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.
Hi Malcolm / Martin / Randy,
I was just thinking. I also sometimes get the altitude hold to work and sometimes inbetween different test, the heli will just climb without holding altitude. So my thoughts is as such:
1. As we know with the Quad, Hexa (mulit rotor copter) we mount the APM in the centre and away from the rotors. But with the Heli we are underneath the main rotor. For me that is already a problem.
2. The heli compress air to get a lift, so for me this is already higher pressure levels underneath the rotor.
3. If I understand it correctly the funnel of the main rotor is right in the middle,centre of the blades (exactly where our APM electronics is).
4. So for me (please correct me if I am wrong) when the main rotor starts up it creates higher pressure, we hover, we then switch to Altitude hold, this gives a higher pressure reading the APM (higher pressure means we are loosing altitude , therefor it needs correction) gives more throttle (in our case more Pitch), this creates again more pressure, the APM things we are loosing more altitude.....so for me it looks like we can have a run away train.. (or heli).....alright I think up to a point of cause but this explains to me why the heli keeps on climbing sometimes.....with out no reason.
Don't you think we need and external pressure sensor away from the rotor? Yes it will be a challenge where to mount? maybe just above the tail rotor with some extention mounting material.....not sure.
What you think?
Good thinking Hein! Yes, I agree that this is potentially the problem and I've been thinking about this as well.
A download of the logs from such an event should clearly show us the issue because we should see the barometer's value is crazy.
I worry that even placing it at the back will cause problems unless it's on a boom stick out from the back of the heli at least 5 ~ 10cm so that it doesn't get any interference from the back prop. It's very possible to create such a mount using 3d printing. I've recently been working on a sonar mount to get my sonar away from the engine and esc. We could do something for the barometer as well...although some additional software changes would be required as well.
I see what you both mean about the sensor being in the high pressure funnel.... but i went out to do some more ALT_hold last night.... added some D in and now it gives a little flick when the ALT_command is given... then it really is rock solid in both sonar and barometer. It has only ever kept rising after coming out of loiter mode.?
Did you get a chance to look over those logs i posted randy?
That Sonar mount looks great. Will it be available in the future at DIYDRONES store or from you? Did you guys do any modifications on the traditional Arducopter code so far? I see you are using 2.0.38....
I'm using 2.0.38 as its easier for me to download from the repo rather than using svn. I made a couple of little changes but nothing serious.
The tail on mine uses a mix of both the Acro/Stabilise PID's for yaw. The rest of the Control is straight from the PID Stabilise. I have found it easier this way it gives me a faster manual tail response without oscillations.
Are you building a Traditional Helicopter?
No specific plans to put that in the store. It's just small part and people can just print it from shapeways if they want it. I suppose if we get a decent collection of parts we could put them into a little kit to make it more convenient for people to get everything they want at once. There's a GPS mount as well.
I'm using the regular code from subversion. I occasionally load the code straight to AP_MissionPlanner to make sure that it works.
Here are two graphs taken from Malcolms logs.
The 1st one just shows the difference between barometer vs sonar. You can see that they start together but quite quickly the barometer shifts to indicate about 2m less than the sonar. Hein and I talked a bit about this in Skype today and his (likely correct) theory is that we only initialise the barometer when the APM first starts. In the regular quad code we do it twice..once when the APM starts and then again when you arm the motors. Because we don't bother to arm the motors in the heli we miss this 2nd initialisation. Now it's actually possible to "arm the motors" on the heli too. Just hold collective down and push yaw to right for 2 seconds. You should see the red, yellow, green lights flash quickly for a few seconds. We probably need to do this better somehow for the heli.
The 2nd graph shows the climb after malcolm came out of loiter. The sonar and barometer show the altitude has risen..so doesn't look like a hardware problem. I.e. the sensor's know the heli is going up. the blue line shows the "throttle out" = collective pitch. So this does go up..so my guess is that as you switch you of alt-hold, throttle returns to human control and this is a little higher than what the autopilot was doing so it rises.
I think it's just not trying hard enough (i.e. P is too low).
In the graph below:
red = longitute error .. how wrong it thinks it's longitude is
blue = the roll/pitch correction it's applying to try and correct the lon error
green = lattitude error
white = roll/pitch to correct the lattitud errorr
So the white and blue lines show how much it's leaning..but i think the peak of the blue line is only about 5 degrees..although that does seem to correct the red line.
The white line never seemed to be enough to correct the wandering green line.
I definitely think removing the flybar will help!