openinverter forum

Romale wrote: ↑Fri Oct 14, 2022 8:23 pm 3 parallels?? and what is the cross-section of one wiring? it is difficult to wind a good amount of cross section with very thick wires

I don't remember exactly anymore, it seemed to have a diameter of 0.67 мм

The winding is applied two-layer

yaroslav wrote: ↑Fri Oct 14, 2022 8:27 pm The winding is applied two-layer

very strange. 3 parallel wiring 0.67 and only 2 turns is only 12 wires in one groove with a two-layer winding. this is very little, there is actually an empty groove left. or did I get something wrong?

Romale wrote: ↑Fri Oct 14, 2022 9:46 pm very strange. 3 parallel wiring 0.67 and only 2 turns is only 12 wires in one groove with a two-layer winding. this is very little, there is actually an empty groove left. or did I get something wrong?

so there are several wires wound up there

continue.
I equipped the motor with a new bearing, which includes an A/B signal sensor, launched a power module from Nissan note-e power, but I have wild overcurrent and strange sound.

I am not at all strong in the work of asynchronous motors and I ask for help in the settings.

I tried swapping several phase pins, but at the same time I got a slow trembling rotation, as if the sensor a/b was not correct.

What did you use to cut that engine wall off? It looks very nicely done.

arber333 wrote: ↑Tue Dec 06, 2022 9:50 pm What did you use to cut that engine wall off? It looks very nicely done.

the most common grinder with a cutting disc with a diameter of 150mm

it was not in vain that I decided to use part of the native internal combustion engine. the perfect alignment out, I easily rotate the motor by the shaft with one hand with the gears turned on.

we continue! today, the engine has finally started adequately. there are unnecessary noises, but this is a slight misalignment, the motor is not fixed at all yet and its shaft rests on a bearing in the connecting plate. one time I even got the error current limit (I think when I assemble a high-voltage battery, I will have to play with the settings)
Today I conducted tests from 70 volts.
I noticed that there is no regen at all to reset the throttle.

the original sensor from the Tesla Model S is used as an encoder, the encoder wheel is a gear from the camshaft of the car. I ground it in a lathe to the point where the length of the tooth began to correspond to the distance between the teeth and wiped off the excess from its width.

so, let's continue.
after trying to configure the fweak, I burned the controller, I had to replace the power module.
but this was not enough, it turned out that part of the brains on the v3 board also died in an incomprehensible way and the controller stopped seeing the gas pedal signal.
I also replaced the whole board, getting a new controller in principle.

Next, I started looking for the cause of instantaneous overcurrent at high voltage and it turned out that asynchronous motors are wound somewhat differently than synchronous ones (I also used the winding scheme from the bldc motor), having deployed the third phase of the beginning and end, I finally got an adequately working motor with a very decent torque.
While the transmission and the motor are poorly fixed, I left a limit of 100 hertz, which gives me 1500 rpm of the motor, but in fifth gear it instantly accelerates the wheels of the car to 60 km / h. it is necessary to take care of the stove (-15 ° outside) and replace the tires with winter studded ones, after that you can go out to test on the road.
Now the system voltage is 220 volts and 116 ah. later I will add batteries up to 360 volts.

BLDC scheme
ACIM scheme as you can see, for the acim motor, it is necessary that all three beginnings of the coils have the same current direction

does anyone have any thoughts on how, instead of getting an overcurrent error, to make the controller simply not supply current beyond the set limit, but just work?

However, I have already received more than 80 kW of power from a 4 kW industrial motor! This can be considered a victory!

That is a superb achievement!
You can try catphish current control firmware, play with iaclim or reduce boost, fslipmax or increase fweak

and which version should I change the firmware to? I have 5.20 R-sine, I thought it was the last one.

johu wrote: ↑Sat Feb 25, 2023 7:31 pm That is a superb achievement!
You can try catphish current control firmware, play with iaclim or reduce boost, fslipmax or increase fweak

by the way, I have not met the iaclim setting field, but I noticed that many people in Tesla ldu have an ampmin value much lower than the default.
Thanks for the hints, I'll play with the settings again.
Now the limit is 200 hertz and the slightly poor alignment of the motor and gearbox gives a lot of vibrations, I'm still afraid to increase the speed. but I think I will raise the limit so as to have 6000 rpm.

wow i have seen your all videos this is very good achievement. carry on my friend best of luck

johu wrote: ↑Sat Feb 25, 2023 7:31 pm
You can try catphish current control firmware, play with iaclim ....

I tried to send a command to show the hidden setting of the iaclim, but received a response that such a parameter does not exist.

I slightly filled the whole thing with silicone, and this repeatedly saved me from accidental short circuits with small objects that fall anywhere in the car. I'll make him a more decent case later.

for such an arrangement of current sensors as I have in the controller, it is necessary to use NEGATIVE calibration of sensors and current limits. in order for the values to be positive, the following photo.


here is this settings file for this sensor arrangement, as in my controller. it works perfectly with them on the native Nissan engine:
this is the settings file for working with the acim motor that I use:

I used 3 electrolytic condecators with a capacity of 680 mf each with a voltage of up to 450 volts, and one 1 mf film to smooth out high-frequency interference. this will allow you to use a controller with a 360 volt battery.