Another solution, but still requiring hardware mods, would be small resistors in series with both the ground and + lines to the pot. Min and max values of  0 and 1023 would then never occur unless the + or ground wire was disconnected from the pot and the software should recognize those as such and probably go to mid-range or to last value. A properly dimensioned high-value pull-up or pull-down resistor would also be required to handle the viper diconnect situation. For maximum reliability, all three resistors would have to be (surface) mounted on the controller PCB and would require 3 separate connectors (pins) to the pot. No sharing of ground and + with other items.

Finn

Finn Lassen wrote:

Assuming input to chip is high-impedance, a couple of resistors on the board a factor 10 higher than the pot in parallel with the pot making input to chip mid-range when wiper disconnects. Wouldn't handle the + or ground disconnect, though. More fancy current sensing devices on + or ground side to tell controller that a wire is disconnected is probably overkill.

Finn
 
Tracy Crook wrote:

Good question Perry.

 

  Answer would depend on the exact nature of the 'open' .  Open at ground end, hot end, or wiper.   Answers in order:  mixture goes full rich, full lean, wanders around at last set point.  The pot used is a high rel part rated for 500,000 cycles but of course anything can fail.

 

Most pot failures are 'dead spots' where the wiper goes open at a certain spot (usually near where it is used the most).  Mixture will drift at this point and proper pilot response would be to move mixture to a slightly richer setting, land, get the pot replaced before flying again.  Thanks for the mental nudge, I need to add this to the instruction manual.

 

   I have pondered various fail safe strategies for this scenario (for example, make the B controller fixed at midrange mixture) but so far have not been satisfied with any of them.  Anyone have other ideas to be considered?

 

Tracy