My quad is flying better than it should. I almost feel guilty.
I slapped together a frame in a few hours. I use cheap props mounted on long shafts with cheap collets. The motors are not precisely aligned to the frame. The props are not balanced perfectly. Motor centers form a rectangle not a square. APM is strapped to the frame. I never bothered to calibrate compassmot. CG is not in the center of the frame.
However...
Loiter is freaky solid, missions are as boring as dressage, no fly aways, toilet bowling or any other stability problems. Gopro video is jello free. With each new firmware version it just keeps getting better and better. Holds position in wind well. I have flown several versions of the firmware on this frame and never had a problem with any of them. Just Dave C tune and go...
I started to think about why it works. Some of the ideas may help others designing their own frames.
1. vibration has a very hard time reaching the APM.
2. the power distribution board is located far away from the APM
3 the power distribution board is not an electromagnet printed on a circuit board
4 motor wiring is away from the APM
4 The GPS is located away from the power distribution board, ESCs and wiring
5 The frame is very light so current and therefore magnetic fields are reduced
Vibration is low. Around "2" in hover
Even with motors just taped to the frame...
APM is just strapped down with velcro.
Good power distribution method? Bad power distribution method?
GPS is away from most RF and in null of telemetry antenna. GoPro on or off does not seem to bother it. I do shift the battery a little to help balance.
Hope some of the ideas are useful. This type of frame makes a great EPP "trainer". There is more info that was posted here:
Comments
I loaded 3.0 today and ran compassmot calibration. 15% interference at 23A
Loiter held position better but I need to reduce the PIDs to get rid of the small oscillations.
perfect use of swimming noodles without doing sport!
Ingenious! Love seeing folks experimenting with materials. In addition to the foam noodles, give PEX tubing(brand names: SharkBite, Apollo, etc...) some consideration as a frame construction material. It's light, strong, easy to work with and is noodle compatible. :) Here's a link to a video of an all PEX construction quadcopter:
http://youtu.be/dn3U07NhQhc
I've used noodles for landing gear and floats on the copter but not to reduce vibration. I think it's time to experiment with a noodle core on the HooperFly. Great build on the Noodle copter; very original. Looking forward to seeing footage of it flying. Cheers!
I purchased the rods from intothewind.com - kite supplies
The hard part will be waterproofing the APM without adding a lot of weight.
DIY at its best ; )
Wal-Mart and K-Mart sell them. I have seen them at other stores as well. Funnoodle seems to be the most common brand. Diameter is about 2.375". Color seems to vary with each batch so if you find a color you like buy several. They are seasonal items and hard to find in the winter. When I found them in January at a K-Mart the clerk said "Going to Florida?" I just answered "yep" since the truth would have taken too long to explain.
There is a larger 3.5" diameter type with a bigger hole but I have not had as much success with those unless you count this...
I am putting together an order list in hopes of making one of these! Where do you get your noodles, Glenn? I'm afraid that if I order them online then they will come all bent up. Also: what is their diameter? I want to make sure I can find the same kind.
I love the fact that you can measure the vibration with the logs. When I was using Wii motion+ sensors I learned that one piece of tape or velcro could mean the difference between a quad just flying and flying great. I think having the battery coupled to the frame near the APM is a very effective "vibration capacitor".
I suspect that you can do only so much with software. I would rather have precise positioning than risk filtering out useful data.
Hmm, I think that your frame should teach us all the following lesson: Vibration control is everything. And because your frame is not super rigid (and has, apparently, a very high damping coefficient) it soaks up all the vibrations when compared to aluminum, carbon fiber or even nylon. My quad flew very erratic during all automatic modes until I mounted my APM on kyosho gyro gel. Four tiny 7mm square pads, one per corner. Once I did that it flies perfect.
Maybe this suggests that some tweaks to inertial navigation could be made to desensitize somewhat to vibrations, of course, CPU permitting. I've been told other autopilots aren't as sensitive to vibrations as ours.
In any case, congrats on your build, resourcefulness is key!
-
1
-
2
-
3
of 3 Next