openinverter forum

Hello everyone,

This is my journey on converter ducati Panigale V4s to electric version and i need help.

Most of the parts is already onhand such as the following,

- Motor (5 poles IPMSM )
- Inverter ( Tesla SDU powerstage with Open inverter board )

But for make everything fail proof i still hesitant to connect the wires as ,

1st - i dont have the actual wiring diagram..
2nd - i am having problem with the encoder. since i am using different motor. ( can someone refer me a sensor that is ready and available for all markets to purchase )

Can someone help me.. for those listed above problem ..

Hope all keep in touch as the progress continues..

Interesting! Post some pics of the project

Correct me if I'm wrong, but wasn't the Panigale one of the first motorcycles where the engine is part of the frame? So if you remove the engine, the bike is in two parts?

It would be interesting to see a photo

iccie2 wrote: ↑Sat Jan 24, 2026 9:56 am Hello everyone,
...
Can someone help me.. for those listed above problem ..
...
Hope all keep in touch as the progress continues..

Photos please

For encoder I propose you use RLS encoder circuit with polarized magnet. You would have to make a mount for magnet on the rotor and mount for PCB on the casing side.
https://leafdriveblog.wordpress.com/2016/11/16/encoder/

https://www.rls.si/eng/rmf44-rmf58-onax ... al-encoder

I'm not sure if this post is genuine. A Panigale is a very attractive and sought-after motorcycle that stands out thanks to its low weight and exceptional, powerful engine. It is completely unsuitable for conversion.

arber333 wrote: ↑Wed Jan 28, 2026 10:41 am Photos please

For encoder I propose you use RLS encoder circuit with polarized magnet. You would have to make a mount for magnet on the rotor and mount for PCB on the casing side.
https://leafdriveblog.wordpress.com/2016/11/16/encoder/

https://www.rls.si/eng/rmf44-rmf58-onax ... al-encoder

Hi,

i have now encoder that can do 0-3.3v with mid of 1.65v. it is MT6701
- i found that is is putting out signal depending to input voltage,
- i did use SMD chip to 5v to 3.3v from Tesla SDU openinverter board to the encoder.
- on multimeter it is showing 0-3.3v depending to rotor or magnet angle.

but the once i go to openinverter board webpage it is not giving any angle signal on plot and guage settings,

what could be the problem .?

nkiernan wrote: ↑Sun Jan 25, 2026 4:44 pm Interesting! Post some pics of the project

Hi,

at the moment i dont have any good pictures but this is the sneak pic I can share.

Most of the parts are from ducati panigale v4s , some are upgraded to fit my needs.

-Modifier Brembo Rcs Corsa corta 17 and 16,
-Stylema Caliper set,
-Ohlins electronic Fork FGRT 231
-Ohlins Rearsuspension TTX electronic .
- 51t Rear Sprocket and 12t front sprocket
-Customized ECU and Suspensions ECU.
- Modified IPMS motor
-SDU openinverter
- Customized battery - 96s 50ah 750amps ( safe current )
- Electronic Throttle ( i replaced original the original throttle since it have some play)


hope i can make this run soon. solving the problems one by one

muehlpower wrote: ↑Wed Jan 28, 2026 10:48 am I'm not sure if this post is genuine. A Panigale is a very attractive and sought-after motorcycle that stands out thanks to its low weight and exceptional, powerful engine. It is completely unsuitable for conversion.

Hi,

as far as i know stock ducati panigale v4s can only do 200-210hp,

but the openinvert can do over 270hp i assume, plus it looks good and exciting vacation project for me. ..

above are just my calculated output , all are based on my assumptions until tested..

cheers.

crasbe wrote: ↑Sun Jan 25, 2026 10:41 pm Correct me if I'm wrong, but wasn't the Panigale one of the first motorcycles where the engine is part of the frame? So if you remove the engine, the bike is in two parts?

