DIY FFB Steering wheel (MMosFFB ) (In Progress)

you need a 350w or bigger 24v motor = the MY1020 is a good design - search for "Cousin of OSW" for all the info you need

Hi Fernando

I have converted a broken Thrustmaster T500 using the Discovery board and an IBT2, running MMOS. I have a problem where the wheel has friction/resistance in addition to the force feedback. It feels similar to how the MMOS 'friction' setting should work on a normal wheel. This extra resistance is not the gear mechanism's friction - I know this because when power supply is off, the wheel becomes 'free' to turn. Do you know what is causing this?

I've tried one other wiring configuration with MMOS in PWM & Dir mode, which kinda worked: PWM (Discovery pin 9) to ENR & ENL and then Enable + (pin 11) to RPWM, enable - (pin 10) to LPWM. This method has some weird characteristics - It seemed to cure the resistance but added a huge FFB nul zone - no movement until at least 50% pwm (I could still hear the pwm noise at all pwm levels.)

I've been researching endlessly to find a way to allow the ffb to work without the unintended friction but I can't figure it out and I need help from someone more experienced with electronics - I hope you can help!

Many thanks.

I understand this frictional force using IBT in my Arduino project. The counter-energy generated by the motors and an active activation R (or L) seems to have this friction effect.

I have not had the opportunity to use the STM32-based design, but check the PWM frequencies available for this 2-PWM and 2-Enable case. This dead zone in the FFB is usually caused by the frequency with which the H-bridge or motor doesnt react.

Hi !
I am trying to do the same setup with the two dc motors of my G27. The game (pcars2) detects the position of the wheel (although 1° in reality translates in 5° in the game). I checked a couple of time your schematic (and others), but I don't get any feedback. To make things easier, I am trying to powering just one motor for the moment.
I think the problem could come from the mmos settings, could you say me if they are good ?
And do we need to connect A0 to VCC to enable the motor or something ?
Thanks in advance.

First, test by the wheelcheck program (cycle effect).
the outputs.
https://www.dropbox.com/s/3zxf5yd1bvujyi5/WheelCheck_1_72.zip?dl=0

This h-bridge needs 5vcc and gnd power,
and the input enable A, enable B, and PWM. (connecting according to the image you reply)

If it doesnt work, check with a multimeter (or LEDs) to understand the behavior of the outputs (PE09, PE10, PE11) to determine problem, if is configuration or h-bridge.

I tested wheelcheck, it gets everything (wheel position...) but when I test outputs, nothing happens...

I found that one cable (direction) had one bad soldering. Now when i turn the wheel, I got a little vibration (the wheel is shaken, but don't move at all). The dir led blanks very quickly.

The wires with that connectors can also couse issue.

Ok the problem came from the tension divider bridge...
I use the logitech G27 PSU that is 24v/1,75A, so I used a divider bridge and it looks like there is a problem with it because if I plug the 24V directly it works...
Now I need to rebuild everything and try again.

I guess we can't use the 24 V psu with the moto monster ? Maximum is 16V right ?

VNH2SP30 max voltage is 16v.

You can use an L298N bridge (2A), overheat a little (i tested g27 motors), but if you can use a forced ventilation it works.
Or IBT2 (bts7960), very used in these projects because of its power.

I've found a way to stop the non accurate friction Force-feeback has improved massively. It means the H-bridge never lands on the ground loop configuration caused by having R_EN and L_EN always at 5v.

Mmos needs to be set to 'PWM & dir', 16.8khz. Put pin 9 to L_EN and R_EN. Pin 11 to RPWM and pin 10 LPWM (depending on the polarity of your motor, you may need to reverse pin 10/11)

In this configuration 50%pwm=0motor power; i'm not sure why this is yet - but it is true, at least for me. The work around is to set the in game force feedback minimum gain to 50% - I will find a more permanent work around using hardware but for now it works just fine. In addition to the in game FFB minimum, use the mmos config's minumum gain setting for fine tuning (around 1% up to 5% normally). Tested for over 40 mins and the motor was no hotter than it was in the old configuration, might even be running cooler.

As an experiment I've removed the small., 300uf capacitor ( it sits on the supply rail between positive and neutral) from the IBT2 and it works fine. However, I sense a bit more electrical 'chatter' from the components- if that's the right term . Anyway, I think it acts as a filter for very high frequency voltage spikes caused by the PWM switching under load - i'm pretty sure the small capacitor should be left in place.

The following paragraph hasn't been tested yet, so please only try it if you know yourself the implications are of what you're doing - capacitors are potentially very dangerous:
In addition to the small capacitor, I think a large 2000uF+ capacitor in parellel would further help regulate the supply voltage. Hopefully it means we can turn the wheel with or against the motor, and the supply rail maintains voltage.

Lastly, in order to avoid the high in game minimum force setting, I will try to design a simple pwm post processor to make ...... <50%pwm=0%pwm ....... 50%pwm=50%pwm ........ 100%pwm=100%pwm ....... so that we don't have any unnecessary switching of the mosfets. This should help cooling and a few other little things too unimportant to detail.

My build doesn't work :-(
Pin 11 10 act as dir pins, they alternatively light a LED.
Pin 9 acts as pwm pin, it makes the led brighter as I put a stronger force.

But when I plug it to the moto monster, nothing happens. If I put pwm and dir pin to +5v, the motor turns, so power suply line is ok. I don't know where is the problem :/

Later I will try to plug dir and pwm pins to an arduino, and then the arduino will send the signals to the moto monster, I hope it will works better...

It looks like the moto monster doesn't react to pwm signal*... The only way to make it works is to send directly 5v to the pwm pin :-(
*I linked the PE9 (PWM pin) to a LED, and it works like it's supposed to be.
PS : I also got the working led's on the STM32

Could you do perhaps do some pictures of your wiring setup and screenshots of your software setup - It's not possible to figure out your problem without more information.

I put pictures of the setup and wirings a few posts upper.

Ahh sorry I didn't see that. You need to turn up motor max force in Mmos settings to about 100%. And put spring to 50% (when you're in game, turn spring completely off)

Like this ? (thanks for taking care !)

Looks good to me. I see that you've previously ran 24v through the motomonster - it is potentially damaged. I would really recommend getting some female/female patching cables - this will save you so much time and effort finding the correct wiring. Because you're using 2 motors, make sure they're not running against each other. It would be best to disconnect one of the motors until you get one working properly.

Yep I used 24v/1,75A. The strange thing is that it works well with this psu and arduino, but not with stm32... I ordered a 24v/12v converter, I will test it on an other moto monster that is brand new...
By the way, I would like to use a 24v/15A psu, can someone advice a good h-bridge that would fit this psu ?

to use a 24v/15A psu, can someone advice a good h-bridge that would fit this psu ?

All the same BTS7960. But they are 10A on a constant draw current is ok, even though stated 43A on a pick, but that only for a short time. Can double, triple em. As ussually.
(I guess was confused wtih that BS Sim info on stall current up to 85A with 24V15A power supply with MY1016 motor etc, but it seems to be wrong, to double the drivers. i.m. no 85A with that power supply. Even though, somewhere, was also the description the stall currents is emerged from the nominal power supply of the motor...whatever.)

Any way, you need to find, if use BTS7960, the original one. That China copy can be burned down soon.
I also whould like to know about what can be used, in alternative. Or solder the "own one" with good FET mosfets. )

Thanks ! I want to fix the G27 first but then I will try to build a new wheel with stronger motors.