2 (typical car)
Velocity I left at 100km/h
Acceleration to zero.

... it spits out, 17.3kw, or 23hp for power. This is without any drivetrain or electronic losses.

That's 300 watt-hours per mile. That's about right for most cars.

Generally any car is going to be in this ballpark.

If you're going up a slope, most "normal" steeper slopes would about double your power requirements.

If you're accelerating (or braking), in non-racy casual driving, 0.05G is relaxed, 0.10G is pretty zippy. 0.2G is aggressive, might get you ticketed. Emergency slamming on brakes or drag racing (pushing the limits of traction) for most cars is around 0.5g, with top performers getting close to 1G. Since gravity (1G) is 9.8m/s^2, every 0.98 (~1) is 0.1G. So, for typical driving, acceleration of 0.5m/s^2 would tell you the added power needed to accelerate. For your car, the power reqs climbed to 38Kw ~ 50hp. Rough rule of thumb if you were picking a motor/inverter, figure out the power needed to drive highway speed, and then double it for reasonable acceleration, quadruple it for more fun acceleration.

Anyway, presuming you weren't climbing a hill or accelerating, 77a @ 235v @ 60mph, ~300wh/mile for a car seems just about right. Nothing amiss there.

jeffb wrote: ↑Sun Apr 24, 2022 10:03 pmQ1: Do I need to do anything special after changing udcsw to get it out of the error state?
Q2: Is the "STOP - PRECHARGE" error that I see the second time a stale message from before, or is it failing repeatedly? Is there any way to know if it has precharged successfully?
Q3: How do I get these parameter changes to "stick" after reboot?

I realize this part is from a 2 month old post...

I'm using the Gen 2 Prius, not the Gen 3, so I wasn't really keeping up on this thread. However, at the same time as these questions were being asked here, I was asking similar or the same ones in the Gen 2 thread.

What this highlights to me is that there's a real lack in the documentation for the OpenInverter interface, and setup in general. Lots of low-skill people are struggling with really basic stuff, only because no one's written it down or explained it yet. Everyone who understood it without seeing an explanation went onward, everyone who didn't got stuck or gave up.

I'm trying, where I can, to fill in any unknowns for the Gen 2 on the wiki, and I'd encourage everyone to continue doing the same here for the Gen 3, but, at some point the main OI documentation is going to need an overhaul too, and community volunteers to do that.

so, for the questions i had the past few weeks. i removed the V1c Logic board. im hopes that maybe something with the board would be the cause of this issuse. ( issue- when accelerating i get highamp trips. it keeps shuting down the inverter )
with closer inspection i checked to see if any of the circuits were bad or maybe they were shorted on the current sense leads. ?
or is the shorted leads the PWM for mg2?

Hello again knowledgeable people, been away for a while taking the shed and house off-grid with solar and my Ampera batteries, can anyone give me a clue about retrieving the following pickle ? I recently got (another) Yaris board ( the smoke got out of the old one) and built it up carefully, voltages all good, to the point of looking for pwm which I didn't find on any of the 6 pins R74-79. Had reasonable success with programming Wemos d1 wifi and spent a while getting my head arount STlink/Cube programmer but have somehow managed to press too many buttons and get the message "Warning ! core is locked up" . Is there a way out ? thanks , john

thornogson wrote: ↑Thu Jul 07, 2022 7:05 pm Hello again knowledgeable people, been away for a while taking the shed and house off-grid with solar and my Ampera batteries, can anyone give me a clue about retrieving the following pickle ? I recently got (another) Yaris board ( the smoke got out of the old one) and built it up carefully, voltages all good, to the point of looking for pwm which I didn't find on any of the 6 pins R74-79. Had reasonable success with programming Wemos d1 wifi and spent a while getting my head arount STlink/Cube programmer but have somehow managed to press too many buttons and get the message "Warning ! core is locked up" . Is there a way out ? thanks , john

Were the params loading in the web interface before you tried re-programming the board? Have you retried programming the board (after the error started showing up)? Where do you see this error?

