Determined to fix much weirdness with the way my MK AD6 was flying I pulled all of the electronics off of it last night and started by rebuilding the PDB without provision for the I2C isolator board, the BLs are now wired direct to C and D connections and the long servo wire leads are gone as is the isolator board from the stack. From there I put it all back together removing any unneeded wires and connection points, added longer standoffs to space the boards out a bit more vertically, and then zero'd the F/C memory and started rebuilding from a default setup. While I was at it tried out different props to see which worked the best and much to my unhappiness they proved to be APC 11 x 4.7, couldn't get it as happily stable with the Graupners.
Keep incrementally increasing the PIDs for the gyro until I had it pretty darn solid in the gusty wind blowing this afternoon, not as good as the WKM can do but WAY better than the AD6 has been of late. Also have it stripped down to just the basic frame and landing gear with dome over the stack, no camera mount and just a single battery. Actually too light for the big Pulso motors, it's hovering a bit below half stick. By late afternoon I had it improved to the point that I thought that's as good as it's ever been and as good as it's going to get. Then I saw the video at the beginning of this thread, hmmm...
After some cogitation and a bite to eat I popped the dome off, removed the Xbee, and hooked up the netbook direct to the Navi board. Took a deep breath as I fired up MKtool and clicked into terminal to start loading software, started with the Navi, no problem, then Mag which for once went all the way through no problem, and the the F/C that choked around 65% and exited with a CRC error. Ok, hope it didn't brick the board, try again, this time it completed and everything looked good. Closed Mktool and fired up the new version, immediately got error lights, bad compass error. Time to visit MK.DE to figure out this weird new calibration routine, watched Holger do it twice, still fuzzy, then read the step by step underneath in understandable English, Ahhhh...
All green lights after spinning the Mk in the air in front of me a few times, grab a batt fresh off the charger, out to the front yard to fly by street lights and the red LED strip on the front boom. A little strangeness, altitude hold won't and everytime I click the switch for it the Mk does a 45 degree turn to the right. Otherwise not bad, GPS seems to be better but somethin ain't right. Back in, reconnect the netbook and find I somehow set Poti 1 for the value of YAW I in gyro. Poti 1 is also the switch for altitude hold, no wonder it did strange things when I turned it on! Lower the min gas setting and back outside, plug in, GPS lock, start motors, liftoff.
WOW!!! BIG difference! Arrow straight, no yawing at all unless I move the stick, never done that since day 1, always been a bit twitchy on yaw. GPS hold still not as good as WKM but way better than it ever has been on this ship, may have partly been because I didn't take it up all that far in the dark and was getting some ground effect and also next to the driveway with cars parked may have tweaked the magnetic field readings and caused it to be a bit unsteady as a result. In any case the new release of .86 seems to have gotten it right this time, works really well from the little I've seen out front in the dark, have to wait for daylight and hope the expected rain holds off long enough to get a good flight in when I can see what its doing up there!
Ken