[Beta] - BackForceFeeder : feeder vJoy FFB pour racecabs

[hadeys]

Hello, I've been having some issues with my PWM2Happ setup. I'm trying to connect my Happ active steering assembly directly to my PC, so I'm not using any additional arcade driver boards. I'm also using Aganyte's Sega FFB Controller Hack. I've tried to read on the forums about how to properly wire and configure the boards, but so far have run into two issues. Firstly, rotating the wheel causes all three (X, Y & Z) axes to change in value. Second, I cannot get the motor to respond to any force feedback effects. Here's the steps that I've taken so far;

Software Configuration

• I've updated the Arduino Mega 2560 to version 20231227T002603Z-001 of the `SegaFFB + Pwm2m2 or Pwm2Happ` firmware
• I've installed VJoy version 2.2.1
• I've configured a new VJoy device with X, Y, Z axes 16 buttons continuous POV hat and all FFB effects enabled as shown in Gojirien's tutorial
• I've installed BFF version 0.7.0.102 & obtained the license
• I've configured my hardware in BFF with Target Hardware = "PWM_CENTERED", "Pin Mapping for FFB" = true, "Digital PWM" = true. Then I restarted the manager, committed EEPROM and reset.

Hardware Configuation

• I only have one control connected to the Sega FFB Controller Hack, which is the 100k ohm pot of the Happ Active Steering Assembly
• I've connected the 5v VCC pad and ground to the outer fixed terminals of the 100k ohm steering pot
• I've connected the center sweeper terminal of the steering pot the the "Steering" pad (pin A0 of the Arduino)
• I've connected the "tx" solder pad of the Sega FFB Controller Hack (pin D14 of the Arduino) to Rx0 of the PWM2Happ board, and I've connected the "-" solder pad (Ground) to the ground of the PWM2Happ board
• I've connected a 24v 15A DC power supply to the "+" and "-" terminals of the PWM2Happ board and verified with my multimeter that it is getting the full 24v.
• I've connected the "M+" and "M-" terminals of the PWM2Happ board to the leads of the Happ Active Steering Assembly's motor

When I rotate the wheel, I can see both in the VJoy driver software and in BFF that all 3 axes change in value, which doesn't seem correct. Additionally, I haven't been able to get the motor to respond to any FFB effects. I have tried running several different types of effects with Fedit, and though I can see that the effect is running through the BFF logs and diagnostics window, the motor does not turn.

Here's the BFF logs from one of my attempts at running a constant force feedback effect;
9:23:00 PM | IMPOR|    2.415718:[MAIN] Running as standard user, trying realtime priority
9:23:00 PM | IMPOR|    2.416239:[MAIN] Setting realtime priority failed, got: High
9:23:00 PM | IMPOR|    2.423637:[MANAGER] Program configured for FFB with vjoy
9:23:00 PM | IMPOR|    2.600328:[MANAGER] Changing control set for Default
9:23:01 PM | IMPOR|    3.646051:[MANAGER] Found io board on COM3 version=0.1.9.0 type=IO BOARD ON MEGA2560
9:23:01 PM | IMPOR|    3.646758:[MANAGER] Program configured for PWM_CENTERED
9:23:01 PM | IMPOR|    3.692763:[MANAGER] Initializing IO board
9:23:01 PM | IMPOR|    3.693070:[USBSerial] Performing IO board initialization
9:23:02 PM | IMPOR|    4.728377:[USBSerial] IO board initialization done
9:23:02 PM | INFOR|    4.728873:[MANAGER] Configuring IO board for pwmmode=5
9:23:02 PM | INFOR|    4.747261:[MANAGER] Configuring IO board for wheelmode=2
9:23:02 PM | INFOR|    4.764303:[MANAGER] Configuring IO board for pedalmode=5
9:23:02 PM | INFOR|    4.781562:[MANAGER] Configuring IO board for ffbcontrollermode=1
9:23:03 PM | INFOR|    4.826862:[MAMEOutputsAgent] Entering HiddenWindow Thread
9:23:03 PM | INFOR|    4.840597:[FFBMANAGER] [UNDEF] step 0   to [DEVICE_INIT] step 0
9:23:03 PM | INFOR|    4.844413:[FFBMANAGER] [DEVICE_INIT] step 0   to [DEVICE_RESET] step 0
9:23:03 PM | INFOR|    4.850877:[FFBMANAGER] [DEVICE_RESET] step 0   to [DEVICE_READY] step 0
9:26:04 PM | INFOR|  186.238677:[FFBMANAGER] [DEVICE_READY] step 0   to [DEVICE_RESET] step 0
9:26:04 PM | INFOR|  186.238736:[FFBMANAGER] [DEVICE_RESET] step 0   to [DEVICE_READY] step 0
9:26:04 PM | INFOR|  186.450447:[FFBMANAGER] [DEVICE_READY] step 0   to [DEVICE_RESET] step 0
9:26:04 PM | INFOR|  186.454243:[FFBMANAGER] [DEVICE_RESET] step 0   to [DEVICE_READY] step 0
9:26:05 PM | INFOR|  187.677784:[FFBMANAGER] [DEVICE_READY] step 0   to [DEVICE_RESET] step 0
9:26:05 PM | INFOR|  187.678902:[FFBMANAGER] [DEVICE_RESET] step 0   to [DEVICE_READY] step 0
9:26:19 PM | INFOR|  200.845773:[FFBMANAGER] [DEVICE_READY] step 0   to [DEVICE_RESET] step 0
9:26:19 PM | INFOR|  200.848578:[FFBMANAGER] [DEVICE_RESET] step 0   to [DEVICE_READY] step 0
9:26:22 PM | INFOR|  204.190546:[FFBMANAGER] [DEVICE_READY] step 0   to [DEVICE_EFFECT_RUNNING] step 0
9:26:36 PM | INFOR|  218.419004:[FFBMANAGER] All effects done
9:26:36 PM | INFOR|  218.419022:[FFBMANAGER] [DEVICE_EFFECT_RUNNING] step 0   to [DEVICE_READY] step 0
9:26:42 PM | INFOR|  224.353534:[FFBMANAGER] [DEVICE_READY] step 0   to [DEVICE_EFFECT_RUNNING] step 0
9:26:50 PM | INFOR|  232.037500:[FFBMANAGER] [DEVICE_EFFECT_RUNNING] step 0   to [DEVICE_RESET] step 0
9:26:50 PM | INFOR|  232.039369:[FFBMANAGER] [DEVICE_RESET] step 0   to [DEVICE_READY] step 0

