Fixing a Turnigy Basic ESC

As mentioned in my previous post, my old tricopter went for a swim in a lake a while back. I’ve just now gotten around to rebuilding the entire system, but in addition to making my old tri fly, I’d like to rescue some of the electronics that took a plunge.

The ESCs in question are some Turnigy Basic 25 A. All of them remained functional, but would heat up consistently under minimal load. I was still able to fly the tricopter (poorly), but my battery life was noticeably reduced.

The entire system was submerged in water, energized for about 10 minutes while I swam out to retrieve the broken tricopter. This lead me to suspect that the circuitry may have been damaged due to the crash, or subsequent contact with water. I was using a 3S battery at the time, which meant that about 11 V would have been present at the input terminals to the ESCs.

DSC_6867

One of the problem ESCs

The first thing I did to try and fix the ESCs was attempt to clean off the circuit boards with flux remover and a brush. This got the ESCs clean, but didn’t show any improvement in functionality. In order to verify that the microcontrollers on the ESCs were still functioning, I attempted to flash new SimonK firmware onto them. The programming completed successfully on all the ESCs, indicating that there were no issues communicating with the microcontrollers. I also probed every DIO pin on each device using an oscilloscope because I wanted to be certain that every MOSFET was receiving a control signal. I was not able to find any issues with the microcontrollers, so I moved on to testing the MOSFETs.

Usually when a MOSFET fails in a power switching application, there will be a short either between gate-to-drain or source-to-drain junctions, causing large amounts of current to flow through the device. This failure will usually cause the device to overheat, and in extreme conditions, catastrophically fail. It is also possible for the diode between the source and drain (circled in red) to fail as an open or short, causing some current leakage from the drain terminal to the source terminal and excess heat. In a properly functioning device, there will be about a 0.45 V drop across this diode, easily measured using a multimeter and the diode test function when the device is removed from the circuit.mosfet_diode

The MOSFETs used on this particular type of Turnigy ESC were manufactured by two different companies, International Rectifier for the N-Type and Alpha & Omega for P-Type. Additional information and datasheets for these devices can be found in the following forum link.

In order to properly test the MOSFETs, they must first be removed from the circuit. Before going through the trouble of desoldering every IC on each PCB, I decided to narrow my scope as much as possible. Since we know that a broken device will most likely conduct excess current, this should produce a “hot spot” on the PCB. I wired up a test motor to the controller and removed the heatsink covering the MOSFETs. I then measured each device’s temperature using an infrared thermometer while the motor was running. I noticed that there was a temperature differential of about 20 degrees on one of the MOSFETs, so I decided to remove it from the PCB and measure it.

DSC_6876

Testing MOSFETs with an infrared thermometer

Sure enough, I used the diode setting on my multimeter and verified that there was 0 V measured across the source and drain terminals of the MOSFET. I flipped the multimeter over to the continuity test mode and probed the same terminals; as suspected, I detected a short. I had another crashed and partially destroyed ESC lying around, so I scavenged an P-Channel MOSFET and used it to replace the shorted device. I apologize for the bad solder job in advance!

DSC_6885

Testing the internal diode of the MOSFET (I don’t normally solder a probe to terminals directly, but I needed one free hand to take pictures!)

After reworking the board and checking for shorts, I hooked up a power supply to the controller and tested it. Success! The board no longer has any hot spots and the motor spins correctly! For reference, the diode of a good MOSFET should read about 0.45 V ~ 0.50 V as shown below.

DSC_6890

Voltage measurement across the Source-Drain terminals of a working MOSFET

This method of diagnosing issues with MOSFETs may not be the best, or most accurate, but it got the job done! I don’t feel comfortable flying my main tricopter with these repaired ESCs, so I’ll put these controllers to good use on my bench test setup. I hope this writeup helps someone out there trying to troubleshoot ESCs as well as anyone looking to learn some practical debugging techniques! Good luck and let me know in the comments below if I helped you out!

Comments

comments