About two weeks ago my Yaesu G-5500 rotator controller developed a strange fault. The Azimuth (AZ) meter would not display correctly, the Elevation (EL) meter was working fine. The AZ meter showed only ~20 degrees but the EME antennas were pointing at 213 degrees. Switching the controller off caused the meter needle to return to 0 degrees, the needle did not appear to be physically stuck.
This sort of fault normally points to the rotator itself (most likely internal 500 Ohm potentiometer) or the cabling. I was fairly confident that in my case the problem was within the control box as my K3NG Arduino controller was displaying the correct readings for both AZ and EL. It works in parallel using the same control voltages.
A quick Tweet about the fault resulted in a number of useful paths to follow to try and fix the problem. Close examination of the circuit diagram showed that the AZ and EL circuits were basically identical so I was fairly confident that troubleshooting would be straightforward. I suspected that the problem would most likely be a dry solder joint.
With the control box opened up, it was fairly easy to check voltages at various points and with everything switched off and the circuit board unscrewed from the meters, check for any obvious solder dry-joints. Everything looked fine, clearly without remaking each soldered joint I couldn’t be 100% certain but certainly none looked dull or had an obvious matt finish. I reattached the circuit board to the meters, switched on, and both meters read correctly! I switched the control box off and back on again a couple of times and we were back to the ~20 degree reading… clearly something was intermittent.
I rechecked the tightness of the meter screws and suddenly the meter was reading correctly. I then removed the four meter screws and cleaned both the meter terminals and the surface of the circuit board where the meter terminals made contact. The board was reattached to the meters and checked. They read correctly first time and have been working correctly since.
In my opinion there should be a star-washer (or similar) between the end of the meter terminal(s) and circuit board to ensure there is a good connection. I didn’t have any available at the time but should the fault occur again I will be certain to fit them. In saying that I can see that it will be a real challenge to do so, which is most likely the reason for their omission at the time of manufacture.