Here's the wiring


BFF hardware config


VJoy showing the axes in different wheel positions (Centered, fully left, and fully right)




Any information or further debugging steps would be greatly appreciated. Thank you!

[Aganyte]

Can you connect potentiometers on accel and brake ? (5 Kohms if possible).

Can you also test your motor directly with 24v power supply ?

[hadeys]

Thanks for the advice on connecting both pedals as well, that fixed the issues with the multiple axes changing together.

As far as the FFB and motor goes, I reconnected the motor back up with my old Immersion Force Feedback board that I was running it from previously. With the Immersion Force Feedback board and the same 24v 15a power supply, the force feedback effects are working correctly.

When I run any effects in Fedit using the PWM2Happ board, the voltage reads 0 across the M+ and M- terminals and the motor doesn't turn. I can see 24v DC on the High Power solder pads on the PWM2Happ board.

I tried running effects from both Fedit as well as from the M2 Emulator, both of which worked correctly when I tested them with the old Immersion board.

[Aganyte]

I can prepare a program for you to inject into the FFB controller which will request constant torques (for example 5 seconds on the right, then 5 seconds on the left). This should allow us to know if the problem is software or hardware. What do you say ?

[hadeys]

That sounds great, thanks for all of your help

[Aganyte]

I'll do that's asap.

[Tathan]

Hello tout le monde,

En cherchant à optimiser mon setup, j'avais déjà eu l'occasion de remarquer que même minimisés, le BackForceFeederGUI consommait + de mémoire que le BFF.

Quand je mettais l'option "originale" du BFF qui consistait à démarrer l'outil en même temps que le PC, je tournais toujours autour des 45000K~70000K de mémoire.

Avec ce petit script qui lance le BFF en fond, je tourne plutôt autour des 12500~30000K.

Ce script est moins "pratique", car il ne fait même pas apparaître l'outil à l'écran, si vous voulez le fermer, il faut aller dans le Gestionnaire de Tâches. (Pas super pratique si vous êtes en train de faire des tests, et encore moins si vous êtes en train de monter une config)

Mais si vous le faites fonctionner au démarrage du PC avec lequel vous comptez jouer, il suffit de mettre ce script dans AppData\Roaming\Microsoft\Windows\Start Menu\Programs\Startup et c'est, pour l'instant du moins, la solution la plus efficace que j'ai trouvé!

N'oubliez pas que pour pouvoir avoir la priorité "en temps réel", il faut que le raccourci auquel fait appel le script s'ouvre en tant qu'administrateur!

https://mega.nz/file/qepyATSD#TCI349WXQXaiSi7o7TX9a52fpmXn3XuKIgriYem9KUw