Once upon an idea...
okay, okay, you all know the great idea about building a convertible tri - to planecopter. Oh yes Avatar like, old story no news. However, I started to work on one. I am still far away from anything that is convertible, but it might be soon a flying tri.
I'm using APM v1 with the 2560atmega the oilpan, gps, sonar EZxx. The frame is made out of CFRB, re-enforced with honeycomb. The arms, about 36cm each, are made from titanium, with a cold coated wolfram-aluminum coating. The three motors, 750kV; NTM 2836; about 250W each on 4S 2200mA LiPo. The props are 12"4.5; CCW/CCW/CW. I use a digital servo to rotate one of the motors, but I've already the setup to turn each motor 90° fwd/aft. All together, including a bunch of stainless steel A4 grade bolts and sel flocking nuts, the tri is on 1.7kg. Heavy beast. If I replace the titanium arms with carbon fibre, I could save about 150g. My body is more or less two round carbon sheets, 20cm diameter. Way to much material, but - who cares. I use as ESC the HK Blue Series 30A.
The pictures shows a partly disassembled tricopter. Troubleshooting...
The Firmware is now 2.2.6b. To reverse the yaw servo I had to change the fiftyHz loop in Arducopter.pde
APM_RC.OutputCh(CH_TRI_YAW, ( (-1 * (g.rc_4.radio_out - g.rc_4.radio_trim) ) + g.rc_4.radio_trim ) );
Time for takeoff...soon. I've to find the best CoG - but I'm not completely sure on how exact the CoG must be - clearly, as good as possible,...but maybe a few mm off? Fine too? Who knows.
I had some troubles, most are solved:
*) Vibration due to loose turn table mechanic. Some preload on the bearings and the problems was solved
*) Vibrations caused by unbalanced blades. The CW blade (from Arducopter Store) had a nice inbalance. Trimming was necessary
*) Yaw servo reversed in auto trim. Solved it with entering the update line, see above.
*) Took me a few minutes to figure out on how to load AC through Arduino. But finally easy.
*) Then was the little software updates when suddenly the YAW servo wasn't supported anymore (2.2b-2.2b5).
*) Too heavy, first I had two 2200Ah LiPoFe A123 6.6V in series. I've now one single LiPo 4S 2200Ah; reduced the weight about 150g.
*) Some flips because I connected the esc left/right reverse....STUPID I AM
*) Some flips because one ESC did not provide the same output power as the other one.....solved new ESC; I hope
*) Last but not least...working on the CoG.
I use a 6 Channel TX/RX Skytech Ts6i.
I hope I can lift of at about 60-70% throttle max. The next days will proof it. Then next step: full conversion. Mechanic is easy, but coding will take time.
Thanks for Igor and Don for support.
Update: after some fine tuning everthing works fine. Updated to 2.4. and no big issues. Except...I lost a prop mount during flight. Lucky as I am I had no damage on the equipment, first, I landed in snow, second, I landed on the legs, third, titanium and carbon fibre can take the load of a four meter drop.
But since it looks promising, I think I will advance to the next version soon. Some minor tests with this frame, but I've some ideas....
Update: (still version 2.2.6b); still fwd/aft revered, but I mentally changed it ;-)
Went flying today. 5m above ground. Easy takeoff, hard to land softly :-) Big issue is that the tricopter heads forward without RC input. Possible reasons:
1) tail motor is not mounted straight
2) esc's are not properly trimmed/aligned - but I doubt it since the aircraft is stable
3) PID's tuning (which....Pitch...?)
4) CoG ? I actually don't think so, since see 2) the aircraft is stable
I lost a prop mount at my last T/O, but I was so close to ground that nothing happens...now I need for the NTM motors a new prop mount...
@Randy, nice hint, thanks, easy to solve.
@Don, okay it is/was installed the right way - I just had a brain lock yesterday night and thought I installed it wrong way...for whatever reason.
My guess is the pitch channel in your radio needs to be reversed.
When you did the radio set-up using the APMissionPlanner did you see the green bar for the pitch moved in the opposite direction from the physical stick on the radio? That's what it should do! :-) Not too intuitive..but if you actually look at the raw signals coming out of the radio it makes sense.
RC pins point to the back. If you look closely at the IMU up near the I2C port and SW1 (the other button that isn't reset) you should see FRONT and a little arrow.
Gosh dammit. APM reversed. Pins should point away from the tail (fwd!).
The beer trimming function (in short BTF) from last night works. First take off and a few touch and go. Stable, but always a slight fwd heading without control input. Moreover I've the feeling that when I push the EL control stick aft, the tail motor increases rpm, while when I move the stick fwd, the two fwd motors increase the rpm. Shouldn't it be revered?
Sorry for the poor video quality.
Beer fixes everything :)
Found finally a way to determine the ideal CoG of my little tri!