Pixhawk total power failure in flight

I was flying my custom hexacopter with Pixhawk running Arducopter 3.2. This was my first flight after a previous crash took out a few speed controllers (I made repairs, updated firmware, and ran through the initial setup again). At first the hex flew exactly as expected. Alt hold, loiter, etc worked perfectly. Then I flipped CH8 on to start the autotune. The hex started to make the first lunge and then everything just shut down. All motors stopped and the hex fell out of the sky (luckily from only about 10 feet high, no serious damage).

Flight logs show all data collection just stopping in-flight, prior to the crash. So I’m fairly sure it was a power failure, not a bad motor, prop, esc, etc. Otherwise the logs look normal.

The Pixhawk appears to still work as everything looks normal when I connect to Mission Planner. Each battery cell looks OK. Battery cable tests OK with a voltmeter. I had two separate ESCs supplying power to the Pixhawk, one through power module connection, one through the AUX OUT pins. I double checked the parameters after the crash to make sure CH8 was still set to autotune. I can’t find anything else wrong with the hex.

Any idea what happened? What would cause a total power failure like that? And how do I make sure it won’t happen again?

Sounds like an intermittent short. Auto Tune’s pitch and roll actions can be real aggressive initially. You might try a props off/ power on bench test. Then try shaking it the same way and see what develops. Just my 2 cents.
SC

[quote=“entomophile”]I was flying my custom hexacopter with Pixhawk running Arducopter 3.2. This was my first flight after a previous crash took out a few speed controllers (I made repairs, updated firmware, and ran through the initial setup again). At first the hex flew exactly as expected. Alt hold, loiter, etc worked perfectly. Then I flipped CH8 on to start the autotune. The hex started to make the first lunge and then everything just shut down. All motors stopped and the hex fell out of the sky (luckily from only about 10 feet high, no serious damage).

Flight logs show all data collection just stopping in-flight, prior to the crash. So I’m fairly sure it was a power failure, not a bad motor, prop, esc, etc. Otherwise the logs look normal.

The Pixhawk appears to still work as everything looks normal when I connect to Mission Planner. Each battery cell looks OK. Battery cable tests OK with a voltmeter. I had two separate ESCs supplying power to the Pixhawk, one through power module connection, one through the AUX OUT pins. I double checked the parameters after the crash to make sure CH8 was still set to autotune. I can’t find anything else wrong with the hex.

Any idea what happened? What would cause a total power failure like that? And how do I make sure it won’t happen again?[/quote]

do you have retracts?
If yes, are they connected to Rx or to Pixhawk?
If to Rx, make sure you remove +5(red) wire from the RX-Pixhawk link and power your RX with a separate BEC (non-ESC) some retracts take 1.5amp @5v while moving so make sure BECs can handle 3amp pikes.
I had the same issue during bench testing and added a separate BEC just for the retracts.
If you connect retracts directly to Pixhawk, connect only signal wires and get power from a separate BEC. For some reason my pixhawk was dropping FMU voltage line below 3v even with the separate BEC attached to the servo rail when the retracts were active.

Fullscalepilot2, good advice. I’ll try that.

Artemen, no retracts or any other periods attached to pixhawk.

I’m going to try adding a zener diode to the power rail.