Late Night Repairs

When we arrived in Malta on Monday, the LCD in one of the control boxes for the VideoRay was broken. (It was working before we left). Since we have two control boxes, we managed to use the working control box with the VideoRay to continue exploring wells and cisterns. On Wednesday night, it was time to start analyzing the data to verify whether or not it was valid. Only after about 15 minutes, it became apparent through a sanity check that most of the depth measurements were invalid. (The invalid data sets had a depth of -17 meters). After discussing this with Joe and Chris, Joe emailed VideoRay about the problem. They responded within the hour suggesting we disassemble the control box and analyze it for loose connection or faulty components if we were brave enough.

Immediately, I brought out the tools and began tearing it down. During this breakdown process, Joe, Jane, and I began brainstorming about what the problem could possibly be. I decided to start with fixing the LCD screen on the broken control box in order to check if that control box was recording a depth of -17 meters. I began switching out components from the broken control box with the equivalent components in the working control box to consolidate the problem. After ruling out the LCD screen itself and the removable Atmel microcontroller, I melted off the epoxy on the PCB connectors of the LCD control board with a soldering iron and found that switching the LCD control board out with the other control boxes LCD control board solved the problem. When switching these boards, both of the control boxes began to function properly (as far as the LCD screen goes). After discussing this with Joe and Jane, we concluded that the problem must have been from the epoxied connectors since the both of the LCD screens worked after switching them out.


It was about 5am, Malta time, when we reassembled the boxes and calibrated the compasses for the next day. We then logged data using both of the control boxes separately with the VideoRay out of water and found that the depth reading was about 1 meter. We decided this value was better than -17 meters and we only needed to calibrate the depth sensor, which could be done in the morning. This was about the time we called it a night.