Thanks to everyone who offered suggestions to this problem. We finally have it solved. Hope this saves someone else some time and effort. The Problem: The Autohelm 7000 with linear drive would work well for the first five to ten minutes, then doze off. Given a nudge of the wheel it would come back to life for a few minutes, and finally give up completely with the message 'Drive Stopped'. The computer worked fine. A cooling off period seemed to bring the drive back to life to repeat the cycle. The Journey: 1) A couple of calls to Raytheon produced the diagnosis that the problem was in the wiring between the autopilot computer and the drive (about 15 feet). The wiring was 10 gauge and 8 gauge was required. This was it, absolutely, no question, the wiring was overheating and eventually creating too much resistance to allow adequate current to the drive. I replaced the wiring with 6 gauge for good measure. Same problem persisted, no change. 2) Next call to Raytheon produced the diagnosis that the drive transistors in the control computer were failing. This didn't sound right, since it seemed unlikely that both transistors (left turn and right turn) would fail in the identical fashion. The computer was back from the factory three weeks and $300 later with a full new circuit board. Same problem persisted, no change. 3) Next call to Raytheon produced the immediate diagnosis that the brushes in the drive motor were worn. The way these diagnoses were so quick and not open to discussion made me begin to suspect that a dart board might be involved. Fortunately, dart number 3 hit the mark. Opening the motor revealed worn brushes which were barely making contact. A few minutes of heat build-up probably caused the brush springs to soften and the brushes to make poorer and poorer contact.
