How do you make the DFGT spin faster? I'd like to do a mod similar to this.
It's not completely straightforward if you want to do it properly. Ideally you'd build an entirely new wheel and only take the DFGT's logic board, but if you can find a replacement higher power motor that fits and it doesn't strip the feedback gears, it can be done using the DFGT's chassis.
To make a motor spin faster you need more voltage, but that can only be done if the board components have enough headroom. The other possibility is changing to a different motor wound for higher RPM, but that usually draws more amperage, which may also exceed the board components' headroom.
I think the components you're referring to are the FETs, I think they're probably MOSFETs. It's been a while since I looked at my DFP's board (because I lost it) but if I recall correctly (though I may be getting mixed up with something I read about the G27) there are two surface mount dual FET ICs that interface the microcontroller with the motor. A FET (field effect transistor) is a voltage-controlled electronic switch, in theory if you got similar FETs, i.e. with similar gate voltages, maximum switching frequency, etc. but with higher power handing ability (probably something in a TO-220 package or bigger) you could switch whatever you wanted; speakers, stadium lights, an air horn... I'm not sure what use any of that would be, but you totally could do it.
Anyway, Google 'full bridge motor driver', that circuit is (if I remember correctly) basically what the DFP and G27 use; one pair of FETs to turn the motor one way, the other pair to turn it the other way. If you look at the two small ICs nearest the motor's connection to the board you should be able to read the label on them to find the datasheet for the FETs inside them, then you can use the spec on that datasheet to help find good replacements. Also look into 'pulse width modulation', I'm not 100% on this (I don't have an oscilloscope so I can't test it for myself) but I'd be surprised if the microcontroller in Logitech's wheels don't use PWM to control the feedback of their motors. PWM is basically a method of getting 'analogue' signals from a digital source; you have a rectangular waveform with a regular period (frequency), but the duty cycle (ratio of 'on' time to 'off' time) changes from 0% to 100% to give the effect of an analogue voltage. I'm guessing here but I think the microcontroller sends the output as four PWM signals to the four FETs to control the feedback motors. Replace the FETs, turn up the current and/or voltage supplied to the uprated motor (where you'll find one that fits AND allows you to mount the optical encoder on it is beyond me, though, hint: look for a dual shaft motor) and you should, theoretically, be good to go.
Three practical problems, though:
1. What if the 12V input is powering more than just the motors? You'll need to work that one out yourself because I've lost all the research I did into mapping out the DFP's logic board. If the barrel jack connector does just go straight into the MOSFETs separately from the rest of the board you're ok, if not then you'll kill something else when you increase the voltage.
2. The added strain on the DFGT's feedback gears will shorten their life. Best to consider converting it to belt drive... Now there's a thought, it'd be quieter and smoother too. That might also solve the optical encoder problem since you could just mount it elsewhere, but then you have to design the pulley system so that it rotates in the correct proportion to the wheel.
3. Cooling. If you were to find a 48V 2A motor the FET will be handling almost 100W (I think, it's late and I'm tired). That's a lot of power, you'd need a heatsink for each FET and definitely a fan if you wanted to put it in an enclosure. Also maybe casts for your freshly fractured wrists. While you
could go that high, you'd have to be mad. I think it'll all depend on the motors you find, really, if you can find some 24V 2A dual shaft motors with the same diameter, length and shaft diameter then I'd start there.
There are probably cheaper, easier, hackier ways of doing it, but this is what I planned to do until I lost my DFP's logic board, failed to source suitable motors then lost interest and bought a G27.
. Everything you posted is good accurate information though... guess we'll see if he makes any use of it. FWIW, the FETs on the DFGT are hexfets made by IR IIRC, but I was unable to find anything related to the number on them. Also, I'm pretty sure they are using PWM, as Fanatec definitely is (Ive scoped mine) and I'm sure Thrustmaster is also since they're running a motor that stalls at about 9+ amps with something like a 5-6 amp power supply.
