I have just bought the ArduPilot Mega 2560 Full Kit... plus the Triple Axis Magnetometer HMC5883L. I should have it built over the weekend.. then i'm off to a my mates machine shop, and i'll make the splitter plate for the lower frame onto which i am going to fit the Ardupilot. the magnetometer i'll put on the tail as suggested. I am going to use my X-cell Razor 600E. For those that dont know its around the 50 size nitro heli.
Do i need to buy any other sensors i.e. the Sonar... i can fly a helicopter all day long but this is my first time flying one with stabilisation so any tips would be gratefully appreciated.
If this works well then i am going to try and use it on my x-cell gasser. i can get up to 20 minute flights with this. But i think for the time being a battery 600 is the way to go.
i'll post some pictures when i have it all together.
Replies
Oh, and 2.5 has been released, it's in the Mission Planner now! I'll try it tomorrow.
I did another 2 flights today with 2.4.2 from ~Monday.
I might have found a bug in 2.4.
When in loiter or position mode, a manual stick input in pitch or roll, which usually resets the loiter position, disables a correct function of loiter mode. Nav goes all of a sudden down to zero. The funny thing is, even switching out of loiter and back to loiter mode, does not help. Loiter becomes unusable until you reboot the APM.
I think other pilots might have had the same issue, because I read in some posts, that after some nice loitering, all of a sudden the quad drifts off to one direction and never comes back. Same happened to me. Reason, because I touched the roll/pitch stick and freaked up the loiter function by doing that.
Is it a known issue?
Does manual stick input not reset anymore the position?
See, when I put in a roll signal, the nav values go back to zero, but the long and lat error freeze.
Btw, when the long and lat error go down to zero, that is when I switch out of loiter. After that I switch back, but no effect anymore. Loiter seems to be not working then anymore.
Just a quick note on those Mi servos that I bought from Hobby King. I'm not quite sure what to make of them. On the face of it they seem like good value, they look to be well made with titanium gears for a nice price/performance. However I'm not convinced of their reliability.
I had one fail very early on. I ran all of them on a servo tester for 30 minutes without problem, but then one of them died while I was setting up the heli. I'm not sure if it just died, or if I plugged a servo lead in the wrong way. Obviously that's a mistake, but every servo I've ever used has been capable of withstanding this.
Then I made the mistake of plugging 3S direct into them. One of them immediately let the smoke out. Obviously this is my fault. However I find it interesting that the 2 remaining ones are fine. There seems to be great variability in their built in that regard.
Finally, the HK470902 that I have, the fastest one, seems to exhibit a really bad flutter around center. It's similar to the way normal digital servos can flutter a bit if you plug a freshly charged 5-cell NiMh battery. I think the internal PID loop is being overdriven with the high voltage. So, it's just like that. It appears to go away if I run it with a 6V regulator. And it only happens right around center. If I move it to a position off-center, it doesn't happen. Weird. The arm doesn't really move, but you can hear the thing basically banging back and forth at super high speed. A buzz. I only have one of these, so I don't know if it's indicative of a design problem, or a one-off quality issue.
I'll continue to experiment with them, but at this point I can't recommend their use on a really expensive machine. Of course, I'm sure people with $1000+ machines wouldn't be interested in these $35 servos anyway. ;)
The heli holds very good with external gyro alone. However if I route receiver output to APM channel 4 input, route gyro to APM channel 4 output, the heli will wag seriously even if I dial the gyro gain to zero. So it seems the APM is trying to act as gyro in this configuration which I don't need it to.
So what setting I should change?
I just completed 2 good flights with the code from trunk (from
yesterday) in my heli.
Overall, it went quite well. The compass was perfect. I flew at the
suspected Bermuda triangle field, and everything was fine. I have no
idea what the issue was last time.
Again, this is with my backwards APM and compass, on a highly
yaw-active heli, no problem with heading that I could tell... so the
code seems pretty robust. Well, I had a huge slew to the left
everytime I stabbed the throttle, but I assume this is just due to
rotor torque which we can work around. I would like to have a
look and see what my magnetic field looks like to confirm it's not
that.
I played with h_collyaw a bit for the first time, and got it up to 5
without really feeling like it was helping. Then I looked at it on
the ground and... I think it's working backwards! Randy, I'll
confirm.