Hi,

Yes it is thru, and thats the main advantage why I choose the ducati panigale v4,

1.) i can easily fit any battery size without cutting the frame and playing with tig welding.
2.) Transportation - you can just easily pack is into small box within airport check in lagguage size limit , which i actually did.
3.) easier assembly and disassembly, you can just easily remove and attached the battery to the frame without using any lift or special tools, you can just easily fit a socket wrench and unbolt it , then just that you have 3 parts that can easily checked and maintained if needed,,
4.) most other parts of ducati v4 are independent from the frame so i can move and modified them at will.
5.) Motor size - can fit a larger motor if want to without modify that frame, i can also easy tune up the center of gravity of the motor without worrying future changes on frame and motor geometry.


hope i answered your doubts and share my insights,

Best regards,

Iccie.

iccie2 wrote: ↑Mon Feb 16, 2026 12:39 pm Hi,

i have now encoder that can do 0-3.3v with mid of 1.65v. it is MT6701
- i found that is is putting out signal depending to input voltage,
- i did use SMD chip to 5v to 3.3v from Tesla SDU openinverter board to the encoder.
- on multimeter it is showing 0-3.3v depending to rotor or magnet angle.

but the once i go to openinverter board webpage it is not giving any angle signal on plot and guage settings,

what could be the problem .?

What kind of sensor you are using again?
The way you describe it you are using polarised magnet on the end of your axle with sensor chip logging angular change rate. But the most important is what kind of output you are receiving on the other end. There are several outputs usually available:
- UVW hall sensor commutation output
- ABZ 90deg overlaping output
- RS422 binary serial output
- Linear raising edge output

What kind of output do you get on a scope when you rotate the motor shaft?

arber333 wrote: ↑Tue Feb 17, 2026 8:02 pm What kind of sensor you are using again?
The way you describe it you are using polarised magnet on the end of your axle with sensor chip logging angular change rate. But the most important is what kind of output you are receiving on the other end. There are several outputs usually available:
- UVW hall sensor commutation output
- ABZ 90deg overlaping output
- RS422 binary serial output
- Linear raising edge output

What kind of output do you get on a scope when you rotate the motor shaft?

Hi ,

I have PWM simulated signal, means it has only 1 output signal rather than 2 from sine and cosine encoder.
- i was thinking that the settings "single" can read it

what i have done so far
+ installed RMB29 chip from rls with UVW signal to emulate hall sensor signal.
+ ordered mlx 90380 from digikey, it is still 10 days away to arrrive.
+ tried running RLS rmf44 and some encoder i have with various signal output such - AB , simulated 3.3v single PWM, ABZ .
+ tried working and running the sensor to 5v to 3.3v using SMD chip.
1.) SIngle PWM encoder dont read by SIngle channel encoder settings from the web interface but SIn/Cosine Settings reading with over 80-90* degress angle since this settings requires 2 signal with 90 degress out of phase. ( correct/)
2.) AB and ABZ -throwing wierd trapezoidal angles from the angle monitor ang guages. \

what i am currently thinking are,
1.) is Tesla SDU drop in board really capable reading SIngle sensor?, if yes which sensor does it needs and which lines should 1 use from the wire is 1 signal wire required>?
2.) is the board capable reading UVW signal , if yes which settings.

Hoping to get a reply and help,

Best Regards.

Iccie

iccie2 wrote: ↑Thu Feb 19, 2026 7:04 am I have PWM simulated signal, means it has only 1 output signal rather than 2 from sine and cosine encoder.
- i was thinking that the settings "single" can read it
......

what i am currently thinking are,
1.) is Tesla SDU drop in board really capable reading SIngle sensor?, if yes which sensor does it needs and which lines should 1 use from the wire is 1 signal wire required>?
2.) is the board capable reading UVW signal , if yes which settings.
.....

Tesla SDU is ACIM motor and that only needs somewhat 60 pulses per turn which will only tell inverter how fast motor is turning. This is enough for OI sine firmware.

If you want to have rotor position as well you need two signals 90deg at each other - A,B signal. This will also give you direction of rotation. This is used with ACIM motor as well.

