Hello fellow simracing enthusiasts.
After spending a lot of hours searching the web for answers or atleast some clues that could help me fix the problem i found a lot of interesting info on DFGT but nothing was particularly helpful, unfortunately, so i decided to post a thread here as a last resort.
I've had my DFGT for a few years, bought it used and it was in great condition (apart from the cracked main bearing case, which is a classic problem). It was working great and i never used the wheel to the fullest, never ran 100% FFB, never used FFB effects, always monitored motor temperature to allow it to cool down when it gets too hot. Then, a few months ago it started losing the center position by a couple degrees mid game, i recognized the problem but never got my hands on it to fix it, since it wasn't really a big problem for me (the reason why it did this was a faulty positive wire going from the mainboard through the wheel shaft to the wheel board). It was working like that just fine, calibration process worked as intended, despite the wire problem was present during every single calibration, but then one day the wheel lost its center position by 30-something degrees mid game and i decided to replug it to recalibrate it, nevertheless it was working just fine, it just wasn't centered properly. After i plugged it back in, instead of going to the right, then to the left and back to center, it went fully to the right and just kept on pushing the motor, despite the wheel being locked by the rack.
So, now the problem is, when i plug the wheel, it starts calibrating but only goes to one side and neither switches direction, nor stops pushing the motor in the same direction, thus it can't calibrate.
What i've already tried: 1) Checked all MOSFETS, all of them work. 2) Checked all transistors (most of them in-circuit, but still) they all seem to be fine, i get 0.6V diode drop where i should and no continuity elsewhere. 3) Checked the TVS diode, since in-circuit it showed 1.5V drop, but it out of circuit it has no continuity so its fine. 4) Checked the encoder opto-couple. I don't have an oscilloscope, so i can't be 100% sure, but simple voltage measurements show that the optocouple works as intended, both A and B out channels fluctuate between 0 and 4.5V depending on the light the reciever gets. What also adds to the point that encoder is working, the wheel will not turn on the motor at all if i disconnect the encoder from the mainboard. 5) Replaced LM393 comparator since there is no way to check it out of circuit, old one works too so it wasn't necessary. 6) Checked all capacitors and resistors for shorts\continuity, found no shorted ones and the values on all resistors match up with my measurements even in circuit. Electrolytic capacitors, even though they're not active components in this case, they're all fine and all have their capacitance. 7) Considering the earlier wire problem, especially since the wire was visually fully intact, i tested all other wires, they're all good. 8) Tried plugging it into a different computer (and reinstalling drivers on my main one)
The rest of the wheel works perfectly fine, pedals work, all buttons work and of course my PC recognizes the device.
I would really appreciate any help, thoughts and ideas on that situation, or maybe someone had a similiar or even the same issue as me and fixed it.
I traced the board in KiCAD, not 100% of the board but the H bridge circuit and what drives it is there. I'm attaching the .zip with the KiCAD project as well as the screenshot of the schematic.
What is also interesting and rather confusing, is that the board can drive the motor in reverse, it switches directions when i short pins 1 and 2 (Source to gate) on the AP9977 MOSFET. (schematic reference M1A AP9977).
What also concerns me is that the AP4951 MOSFET recieves full 20V on both inputs, S1 and S2 are shorted, i guess thats how it meant to be, but G1 gets full 20V and G2 gets around 10-11V all measured to GND, which is rather confusing since according to the datasheet, gate threshold is around 3V for this MOSFET so both transistors (its a doubled P-MOSFET) should be open in the same time which would put full 20V from the power supply on both pins going to the motor and so it shouldn't rotate in that case. If anyone here with a Logitech wheel (they all have more or less the same H-bridge driving circuit) could measure that on their wheel, i'd appreciate that.
Please note that the polarity of the motor on my schematic is reversed because of the modified gearing. This is mainly to those who might find and use my schematic later to troubleshoot their issues and has nothing to do with my issue.
TL
R
When i plug in the wheel, it only goes fully to the right during calibration, thus it can't fully calibrate, the rest of the wheel works fine.
