openinverter forum

So, On this forum, no one has finally launched the Honda Civic inverter, or I have not found it.

since the Nissan leaf gen2 inverter was secondary here to check the performance of the motor, I removed it from the title of the topic.

It seems to me that this is not done in vain, for small motorcycles and ATV it would be an ideal solution, for example, using the rear engine estima or Alphard.

Before racking my head with the modification of the Honda inverter, I decided to make sure that the prius mg 1 motor was working properly.
I cut it out of the total mass of metal and got a neat motor, dragged it home, connected it to the Nissan leaf gen2 inverter, loaded the settings for Nissan that work perfectly with the Nissan motor.
Then I changed the number of poles of the resolver by two and found my angle of displacement. now all the most incomprehensible, it can be seen in a short video.
it works fine in rune mode, but when it's just starting, there are a lot of strange stutters.

The second oddity is the +12 feed on the brake only reduces the speed, but it is not possible to stop it ( I tried it with my hands) , the motor also has its own idle speed, although in the settings the desired field value is -100.

Who has any thoughts on this?

I will add more interesting photos))

While this is happening can you capture the content of 192.168.4.1/cmd?cmd=json and attach it as text file?

Nice work, that is the IMA inverter?

johu wrote: ↑Sat Jun 25, 2022 8:17 pm While this is happening can you capture the content of 192.168.4.1/cmd?cmd=json and attach it as text file?

Nice work, that is the IMA inverter?

Yes, this is an IMA inverter. I would like to install it with this small motor in my four-wheel drive buggy to connect to front-wheel drive, connect it only in difficult road conditions.
Do I need to save the settings file exactly when it stutters itself?

johu wrote: ↑Sat Jun 25, 2022 8:17 pm While this is happening can you capture the content of 192.168.4.1/cmd?cmd=json and attach it as text file?

Today I took the Nissan inverter and took it to the garage, where I connected it to the Nissan leaf motor to make sure that the inverter itself was working. I changed the offset angle to 0 and the number of poles of the resolver to 4 in the settings, the motor worked fine and applying a high signal to the brake stopped the motor, as it should be. Also, the accelerator pedal reset also stopped the motor, as it should be. I used the same accelerator pedal and the same battery to guarantee to find the culprit of the poor performance of the prius mg1 motor.
However, when I brought the inverter with the battery home, the prius motor behaves as before.

Conclusion, it's not just about the settings, that this motor needs something special.

Can I change the polarity of the sine or cosine?

Has anyone used prius motors with a homemade inverter or with a nissan leaf gen2 inverter ?
If yes, what settings did you use?

I've found syncofs around 10000 for MG2 but I can only idle it due to a lack of driveshafts. But that works as expected.
Will dig deeper once I can actually drive

johu wrote: ↑Sun Jun 26, 2022 10:41 pm Will dig deeper once I can actually drive

hello. I was thinking, maybe the reason for this behavior is prius mg1 if there is no connection of the motor temperature sensor?

Eureka! I reversed the polarity of the sine and cosine, passed a new angle search and everything works fine! now I will assemble the honda ima inverter and conduct tests with it.

very nice!
always surprised how many combinations there are that just seem to work

johu wrote: ↑Mon Jun 27, 2022 6:59 pm very nice!
always surprised how many combinations there are that just seem to work

however, the weakening of the flow was not disabled by the specified setting ((

can I try something else?

Romale wrote: ↑Mon Jun 27, 2022 7:12 pm can I try something else?

yes try playing with syncadv

Today I continued assembling the Honda ima inverter. I don't know what will happen in the end and whether it will work.
we'll see soon.

I broke my head searching for the cause of this error. why can it occur? there is a weak description in the wiki, saying that this is such a mistake, but what are the possible reasons??

Tie the "reserved" JP2.17 pin to 12V. Gotta ignore that one in software

johu wrote: ↑Mon Jul 04, 2022 9:57 am Tie the "reserved" JP2.17 pin to 12V. Gotta ignore that one in software

thanks. I'll try it today. I will continue testing in the evening

HICUROFS2
now this error is constantly present on the Honda ima inverter.
I noticed that I couldn't calibrate the current and udc sensors in any way. perhaps I made some mistake in the manufacture of the sensor board v3. for example, it says dc/dc 24 volts to 15 volts, for some reason I used 24 volts to 5 volts, thinking that it was stupid to connect 15 volts of the touch board and 5 volts of the main board version 3. maybe I did it in vain?

I also had to remove a piece of the native board powering the drivers in the igbt, since two channels stopped working for an unknown reason. I replaced them with 2watt dc/dc from 12 volts to 15

in general, after studying the circuit of the touch board, I realized that the indicated dc / dc is erroneous. I went to the store and bought a new one with an input of 5 and an output of 5 volts. now a 5 volt supply voltage has appeared on the Attiny chip. however, I still can't pass the voltage calibration. performing all the described actions, I do not get the final result. whatever voltage the console allegedly did not show after setting up udcgain, when the input voltage is turned off, this indicator does not change

I also checked the output from the current sensors and it is equal to 2.48 volts from each. the output from the touch board to the main brain v3 is 1.64 volts, which is the norm and therefore it is not clear to me why there is an error (hicurofs2) in this situation?

Romale wrote: ↑Tue Jul 05, 2022 1:16 pm I also checked the output from the current sensors and it is equal to 2.48 volts from each. the output from the touch board to the main brain v3 is 1.64 volts, which is the norm and therefore it is not clear to me why there is an error (hicurofs2) in this situation?

Now I see the problem. v3 Boards have a 5->3.3V divider on board on the current sensor input. So your 1.65V is divided again, therefor hicurofs

With the voltage you'd have to check the analog signal that goes into the V3 brainboard and how it behaves as you change HV

johu wrote: ↑Tue Jul 05, 2022 3:07 pm
Now I see the problem. v3 Boards have a 5->3.3V divider on board on the current sensor input. So your 1.65V is divided again, therefor hicurofs

does this mean that I can directly connect the honda ima current sensors with a 2.5 volt output?