Now PMSM motor needs rotor absolute position and you have several methods:
- ABZ encoder which will send additional Z pulse to mark the 0deg of rotation.
- Sine Cosine resolver signal which can give you acceleration as well
- triangle raising pulse....
- UVW 6 step pulse what you will understand as digital Hall sensors

EDIT:
If i understand correctly you are using PMSM motor? Now This means inverter would require either Resolver signal or perhaps ABZ signals to successfully run PMSM motor with OI brain.
Now we have used Resolver signal with OI...


But does Tesla SDU replacement board implement resolver control?

arber333 wrote: ↑Thu Feb 19, 2026 1:58 pm Tesla SDU is ACIM motor and that only needs somewhat 60 pulses per turn which will only tell inverter how fast motor is turning. This is enough for OI sine firmware.

If you want to have rotor position as well you need two signals 90deg at each other - A,B signal. This will also give you direction of rotation. This is used with ACIM motor as well.

Now PMSM motor needs rotor absolute position and you have several methods:
- ABZ encoder which will send additional Z pulse to mark the 0deg of rotation.
- Sine Cosine resolver signal which can give you acceleration as well
- triangle raising pulse....
- UVW 6 step pulse what you will understand as digital Hall sensors

EDIT:
If i understand correctly you are using PMSM motor? Now This means inverter would require either Resolver signal or perhaps ABZ signals to successfully run PMSM motor with OI brain.
Now we have used Resolver signal with OI...


But does Tesla SDU replacement board implement resolver control?

Hi,

Thanks,

Tesla SDU drop in board have 4 wires for encoder ,
1 for negative, 1 for positive , and 2 for signal which Tesla originally uses.

At the moment the nearest option is to use sine/cosine encoder.

I have some question below,
1.) If I have UVW Hall signal sensors where should I put the 3rd signal wire?
2.) If I have ABA encoder where should I put the 3rd (Z) wire?

Can you have a reference picture of the board where I can put it?

3.) I was thinking that if I'll put 2 PWM sensors and of set them by 90 degrees to make them simulate sine and cosine signal? Got the idea where to put it but it seems on awkward location.

Best regards,

Iccie.

iccie2 wrote: ↑Fri Feb 20, 2026 2:32 am .....
I have some question below,
1.) If I have UVW Hall signal sensors where should I put the 3rd signal wire?
2.) If I have ABA encoder where should I put the 3rd (Z) wire?

Can you have a reference picture of the board where I can put it?

3.) I was thinking that if I'll put 2 PWM sensors and of set them by 90 degrees to make them simulate sine and cosine signal? Got the idea where to put it but it seems on awkward location.

1. Not gonna work. OI interface isnt setup to work from UVW inputs. I used Lebowski with that....

2. There should be a third input on the OI interface. But maybe not because SDU drop in board was prepared with Tesla ACIM motor in mind. Like you said you want to investigate your options with resolver signals. I suggest you see Wiki and your SDU board schematic. The way i setup my OI was to use Z channel (PD2) to excite the resolver winding and then i got return signal via SIN COS inputs which were channel A and B (PA6 and PA7) on the incremental encoder. This means you need 3 channels from your board; one output and two inputs... Tesla SDU only has 2 inputs...
Tesla SDU schematic
https://github.com/jsphuebner/Tesla-Fro ... ematic.pdf

OI rev 3 board with resolver schematic is there...
https://openinverter.org/wiki/Main_Board_Version_3

If you cant determine that or dont have free inputs i suggest you look for different inverter to use for your motor... or you can build your own exciter for resolver... i am not sure if it will work...Johannes?

arber333 wrote: ↑Fri Feb 20, 2026 7:15 am 1. Not gonna work. OI interface isnt setup to work from UVW inputs. I used Lebowski with that....

2. There should be a third input on the OI interface. But maybe not because SDU drop in board was prepared with Tesla ACIM motor in mind. Like you said you want to investigate your options with resolver signals. I suggest you see Wiki and your SDU board schematic. The way i setup my OI was to use Z channel (PD2) to excite the resolver winding and then i got return signal via SIN COS inputs which were channel A and B (PA6 and PA7) on the incremental encoder. This means you need 3 channels from your board; one output and two inputs... Tesla SDU only has 2 inputs...
Tesla SDU schematic
https://github.com/jsphuebner/Tesla-Fro ... ematic.pdf

OI rev 3 board with resolver schematic is there...
https://openinverter.org/wiki/Main_Board_Version_3

If you cant determine that or dont have free inputs i suggest you look for different inverter to use for your motor... or you can build your own exciter for resolver... i am not sure if it will work...Johannes?

