Electrohub, Naze32, Crazy Throttle Response

[Daniel991]

Hi,

I've been flying RC planes for about a year now but this is my first multi-rotor build. I'm building the Electrohub in a quad X configuration using the naze32 board from FliteTest and the power combo from Altitude Hobbies. I've got it built and I followed the steps in the naze32 setup video (http://flitetest.com/articles/naze-32-board-setup). I followed the steps in the video exactly except that I did not check the box to prevent the motors from spinning up when the quad is armed. I used a piece of 0.25" latex foam rubber wit hdouble sticky tape on both sides to mount the control board to the upper plate. I thought this would isolate the control board from vibrations but maybe this is too soft and is causing problems?

I took the quad outside to maiden it. I armed it and advanced th throttle slowly hoping to bring it up into a hover 1 or 2 feet from the ground but with just the tiniest throttle input the thing shot up about 20 feet in the air! I cut the throttle but it didn't seem to be coming down just floating up where it was and starting to go out of control. A little panicked I flipped the arming switch to cut the motors and it came tumbling down to the ground but luckily only the zip-ties broke.

I thought that maybe I had set the minimum throttle too high so I lowered it to a setting about five clicks before the first motor started spinning. I also increased the "Throttle Expo" setting in clean flight by about 25% I think. I took the quad out and tried again but got almost the exact same result. This time two of the booms cracked. My PID gains are all at the default values. I've read that the over sensetive throttle may be due to a too high P value but there is not a P for throttle. I see Roll, Pitch, Yaw, ALT, and VEL. Should I reduce the P-gains on all of those by say 25% and try again?

Both flights were in acro mode. No auto-level.

Thanks,
Daniel

 

[makattack]

Hey Daniel,

Sorry to hear that you've had an eventful maiden of your first multirotor, but it sounds like it's easily repairable. I would definitely double check all the electronics/wires/etc, after your repairs. When people speak of too high a P value, it's most likely on the roll or pitch axis. I do find it surprising the default PID values for that hardware configuration you have would be too high. Did you also calibrate your ESC's following Eric's method of using the cleanflight GUI with the props removed?

 

[Daniel991]

I did do the ESC calibration as shown in the video and got the beeping sounds from my quad so I believe it worked correctly. Yes the damage looks easily repairable I should have it ready to try again tomorrow morning but I'm not sure what to try to change so that I won't get the same result on the third try. I will admit that I was in a hurry to maiden it before dark yesterday so i didn't take the time to balance the props. I will do that before trying to fly again but I doubt that would be the cause of this problem.

 

[Snarls]

Can you give us a screenshot of your Cleanflight configuration tab? I'm thinking something might be up with your minthrottle/mincommand values. Also I think a better idea for testing if you fixed the issue is to take off all your props and then run the motors. If they start up slow and you can increase their speeds with throttle then you know you're good to go. Otherwise if you risk having the same crash happen again, like you've experienced.

 

[Daniel991]

here are my configuration and PID tabs.

 

[NHS77]

I don't see anything too suspicious in those screen shots other than the low minimum throttle of 1064... on my Naze32 boards this is usually around 1150. However I had a similar experience with a 250 carbon frame running 4S with 6x4.5 props without any additional hardware on it. The thing was ridiculously light and overpowered that way and I had trouble getting it to hover. Hence I started messing with the the throttle curve in the receiver tab. I had guessed the hover point would be around 25% (instead of 50ish as usual), so I set it up with a flattened curve at 25%. The thing shot up as soon as motors spun up and I had to cut throttle to get it back down. In the end i used a linear curve (aka a line and it went back to being controllable.

My first Ehub build was fairly light without any add-ons like cameras, etc. and pretty powerful even on 3S. If you were running 4S I could imagine a similar effect where it's just too much power for the little weight.

Post a screenshot of your receiver tab and also double check the throttle end points on your Tx (should be 1000 / 2000)