MMOS controller with cheap servo drive and motor

Any improvement is well received

Now you are using the motor configured to 200 steps per revolution, surely if you increase this value the cogging will decrease, or I think, I do not have the hardware to check it.

I tried changing the value to 2000 steps per revolution, but it didn't seem to make much of a difference. I'm afraid the cogging comes from the hardware, there's always a bit of it when I turn the wheel, even when the drive is turned off. I guess that's the price you pay for buying the cheapest motor in all of China

请问有出过中文教程吗，很想尝试这个方案，望回复，谢谢。。。

See the author from China, so ask him a Chinese tutorial?
-from google Translate

I do live in China, but I am actually from Europe, so even though I understand a bit of Chinese, I will not be able to make a tutorial in Chinese, sorry.

Tutorials can be understood through translation, thanks to the author.

Hi @sikjar i've go through your pdf and was wondering can this method used with a servo motor directly with out the need for the arduino in between the stm32 and the driver. as i think (correct me if i am wrong) servo motor drives are compatible with pwm signals.

Sikjar is actually using a stepper motor and its corresponding driver, and these do not work with PWM signal, so use the arduino to interface between the step driver and the STM32 board

thast what i thought. but was not sure so asked. so using this method any servo motor and driver combo can be used alon with this stm32 board right ?

I do not have it very clear, I would say that it is not worth any servo driver, for the OSW they use the drivers of the mark granite devices, that are much more expensive than the drivers + servo Chinese combos.
I guess because they are not prepared to operate regulating torque.

No what he's doing is using an arduino to send configuration changes to the Leadshine driver via it's RS232 port, while the servo is running, then commanding the wheel to move using it's Step/Dir inputs.
Basically the issue with cheap servo drivers is that they usually only have Step/Dir input, or CW/CCW, which in effect is velocity control, however what you need for a FFB Steering wheel is Torque control, the more expensive drivers allow this.
The Cheap driver though basically has to control torque to do anything, it's just hidden from the inputs, in the configuration, you can tweak the way the current control loop works and effectively change the torque, which is what he's doing with the arduino in real time.
The driver/Arduino pair then looks like the more expensive drivers used by the OSW wheels and the STM32 (which is what they use for the USB interface of the OSW) drives it the same way.

You can only do this on drivers that have similar configuration, but almost every cheap EBAY servo is using a copy of one of the Leadshine drivers.

actually i'm from India and the simcube is not available here and importing it will most likely double its cost. so was think if i could use this stm32 board with the mige small servo and driver combo available in aliexpress.

I've never tried this hack, but I would not expect it to work as well as using a Ioni Pro with a Sim Cube or whatever else, the nature of the way it's working guarantees some latency on direction change, and I would expect some odd feeling while not cornering, because it's likely to hunt for center.

The other issue you'd have if you went this way, is you'd need a 19A+ driver to get the 20NM out of the small Mige, which means you'd probably be using a different driver than the original author, and assuming it has similar configuration, you'd have to reverse engineer the serial protocol, because they aren't documented and quite probably different.

those are exactly my concerns right now. i'm trying to contact the sellers for the user manual of the drivers to figure out if i can match the wiring from the stm32 board to the new drivers. Am i right in thinking that if i get the wiring to match then most likely the setup will work? or is there more to it(i know the fidelity is something that can't be confirmed) .

The Mige is an AC Servo motor, you might find it difficult to find a "cheap" driver, that will drive it, the granite devices stuff is really pretty well priced for what it is.
You'll need a driver that will drive the motor, supports the quadrature encoder on the Migi and ideally Torque, using PWM/Dir, if you can get that, it'd minimize the additional effort in getting something working, if you can't get the PWM/Dir part, then your back at using the Arduino, and there would have to be some documented/undocumented method to set the torque level on the driver that you could exploit.
I would assume the amount of work involved to get a working solution is significant, and this thread might prove to be nothing more than a source of ideas, and you could just be throwing your money away, if your not comfortable with that I'd consider just eating the cost of the granite devices stuff.

that's very well put. will check if there is a driver with the pwm/dir support. if not then will have to figure out a way to source a simcube one way or the other. anyway thank-you for you insight. made things more clear for me

@sikjar
Hi, would like to ask if the next use of the motor is also Leadshine?

I would say that your problem is the drive, maybe the original ES-D1008 or the ES-D808, I have already had a problem with the stepper motor being lost and no torque due to drive. I'm studying to do a project like that thanks for the tips.

I'm here studying your schedule and I did not understand some things.

1 - why do you collect 10 samples and calculate the mean?

2 - Why do you set the torque to 0 to make the motor spin?

3 - Do you set the torque after the engine is hard to turn?


In my point of view, the problem may be in the code, but I'm not sure I've never programmed for this specific case. I have a code idea that might work maybe you've already tested it and it did not work. If you want my help just say that I give you my idea and you test.