Honda Mileage Device YAG VID

Just a quick note to say that I think changes I made last night to the VID firmware correct a slight problem I've been seeing for a few weeks now. As some might surmise, I have a vehicle simulator hooked up the device on the bench. I run this thing day and night-- accumulating hundreds of virtual miles each day. When the simulator is in the static mode, I expect to see the trip/tank averages remain constant PERIOD. However, I have recently noticed that the engine pulse width (that should remain constant) sometimes jumps about 50 percent several times per minute. This leads to a lowered average MPG. I checked and rechecked the incoming pulse width and could find nothing wrong. I then checked and rechecked the code--- also found nothing wrong.

Last night, it finally dawned on me that due to the way I was averaging pulse width readings in the injector interrupt thread, the main thread was sometimes disabling the injector interrupt before it could complete the last measurement cycle. So now by checking the completion status of the injector interrupt thread before halting the injector interrupt, the problem "seems" to be cured. While this appears to be the cure, I have now entered a new phase of regression testing to make certain that this change does not cause something else to break.