I am converting a juiced crosscurrent x to a base runner. The old controller popped. I seems like the grin units work well with the 52 volt battery the bike came with. I am using the stock dropout torque sensor. It has an idle output of around .4v and will go up to 2v if I stand on the crank. I'm basically guessing with the software config. The bike works fine with the throttle but no amount of stomping on the pedals will get the pedal assist to kick in. Another note is the dropout sensor does have a cadence sensor in it as well. The average pedal speed is bouncing from +400 to -400 rpm when I pedal. I'm not sure if that is related to the bike ignoring the torque sensor value. Is there any other info I need that would be helpful?
Ok, so there's a couple ways you could do this.
The first is using the TS and throttle separately, like you're trying to setup now.
Does your BR have the option in Pedal Sensor Type for just the torque, with no PAS / cadence sensor? Or if there's a way to disable the PAS Sensor Configuration section's values so it ignores them? If not, you would need to have a working cadence sensor to get the BR to respond to the torque sensor.
If the cadence sensor is being detected only as that specific rpm, just as forward or reverse, my first guess would be that the signal being seen by the BR is not actually the cadence, but instead is a F/R line that happens to for some reason be modulated at a specfic frequency instead of a solid signal. I'll have to look up the specifics on how the TMM4 works (the last things I saw didn't show a cadence sensor, just the torque sensor, but I might've missed it).
If for some reason you can't get the existing cadence sensor to work, you could add a crank-mounted magnet ring and sensor. Shouldn't be necessary, but it's a possible fallback.
The second is using the TMM directly as a throttle, by changing the throttle range to match it. That means your regular throttle would not work over it's whole range anymore, it would have a much higher response for a smaller movement range, if you simply wire both outputs to the same throttle input (perhaps with a 1-10kohm resistor in series with each connection to ensure neither can cause a problem with the other).
Alternately for that method you could use a bit of electronics to "compress" the voltage range of the throttle down to match the TMM, or expand the TMM's range to match the throttle, and then send them together to the throttle input of the BR.
There's other methods too, like the Cycle Analyst, but that's more complication and expense, and shouldn't be necessary.