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.



Setup Instructions

Views: 48052

Reply to This

Replies to This Discussion

The HXT900's are analog aren't they?  I don't think an analog servo has any place on a heli, especially not a flybarless heli, and especially not with Arducopter.


The reason is analog servos will NEVER center accurately.  They just can't.  They will always have deadband in the middle which is huge compared to a digital servo.  This is because torque drops to near zero close to the target.  If you aproach the same setpoint from opposite sides, they will stop in different locations.  Also, they are much slower to make small corrections than a digital, much bigger difference than the speed ratings would suggest.  Those ratings are the time to do a full sweep of 60°.  But with analogs, they are slower to make small changes because again, torque production is near zero for small movements.  And also, digitals have much much more holding power compared to a digital.


All of this plays havoc with a PID controller which itself acts like an analog system.


I can't see any reason you would ever *want* a slow servo.  My tail servo is .03 sec/60, and I'm planning on putting them on the swashplate too as soon as I can afford to.


The *only* downside to strong, fast, digital servos is much higher power drain.  I'd say probably double what an analog servo uses in the same application.  So, make sure your BEC is up to it.  This is because they will be more "hyperactive".


In this system, the closer we can get to an ideal model where the PID controllers are acting directly on the inertia, the better it will be.  Any delay of the controls will make the system sloppier.  When the heli is piloted by a human, it's less of a big deal because, as smart as the Arducopter is, it's not as smart as our brains.  We automatically learn to anticipate lags or slopiness in the system response.  Arducopter cannot.

Randy, is that AC2 section used by the Quad guys too?  Any chance we could have "Throttle Rate" changed to "Collective Pitch Rate"?  I think it's a little clearer.


@Marianne: Looks like your Throttle Rate I term is quite a bit higher than Randy's.  That could sure cause the symptom you are describing.  I usually tune a system by making the I term 0, and then work with only the P term until it's as good as I can get it.  Then you use the I term only to "snub" the system so that I can get as close to the setpoint as possible without getting into instability again.


This is about as good as a human can do.  If you have some systems (datalogging, osciliscope, etc.) and some complicated math (Ziegler Nichols method), it can be improved from that a bit.  But since you probably aren't equiped to do that, you'll just have to do the best you can.  Plus, since we don't have a D term, the ability to use these methods is hampered.


Everybody doing this needs to read and understand this:


It's a pretty good article on the subject.

I guess that solves the problem. will try to test that PIDs tom. :) I used the code from the GIT.

Thanks Robert.  I am going to upgrade my servos then.  Do you think the HK-922 is a good and affordable choice?  In my test this servo performs well on the tail.  Would you recommend something different in the $10-$20 range?



Graham, I really can't recommend any servo, as I haven't used those ones and I don't own a 450 heli.  However, the HXT-922 is backordered, and the 933 is not. ;)  So your choice is easy if you can't wait.  The 933 is a bit slower, but puts out more torque.  This is the way people tend to go with swash servos.  I think particularly with flybarless, you need a bit more torque than flybarred.


I don't know if it matters, but I've only bought Turnigy branded stuff, not Hobby King, and so far I've been pleased.  I'm not sure I trust Hobby King electronics.  That's just my opinion, I don't know if it's really worth anything.  It's mostly just based on reading from others.


The Turnigy DS530 looks like a really strong, fairly fast servo.  I see some mixed reviews however, but I don't pay a lot of attention to those.

Robert, from my experience so far I can only recommend you to test as most as possible on the ground. And please try what happens in alt_hold when you start moving the pitch on your RX. Like I wrote I got very strange behaviour. Pitch suddenly jumping to max value when I increase pitch and I could not bring it down again by giving low pitch with the RX. Will try to shoot a video tomorrow and post it here to demonstrate the effect.

Thanks Robert.  The DS530 does look quite good based on the specs, but like you said the few review there are not that good.

Yeah, but like I said, I don't put a lot of weight into those.  The ESC that I have (Superbrain 100A) got terrible reviews, but mine is fine.  I figure they are selling 1000's, of course they get a couple duds.  So those guys post, and that's what you see.  If HK had a more effective customer service plan, if they took care of these problems, you probably wouldn't see it.  So I don't think it's that they have a 50% failure rate or something, might be the same rate as any Name Brand, but the customers are upset because there is no recourse, and they post.


I dunno, that's how I feel about it.  I bought about 20 individual electronic things from HK, some with bad reviews, and they all work fine.

The Rate Roll won't do anything for a heli but for a quad it will be used because for them they still use a nested controller for stabilise mode.


The deadband is only used for the radio input.  There's no deadband for the heli's output to the servos.


"Lock pitch and roll" likely just copies the PID values between the appropriate roll and pitch sections as you update them.


Throttle rate and altitude hold are the two nested controllers for altitude hold.  So the outter controller (altitude hold) will look at the distance from the target and decide on a rate of ascent (or descent).  It will pass this to the inner controller (throttle rate) which is responsible for implementing that rate by moving the swash up/down.

Yes AC2 is used by the quad guys too so we run into the confusion re the wording especially for throttle vs collective pitch.  We could split the screens for heli and quad although that would lead to more maintenance for michael oborne and the possibility that the heli gets left behind as jason makes changes to the quad code that naturally carry over to the heli code.

It sounds like that's what needs to happen, eventually.  If the Pitch and Roll Rate PID's no longer apply to us, that's another reason to split the pages.


I understand that it will make more work for Michael, so we'll have to be patient.  In the meantime, I'll have to try and update the wiki so that this is clearer.

Ok, so to make sure I'm clear:


Quads still use nested controllers.  In stabilize mode, the stabilize PID's operate on top of the Rate PID's.  In Acro mode, they only use the Rate PID's.  Quads do not use "Acro PID" at all.


Helis no longer have nested controllers.  Stabilize mode ONLY works with the stabilize PID's.  In Acro mode, we only use the Acro PID's.


My "death wobble" was most likely due to the the Acro PID's being untuned.  Yeah?


So, I just saw the announcement for the new T3 competition, and I'm thinking of giving it a shot.  Is that realistic, or are we way too far off from getting a heli to fly waypoints?  Obviously I need to do a lot of tuning yet, but is the code there?


Another question: is there a mode whereby we can fly the heli and have the tail always automatically point back from whatever direction the heli is flying?



Reply to Discussion


© 2017   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service