Testing a prototype under severe conditions I was able to get it to run backwards. It involved climbing a hill at full propel until the engine died, then rolled back some before the loss of hydraulic pressure automatically applied the spring applied hydraulic release parking brake. During one of my tests, the machine sounded strange, like the engine was running at low idle, and then stopped. I had instrumented this machine thoroughly. When I analyzed my data, I found it did run backwards for 7 seconds drawing fuel for the common rail supply pump from the return to tank. No problem with any of the 3 hydrostatic or 2 open center gear pumps on the machine. I am at a total loss to explain your control reversal. My instance was a hydro with a closed circuit so it always was filled with oil less leakage. If yours has a gear pump running backwards it could possibly draw oil from the circuits and then supply it to - the hydraulic reservoir where it isn't going to do anything. Wicked engine gyrations as it dies can cause problems. The engine I described above - the next day it spit its innards on the ground. Our engine design group couldn't believe my data so they duplicated it on a dyno and found it possible including the fuel supply I described above and found running backwards, the injection timing was so far off it put the strain on the rods that ended up ventilating the block so a sense was added to the crank to detect engine reversal and cut off the injectors. Could the engine gyrations while yours died have cracked some components in the cam/FIP gear train, or in the pump to crank coupling?