Hi Christian , yes , originally the parameters were showing , led was flashing and so i gave it 12v on pin3 (FWD), set encmode to AB , fslipspnt and ampom testing params each to 1 and got the message 'inverter started' but I found no PWMs , (this may be me not good with my new hantek) , I diverted my attention to programming D1mini with some success and then buoyed by this small win attempted to talk to the STM32 with my chinese STlink , found the utility was legacy and downloaded Cube which, to my surprise worked out of the box . for some reason , late in the day I thought it was a good idea to reload stm32_loader.bin ..and this is the reason we shouldn't let 6 yr olds drive bulldozers . any help would be much appreciated !

thornogson wrote: ↑Fri Jul 08, 2022 5:11 am Hi Christian , yes , originally the parameters were showing , led was flashing and so i gave it 12v on pin3 (FWD), set encmode to AB , fslipspnt and ampom testing params each to 1 and got the message 'inverter started' but I found no PWMs , (this may be me not good with my new hantek) , I diverted my attention to programming D1mini with some success and then buoyed by this small win attempted to talk to the STM32 with my chinese STlink , found the utility was legacy and downloaded Cube which, to my surprise worked out of the box . for some reason , late in the day I thought it was a good idea to reload stm32_loader.bin ..and this is the reason we shouldn't let 6 yr olds drive bulldozers . any help would be much appreciated !

Oh, hmm, I've not programmed a board from Windows, just Mac or Linux. I'm not familiar with Cube, so can't help with the specifics of that.

But it stands to reason that if the error started after reprogramming, then something went wrong with the reprogramming. I'd suggest trying to reprogram the board again with the STlink and double check that you're using the right file and the right settings.

You may have managed to set some of the memory lock/protection bits on the STM32 chip. Only way I know of to clear these is to do a FULL chip erase. Note - this will erase everything on chip but if you're going to reprogram it anyway that shouldn't be an issue.

since the Prius gen 3 has max voltage of 650v, what's a safe full charge voltage to work with to account for any voltage spikes and not wanting to wear out the components too fast.

chrskly wrote: ↑Sat Jul 09, 2022 10:50 am
Oh, hmm, I've not programmed a board from Windows, just Mac or Linux. I'm not familiar with Cube, so can't help with the specifics of that.

But it stands to reason that if the error started after reprogramming, then something went wrong with the reprogramming. I'd suggest trying to reprogram the board again with the STlink and double check that you're using the right file and the right settings.

Thanks Chriskly ,Domt177 I've invested in a 'propper' ST-Link programmer and will set to this task as soon as my daughter's wedding is done ! Reality keeps getting in the way, still, it will soon enough be garage weather! Thanks for your comments

Update ! @Pete9008 , full erase , partial success, I couldn't leave it alone and started again , can now see the inverter management console by wifi but it's not all there, will not load even the freshest stm32_foc.bin , the 'waiting' icons just keep going round and round and the led2 on the board is not flashing...... any ideas ? .. cheers ,John

Just wondering what throtcur everyone is using, no one seems to use the parameter library (shame, please submit https://openinverter.org/parameters/) and how many are in use in running vehicles, so I created a quick poll.

I'm set to 4.

https://www.surveymonkey.co.uk/r/JWZ2Q9D

thornogson wrote: ↑Thu Jul 14, 2022 4:39 pm Update ! @Pete9008 , full erase , partial success, I couldn't leave it alone and started again , can now see the inverter management console by wifi but it's not all there, will not load even the freshest stm32_foc.bin , the 'waiting' icons just keep going round and round and the led2 on the board is not flashing...... any ideas ? .. cheers ,John

I've not got my inverter going yet so am guessing here but what you're seeing sounds a bit like this https://openinverter.org/forum/viewtopi ... oad#p36896?

Has anyone got a Pinout for an unmodified Prius inverter connector, all I can find is the replacement connector Pinout. (Ampseal etc)

Edit: found it here: https://openinverter.org/forum/viewtopi ... 941#p24941

Really struggling with resolver feedback and a HIRESOFS error while going through the FOC tuning steps. Can someone confirm I have this connected correctly? Using jumper wires from the 34 way IDC to the factory Prius connector.

andrewjenkins34 wrote: ↑Mon Jul 25, 2022 3:44 pm Can someone confirm I have this connected correctly?

According to the last image, you connected Enc1 to Cos. I think it needs to be Enc1 to Sin and Enc2 to Cos (the labels are correct, the text right beside it is not.

So then it should be
Prius: Desc: 34 Way:
1 Ex + 21 MG_2EXCA
2 Ex - 18 GND
3 Sin + 16 MG2_ENC_1
4 Sin - 20 MG2_SINA
5 Cos - 19 MG2_COSA
6 Cos + 17 MG2_ENC_2

Does that look correct?

Pleased to announce the first batch of boards in years with almost all smd parts placed has just arrived:) Hopefully a sign of things to come.

ordered new board, still awaiting shipping tracking number. glad to see some hope for the community with new complete boards

Shipped last week. Sorry for the delay.