I tried some Alt_hold, and quickly realized that my sonar is just
rubbish. With any throttle_P at all, it just goes crazy. I don't
know what the problem is. Well, I have an idea. It might be my tail
support booms slapping against the plate which holds the APM and
sonar. But both are isolated from that plate by nice rubber mounts.
So I wouldn't expect that. Regardless, I'll remedy that situation and
test again. If that's not it, then I have no idea. The sonar is
heated, filtered, and the wires aren't shielded but they are only
about 1" long. The sonar is rigidly bolted to a rigid bracket which
is rigidly bolted to the APM case giving a nice lump of mass, and the
lot has some really nice rubber damping, and the heli right now is
about as vibe free as it's going to get. It should have a smooth ride
and not be subject to mechanical vibration.
After I turned off the sonar, I tried some more Alt_Hold and it was
generally fine, though it wandered a lot, that I can work on.
I tried some loiter, which was not great but didn't do anything really
spooky. I tried to let it wander to see what it would do, but it
seemed like kept wanting to wander over my head so I aborted cause
that's not cool. I need to tune that as well, but first have to
figure out how.
Now, I did have something very strange that happened in the included graph.
You can see some massive spiking on Thr_In and RC3_out.
That was NOT me, and I don't think it was (shouldn't have been) in
Alt_Hold. Where did these come from? I noticed it at the field, I
thought maybe my ESC or motor was giving out. It was a really sharp
vertical shaking. I landed, checked things out, everything seems
fine, so I took off again, and it was fine.
I assumed it was a mechanical problem until I got back home and looked
at the logs. Why is Thr_in moving like that? It must be a bug, or a
radio glitch is all I can figure? But I'm running 2.4Ghz, and it
didn't affect any channels other than throttle. I have no idea, so I
think we should look at it.
I am trying to setup a 450 clone but the tail seems not working at all. If I use no gyro, the tail servo moves a limited range that won't hold the tail in flight at all. If I use gyro, the tail will wag even I set the gain on gyro to 0, so it seems the APM is trying to act as gyro in this case.
Can any one share a set of parameters as a reference?
Thanks.
Guys, FYI, there's a problem right now where if you upload the code from GIT, Mission Planner does not recognize that you have a Trad Heli, and won't let you use the Heli setup screen.
Randy and I believe this is due to some parameter names that I changed when I made my H1 swashplate changes. So, this is just temporary.
Malcolm, this is new, since I talked to you last. Your hunch was right!\
And Happy Birthday Malcolm! ;)
Had a 4 flights today, cant really get the heli to stay at one attitude in stabilise, i'm forever feeding in some cyclic, sometimes its slightly pitching forwards, other time backwards, left and right..... I have a Stab P of 1.6, it was so much better with *0.40 code, the heli was always rock solid in stabilise.
The Alt_hold works really well, i did notice i had a blade slightly out of track. one turn of the ball link and it was back in track this helped the Stab, but i still needed to feed in some cyclic. I wonder if this is the "leans" which the quad guys are talking about?
Ok, just flew 1 pack, and I managed to cure my constant need of right stick problem. It was because I had Yaw_stab_I at zero. I had done that intentionally, because it doesn't really make sense to me. The gyro measures yaw rate in deg/time. Rate yaw controls the yaw rate in deg/time. I have P and I terms in yaw rate. The I should have wound up enough to counteract a yaw, but it did not.
Stabilize yaw commands a yaw rate based on compass error. It should not need stabilize I to just not rotate. I would think it is only required for example, to... actually I can't think of any reason it would be required. It commands a rate. Why would you ever want to have a constant turning rate based on a compass heading error?
Anyway, I put some Stab I in, and all is good, it now holds a heading. I don't get it.
And also, why did it fly well before with no Stab I? What has changed? Now that I have stab I, it is commanding a persistent yaw RATE into the yaw rate system, so it should just keep yawing. It's almost like I have a constant yaw rate error coming out of the gyro. But why isn't that calibrated out when you arm the motors?
Edit: hmmm... on further reflection, I have to remember that Yaw_stab_I does not command rate! It goes direct to the servo output. So it will build up to offset rotor torque. So it won't hurt. But it still shouldn't be needed to make it stop yawing. I think yaw rate should be able to do that???
-
10
-
11
-
12
-
13
-
14
of 34 Next