Finally got to testing the controller. It's rough to be a low task on my priorities list. I was considering doing the test on a large dyno, but I'm batting about 0 for 4 on getting an RC ESC to survive the start-up of getting the very heavy dyno roller up to speed before they puke magic smoke, and I really wanted to see it have a fair shake at what it could do for sustaining easy to drive continuous power before that.
I'm fortunate to be friends with perhaps the most experienced and well equipped large scale RC motor controller testing team on the west coast. Here are a few pictures of some of the dyno-stands they were tinkering on when I arrived.
And a larger test running. Notice the big RC ESC's stuck under the front edge of the test stand.
We tried the controller on a number of different motors to find something it sync'd well with. Oddly, the larger Joby motors didn't hold sync well. For some reason the motor would occasionally buck or hicup a bit, and sometimes make a growling sound and shut down and need to be power-cycled to spin the motor again.
This was the largest outrunner I've ever got to play with. Feeling the unstoppable torque at lowest throttle setting on the large-pizza sized Joby blew it away though.
That's a room full of motor designers and motor control designing nerds. My kind of paradise.
When we found the JM1 motor was an excellent match for the controller (never lost sync, had excellent response and smooth torquey starting and relatively quiet operation (quiet for a trap controller, the motors are silent on sinus control). It also didn't show hicup'ing behavior while running the JM1, I'm not sure why.
Next step was selecting the best winding option to not over-load the controller for our first round of tests. I don't recall the exact kV, but I think it was close to 60-70kV.
The experts selected an appropriate load prop (I am not a prop calculator wizard like these guys) and mounted it up.
Then we mounted up the instrumentation and taped down the battery up top and taped down all the wires thoroughly, because the prop-wash from these things is like a tornado. The test stand measures not only input power to the controller, but also output power by measuring the reactive torque on the motor mount with slick strain-gauge setup and the RPM of the prop.
We did some baseline tests (sub 20amps, on a low impedance 40Ah battery at ~90V) to verify the instrumentation we wanted was working well, and that nothing was getting warm on the motor or controller (both stayed below ~35c running at 20amps, about 1800w for a minute).
Here is some video from the pre-run testing/tuning/confirming instrumentation etc.
[youtube]1OkKepn2toI[/youtube]
Now we start at just 30amps, then 50amps. It does fantastic. I interrupt the test early to check on it with the FLIR. It looks fairly cool from the surfaces you can see, which peak around 40degC (104F), pretty chill, and the fans were blowing slightly warm air. No issues for this power range, smooth control with no hicups or faults. We were all pleased and stoaked to see how it would do at step 2.
[youtube]LMCAB3o5Kw4[/youtube]
Then we began our second power handing test run targeting 100amps. It did do 100amps for longer than many RC controllers (you hear how excited we all were in the dyno room!), however as you can see from the plasma flashes I was lucky enough to catch on video it was not 100amp stable, though it did achieve it easily (not too many of the >12s RC ESC's can reach 100a before exploding.)
[youtube]J2ESeGcOVaw[/youtube]
We cut the shrink to see if by chance we could see what failed. The heatsinks immediately after shooting plasma weren't up to 100c. However, the fans did a good job of rapidly cooling them to like 40c again just a minute after it blew (the fans were powered by an external power supply that stayed running).
Looks like FET smoke to me.
I'm not going to be able to fix that easily to try again.
They had a few boxes of various other large RC controllers that performed similarly.
This amazed me, 12x TO247 FETs in an RC controller!! (I wish I would have tried to find the brand but I forgot to look) Wow!!! Sadly it was also blown.
Moar various large RC controller carnage box. These guys actually do this type of testing on a regular basis, and I was very fortunate to have them lend the time to help me get this tested as well as we were able to do.
I know this took a long time for me to get around to completing, I think some of you know what kind of hours I've been working, as well as taking an awesome vacation to Maui and getting married.
I made a point of doing the best job I possibly could with the testing for you to try to make up for it.
Thank you Bruno (I will send it back to you if you want to take it apart further), and thank you JoeBen for helping, and other folks on the amazing Joby team. You are living the dream my friend.
Enjoy life,
-Luke