After spending a lot of hours searching the web for answers or atleast some clues that could help me fix the problem i found a lot of interesting info on DFGT but nothing was particularly helpful, unfortunately, so i decided to post a thread here as a last resort.
I've had my DFGT for a few years, bought it used and it was in great condition (apart from the cracked main bearing case, which is a classic problem). It was working great and i never used the wheel to the fullest, never ran 100% FFB, never used FFB effects, always monitored motor temperature to allow it to cool down when it gets too hot. Then, a few months ago it started losing the center position by a couple degrees mid game, i recognized the problem but never got my hands on it to fix it, since it wasn't really a big problem for me (the reason why it did this was a faulty positive wire going from the mainboard through the wheel shaft to the wheel board). It was working like that just fine, calibration process worked as intended, despite the wire problem was present during every single calibration, but then one day the wheel lost its center position by 30-something degrees mid game and i decided to replug it to recalibrate it, nevertheless it was working just fine, it just wasn't centered properly. After i plugged it back in, instead of going to the right, then to the left and back to center, it went fully to the right and just kept on pushing the motor, despite the wheel being locked by the rack.
So, now the problem is, when i plug the wheel, it starts calibrating but only goes to one side and neither switches direction, nor stops pushing the motor in the same direction, thus it can't calibrate.
What i've already tried: 1) Checked all MOSFETS, all of them work. 2) Checked all transistors (most of them in-circuit, but still) they all seem to be fine, i get 0.6V diode drop where i should and no continuity elsewhere. 3) Checked the TVS diode, since in-circuit it showed 1.5V drop, but it out of circuit it has no continuity so its fine. 4) Checked the encoder opto-couple. I don't have an oscilloscope, so i can't be 100% sure, but simple voltage measurements show that the optocouple works as intended, both A and B out channels fluctuate between 0 and 4.5V depending on the light the reciever gets. What also adds to the point that encoder is working, the wheel will not turn on the motor at all if i disconnect the encoder from the mainboard. 5) Replaced LM393 comparator since there is no way to check it out of circuit, old one works too so it wasn't necessary. 6) Checked all capacitors and resistors for shorts\continuity, found no shorted ones and the values on all resistors match up with my measurements even in circuit. Electrolytic capacitors, even though they're not active components in this case, they're all fine and all have their capacitance. 7) Considering the earlier wire problem, especially since the wire was visually fully intact, i tested all other wires, they're all good. 8) Tried plugging it into a different computer (and reinstalling drivers on my main one)
The rest of the wheel works perfectly fine, pedals work, all buttons work and of course my PC recognizes the device.
I would really appreciate any help, thoughts and ideas on that situation, or maybe someone had a similiar or even the same issue as me and fixed it.
I traced the board in KiCAD, not 100% of the board but the H bridge circuit and what drives it is there. I'm attaching the .zip with the KiCAD project as well as the screenshot of the schematic.
What is also interesting and rather confusing, is that the board can drive the motor in reverse, it switches directions when i short pins 1 and 2 (Source to gate) on the AP9977 MOSFET. (schematic reference M1A AP9977).
What also concerns me is that the AP4951 MOSFET recieves full 20V on both inputs, S1 and S2 are shorted, i guess thats how it meant to be, but G1 gets full 20V and G2 gets around 10-11V all measured to GND, which is rather confusing since according to the datasheet, gate threshold is around 3V for this MOSFET so both transistors (its a doubled P-MOSFET) should be open in the same time which would put full 20V from the power supply on both pins going to the motor and so it shouldn't rotate in that case. If anyone here with a Logitech wheel (they all have more or less the same H-bridge driving circuit) could measure that on their wheel, i'd appreciate that.
Please note that the polarity of the motor on my schematic is reversed because of the modified gearing. This is mainly to those who might find and use my schematic later to troubleshoot their issues and has nothing to do with my issue.
TL
RWhen i plug in the wheel, it only goes fully to the right during calibration, thus it can't fully calibrate, the rest of the wheel works fine.
Last edited:
