Hey guys,I' ve just build a RoboCat270 for some FPV speed fun. As I am a great fan of my APMs in my other quads I thought I should build it with a miniAPM. I added m8n GPS, OSD and telemetry link, as I really enjoy these things.The quad is powered by 4 EMAX 2204/2300 motors, running at 4s with 5045 props.At maiden flight I encountered fast oscillations on roll and pitch, so I tuned down rate roll p and rate pitch p to 0.030 to make it stable.To get throttle to about 50 percent, I had to put throttle hover down to 350.The plan was to let the quad fly in AltHold, invoke autotune and get the parameters tuned.Unfortunately, I can't get it to hover stable in AltHold. Whenever I try, it either lands, whobbles about 1 meter up an down or just speeds off into the sky.I played a bit with Althold P and Throttle acceleration, but to no success.Any suggestions would be highly welcome.Please see the pictures for details
I just tried Althold after getting rid of the vibrations.
It works!
Not perfect but I will get there after some tuning.
I would also like to inform you that in the testing process I tried two different props
Kingkong 6040 and Gemfan 5045 (both supplied by HK).
With the Gemfans my quad hovers at 36 percent throttle and 10.5 amps, while the Kingkong props need 5 % less throttle while using 2-3 % less current. I will stay with the Kingkongs.
Thank you very much for your comments!
I just found out that I was part of the problem myself.
I mounted the miniAPM as usual with gel pads as I do always and did not care about vibrations (it had always worked so far, and the vibe measurements in the logs were good then).
But this little quad just produces a hell more. So vibes were about 3times of the allowed level. Argh!
I reduced the gel pads now to 1/4 size (area) and placed a 12 g metal plate below the APM. Now I have vibe levels lower than I ever had :-).
Don't know yet if AltHold now works, as it is deep in the night, and if it speeds off into the sky...
But I will try tomorrow.
I think it's very likely to be high vibration. Copter-3.3.1 has improved vibration reporting but I think the miniAPM has the AVR cpu so I think you'll have to remain with Copter-3.2.1 so it's a little harder to be sure of the vibe levels but I'm pretty sure this is the issue.
I had the same problem last week. I eventually discovered that the craft would alt hold and loiter much better with the video Tx and camera powered off. I haven't yet had the chance to establish whether it was the camera or the VTx, or both, causing the issue. I did notice that in fast forward in alt hold it slowly dropped altitude and rose up again when I slowed down so I need to deal with that first. If you haven't yet done so I would recommend test flying with video switched off to see if your issues have the same cause as in my case.
I had the mini-APM in my 320 hexacopter. i had the same issues you described. After much internet and soul-searching, I found that the alt-Hold issue is prevalent with the mini-APM. Both the flavors seem to exhibit that.
I flew it on 'stabilize' and 'acro' with the GPS puck to get some experience with the craft, but I eventually gave up on APM altogether and went with this: http://www.pixfalcon.com/
The fact that ArduCopter is only supporting PixHawk based FCs after 3.3 was the convincing argument.
And yes, I have been very pleased with my PixFalcon.
Gustav Kuhn > Tony KenwardNovember 8, 2015 at 10:35am
Well, for me, the pre PixHawk controllers, even the "china copies", works very well.
Easy enough to load the "earlier" code........
I have a PixHawk, still need to buy a reliable RX, with SBUS output.
Replies
I just tried Althold after getting rid of the vibrations.
It works!
Not perfect but I will get there after some tuning.
I would also like to inform you that in the testing process I tried two different props
Kingkong 6040 and Gemfan 5045 (both supplied by HK).
With the Gemfans my quad hovers at 36 percent throttle and 10.5 amps, while the Kingkong props need 5 % less throttle while using 2-3 % less current. I will stay with the Kingkongs.
Again thanks to all,
Mike
Thank you very much for your comments!
I just found out that I was part of the problem myself.
I mounted the miniAPM as usual with gel pads as I do always and did not care about vibrations (it had always worked so far, and the vibe measurements in the logs were good then).
But this little quad just produces a hell more. So vibes were about 3times of the allowed level. Argh!
I reduced the gel pads now to 1/4 size (area) and placed a 12 g metal plate below the APM. Now I have vibe levels lower than I ever had :-).
Don't know yet if AltHold now works, as it is deep in the night, and if it speeds off into the sky...
But I will try tomorrow.
I will keep you posted.
I think it's very likely to be high vibration. Copter-3.3.1 has improved vibration reporting but I think the miniAPM has the AVR cpu so I think you'll have to remain with Copter-3.2.1 so it's a little harder to be sure of the vibe levels but I'm pretty sure this is the issue.
I did as suggested in the link, the log is attached.
2015-11-09 16-19-09.log
I had the same problem last week. I eventually discovered that the craft would alt hold and loiter much better with the video Tx and camera powered off. I haven't yet had the chance to establish whether it was the camera or the VTx, or both, causing the issue. I did notice that in fast forward in alt hold it slowly dropped altitude and rose up again when I slowed down so I need to deal with that first. If you haven't yet done so I would recommend test flying with video switched off to see if your issues have the same cause as in my case.
hi Michael,
I had the mini-APM in my 320 hexacopter. i had the same issues you described. After much internet and soul-searching, I found that the alt-Hold issue is prevalent with the mini-APM. Both the flavors seem to exhibit that.
I flew it on 'stabilize' and 'acro' with the GPS puck to get some experience with the craft, but I eventually gave up on APM altogether and went with this: http://www.pixfalcon.com/
The fact that ArduCopter is only supporting PixHawk based FCs after 3.3 was the convincing argument.
And yes, I have been very pleased with my PixFalcon.
Well, for me, the pre PixHawk controllers, even the "china copies", works very well.
Easy enough to load the "earlier" code........
I have a PixHawk, still need to buy a reliable RX, with SBUS output.
Then, I'm still using XP Pro :-)
Hello,
Sounds like you are familiar with APM. So I'd recommend the following:
1) Calibrate Baromoter using "Set Home Altitude" in Mission planner and wait 30 secs or home altitude to be set.
2) Power Cycle the board and test the drone
3) Re-calibrate the Accelerometer. Be sure the there is zero motion for each direction.
4) Power Cycle the board and test the drone
5) Re-calibrate the Compass, making sure to have many many data points by turning off auto accept.
6) Power cycle the board and test the drone.
If problem still persists, clear the board parameters by hitting reset to default button in the Full Parameter List page.Follow the instructions.
Once the APM board is reflashed to default parameters:
Set Home Alt again.
Power cycle the board.
Test the drone.
Good luck
image.jpeg
image.jpeg