Hi,

I'll be waiting for my sine cosine encoder, it think it should work as the sin alone can be read by sine/cosine on settings.

What bothers me is that,
Tesla is SDU OI is normally a sine inverter to drive ACIM right?

But is it ok if I change the firmware of SDU OI to FOC? Is there any contradiction with that if I am planning to run sine cosine encoder and PMAC or PMSM ?

Best regards,

Iccie.

iccie2 wrote: ↑Fri Feb 20, 2026 12:48 pm Hi,

I'll be waiting for my sine cosine encoder, it think it should work as the sin alone can be read by sine/cosine on settings.

What bothers me is that,
Tesla is SDU OI is normally a sine inverter to drive ACIM right?

But is it ok if I change the firmware of SDU OI to FOC? Is there any contradiction with that if I am planning to run sine cosine encoder and PMAC or PMSM ?

Best regards,

Iccie.

SDU is SINE firmware. If you want to run PM motors you need FOC definitely.

For your aplication you will use sine - cosine encoder and you do not need R1 excitation output your sensor board already signals two analog signals. You will need to intercept that using existing inputs that were marked A and B respectively...

arber333 wrote: ↑Tue Feb 24, 2026 5:06 pm SDU is SINE firmware. If you want to run PM motors you need FOC definitely.

For your aplication you will use sine - cosine encoder and you do not need R1 excitation output your sensor board already signals two analog signals. You will need to intercept that using existing inputs that were marked A and B respectively...

Hi,

Just to confirm, I only need to flash it to FOC firmware correct?

But will the hardware just fine to drive PMSM after programming it to FOC firmware. ?

Best regards,

Iccie

iccie2 wrote: ↑Tue Feb 24, 2026 10:31 pm ....
But will the hardware just fine to drive PMSM after programming it to FOC firmware. ?
....

No idea....
I guess i am implying you test it and take one for the team....
Probably you should use some 60Vdc and one small PMSM motor for testing...

arber333 wrote: ↑Tue Feb 24, 2026 10:45 pm No idea....
I guess i am implying you test it and take one for the team....
Probably you should use some 60Vdc and one small PMSM motor for testing...

Hi,

I just got my mlx30980 encoder.

But the problem is only 1 signal output is work.
And i am still using the 3.3v
Yet to try the 5v supply. But i was wondering it it normal?

As far as i know both should work with 2 output signal.?.

And also the inverter still not showing any angle position.


Beta regards.

Iccie

I was now able to get a decent signal.

It came out that the board that I have is not compatible or defective.

What i have done is I tried wiring my mlx90380 smd manually as I have spare parts as redundancy and it was now giving a decent signal even if I just hand holding the magnet and chip manually.

This style sensor is pretty sensitive to alignment and magnet strength which translates to the functioning air gap. Are you using the correct (diametrical?) magnet?
Do you have a warning light for too much / little magnet force?
Will the rotor spill some magnetic forces out where you'll have the sensor magnet and sensor? It may get flooded with magnetic fields at some power level / just sitting there.

jrbe wrote: ↑Thu Feb 26, 2026 12:27 pm This style sensor is pretty sensitive to alignment and magnet strength which translates to the functioning air gap. Are you using the correct (diametrical?) magnet?
Do you have a warning light for too much / little magnet force?
Will the rotor spill some magnetic forces out where you'll have the sensor magnet and sensor? It may get flooded with magnetic fields at some power level / just sitting there.

Hi,
I am kinda contain that it is centered, but I am having problem with the angle. It is only reading 2- 348 degrees. I am still searching to what I can do to fix this.


Any ideas?

It seems like Johannes is wrong about the mlx90380,
I checked the voltage output it was 0.5-2.8v on multi meter after having doubt, I finally check it on datasheet. And I found that the output range is not really 0-3.3v..