I'm going to tap into it tonight. I'm really curious if I can get=20 access to the individual motor driver to get the low-level drive signal,=20 or if I have to take the signal off the higher-voltage motor wires. I=20 suspect the latter as mass consumer products tend to do this more. But another option has popped up (from one of my robotics-team students)=20 -- the audio data for crashing, skidding, etc are all in separate .wav=20 files in a directory. The thought is to edit the audio (using audacity=20 or similar) to remove the very low-level frequencies from the ones where=20 I don't want rumble, and increase or add low-level frequencies to the=20 ones I want to rumble. I can adjust the cutoff frequency on the=20 buttkicker. Need to experiment a bit to see if this really will work well. Cheers, -Neil. On 9/9/2014 11:19 AM, The Lightning Stalker wrote: > I'd say the easiest by far would be to open up the steering wheel and fin= d the rumble motor. Then use the voltage or PWM going to the motor as an i= nput to a PIC. A square wave sounds more like a job for the digital output= to me, then through a divider or whatever to get the right signal level to= the amp. It sounds like different frequencies generate different rumble l= evels, so maybe a varying frequency depending on the desired effect. > > ---------------------------------------- >> Date: Mon, 8 Sep 2014 15:49:47 -0400 >> From: picdude3@narwani.org >> To: piclist@mit.edu >> Subject: Re: [EE] Force-feedback signal? >> >> I'll add that I'm using a Logitech G25 force-feedback steering wheel. >> >> Cheers, >> -Neil. >> >> >> >> On 9/8/2014 2:44 PM, Neil wrote: >>> Hi All, >>> >>> Need some quick assistance for a project I'm taking to the Orlando Make= r >>> Faire this coming weekend... I'd like to add seat vibration to a race >>> car simulator, which is essentially a PC-based race car game, race seat= , >>> force-feedback steering wheel, pedals and gear-shift, but with real >>> gauges instead of the usual on-screen ones. For the curious, gauge data >>> gets dumped out of the game (LFS) to UDP, Python takes that and spits i= t >>> out the serial port, a PIC18F takes the serial data and converts it int= o >>> a digital stream with 6 parameters, and each of the 6 gauges on the das= h >>> will grab whichever piece of data they need from the bus stream. >>> >>> This is what I have for the seat vibration... >>> http://www.thebuttkicker.com/gamer2 . It's essentially a subwoofer >>> that's bolted under the seat, but there's an issue in that it uses line >>> level audio for it's input so connected to a speaker output, it rumbles >>> more (due to wind, engine and some road noise) as the car goes faster, >>> and that's out of sync with what's expected... ie: rumbling/jolting whe= n >>> the car goes off-road. The force-feedback signal that goes to the >>> steering wheel has the rumbles/jolts correct. >>> >>> I've connected the buttkicker input to a function generator and a >>> low-frequency (~30 Hz) signal produces a nice rumble... even a square >>> wave works well. But now I need to get the force-feedback signal to >>> modify the amplitude of a square-wave signal I generate. There are two >>> options I can think of... >>> (a) Grab/convert the motor signal (from inside the force-feedback >>> steering wheel) and have that modulate the square wave. >>> (b) Use acceleration/velocity data from the game (which I can get and >>> also dump out the serial port) to modulate the square wave. >>> >>> Any of you know about these force feedback wheels and can tell me if (a= ) >>> will be relatively simple? IE: can I just isolate and filter the motor >>> signal to get a signal whose amplitude varies with the force that the >>> game is applying to the wheel? That would then go into an op-amp >>> voltage-controlled amplifier to drive the butt-kicker. >>> >>> If I use the data from (b), I can have the same PIC generate an analog >>> output (proportional to how much rumble is required) and a separate >>> 30-Hz square wave. Both of those would go into a VCA also as in (a). >>> >>> I have only a couple days to do this, so any help in figuring out the >>> easiest path here or any other options would be appreciated. >>> >>> Cheers, >>> -Neil. > =09 --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .