Servers by jDrones

HeliQuad rudder 4th servo not mixing with the collective - acting as tail rudder

(Josh) #23

Thanks Bill, re-reading my posts sounded different than I intended, I did not intend to sound so arrogant or like I am the best pilot, just that I am very familiar with CPPM and have been flying forever so, by choice, I prefer nobar stabilization. I prefer zero gyro support (other than HH for heli tails) but I am going to leave it with stabilized mode and acro mode to start before I tweak any of the params you mentioned. That way I will get the feel of the flight before I strip out all stabilization, I may just leave it, I dont know yet (until I get it in the air). I should have the blades any day so I will definitely be posting the first flight feedback. Let me know if you see anything odd or out of the norm, it probably is not intentional. I know now why my channels were out of order, its on the radio calibration, every 3rd or 4th try, it doesnt move to the next step, the green bars are showing it is moving per the step but its not always registering (leaving it all the way to the left for example), so to get around this I was moving the gimbles a little bit to get it to register (now I know the issue this causes so that wont be happening again), this is what screwed me originally. Now that I know what params they are and what is happening, I can just fix them manually after I calibrate the endpoints.

(Bill Geyer) #24

No. I didn’t take it that way. I know there are many accomplished pilots that can fly without the aid of stabilization systems and I admire their abilities. I’m not one of them. I’ve learned that much of the raw handling characteristics depend a great bit on the quality of the helicopter design as well as the set up.

(Josh) #25

Thanks Bill, here is the latest .param file I plan to fly with. I am still tweaking the RC_3 servo as its acting a tad slower than other 3 servos (all brand new, all bought same day last week). Other than the servo 3 change, everything else is pretty much ready. If you wouldnt mind, take a quick look and let me know if you see anything odd like you did the first time around. Thank you in advance!

6-11-2019-A.param (22.4 KB)

(Bill Geyer) #26

Josh,
Looked at your file. Here are some corrections
ATC_HOVR_ROL_TRM - make this 0. this is for heli’s with tailrotors
ATC_INPUT_TC this looks very small. I would suggest making this 0.1 or 0.15
H_COL_MID 1500 - does this correspond to zero collective blade pitch?
H_LAND_COL_MIN 0 - depending on what amount of negative collective pitch you are setting then you’ll need to adjust this so when in non-manual throttle modes like althold, the collective doesn’t go full negative. If you have your H_COL_MIN set for -2 deg collective pitch then leaving that at zero is fine.

I think that is all for now.

(Josh) #27

Thank you very much Bill! I do have zero at mid-stick, I set the endpoints and radio params to be exactly 0 (knife’s edge) and zero degrees for mid-stick (all in PX4, not radio, radio trims are zero’d across the board). I will set your params and see how it goes, my blades are arriving tomorrow so perfect timing! I plan to test fly tomorrow (assuming the wind holds up). I will try to get a video (if successful) or follow-up with feedback if not successful (but it should fly). I have tested every point of angle of attack, and she pitches correctly across the board for lift, roll, pitch, and collective across the board. Assuming no issues, she should fly. This is how I test all of my helis pre-flight so this is no different, its the same principles times 4 rotors (heli 1 rotor with swashplate) but I have corrected for the tail rotors (no swash) to pitch for the rotors (if that makes sense). What I mean is that all 4 rotors are pitching at the right times, at the right angles to mimic a single rotor swash, the 4 angles are perfect so she should fly straight and true.

I should have led the convo with who I am, I am a senior sys engineer working on wall street as my day job for the last 25 years handling the back-end services for many clients. My night job is to engineer new complex web.services virtualizing large data centers into cloud resources so working with my hands is something I have done since I was 3. I used to take apart my moms vacuums at 3 (yes THREE years old) and fix them (or break them worse), by the time I was 12, I had started my first Harley Davidson rebuild and finished it when I was 16 (complete ground up rebuild of the entire motorcycle, engine, carb, ETC, complete ground up build), so working with my hands and engineering things to work is my forte. Starting in planes when I was tiny and could barely walk then moving to RC cars, then finally helis and remained with helis ever since (only helis) until I found the CP drone. I stayed away from drones (UAVs) and avoided them like a virus because they are way to easy to fly and nowhere near as challenging or rewarding as building and flying a CP heli from scratch (TRex 600-800’s). This is my opinion so UAV fliers dont attack me, I just prefer helis but now I am into CP drones so hopefully it flies as cool a my helis (or BETTER)!

I have been waiting for something like this for a long time, helis are all pretty much the same now, they continue to advance in technology (FBL gyro but I dont use them) so the helis themselves are for the most part, the same across models and makes. MY preference to go NO-BAR means the helis are all mechanical so the advances in technology do not help me. Enter the CP drone, enter the PX4 flightboard open source, and you opened up a new world for me. This is a dream come true for me, a platform where I can program and write my own code for the flighboard and have a community behind me to help is the best thing since sliced bread. I only wish I knew of Pixhawk sooner, I would be proficient by now if I had known of this years ago :frowning: I LOVE the idea with Pixhawk, open source and the ability to write and change anything I want to! I have been a long time developer of Linux so this is right up my alley!

1 Like
(Josh) #29

Here’s my baby, almost ready to fly (blades were just delivered)!

1 Like
(Josh) #30

Bill, quick question (if you dont mind another Pixhawk newb question), I went to the field yesterday, prepping to test fly and I stopped (to ask and confirm the next question). I cannot seem to arm the motors in “stabilize” mode, only in “acro” mode does the motor spin, so am I doing something wrong? When I go through the arming steps (yaw right), the PX4 arms (lights green and tones), button disabled so all servos activate and move (on boot with no button), but if I am in stable mode, the arm (channel 8) toggle doesnt kick in the motor, however in acro, NP, the arm channel (8) fires the motor and the blades turn. My question is how to arm in stable mode to lift off, the reason I halted the test flight and the reason I decided to fly my TRex 600 and 700 instead and not the UAV was to ensure this is normal. I was at the field with all charged lipos but wanted to confirm with you before I tested the UAV. Is this normal behavior, and if so, how to I enable stable mode and lift off? In acro mode, it armed and spun, thus would lift, but I didnt want to chance it until I have confirmation just to make sure I didnt miss a config somewhere.

(Bill Geyer) #31

Josh,
Sorry for the delay. One thing I noticed in your set up is that you are using RSC Mode 1. this is a pass through mode. So I’m not sure whether you are using an ESC with a governor or using a throttle curve. If you are using an ESC with a governor then use RSC Mode 2 and set the RSC_SETPOINT variable. If you are using a throttle curve then set up the RSC throttle curve. My thought is that when you switch between the stabilize and acro modes that your RC channel 8 settings are slightly different. the RC channel 8 must be within so many PWM of the RC 8 Min parameter value. that is my theory. Do you have some settings in your RC transmitter that is changing the RC 8 output to the Pixhawk?

(Josh) #32

interesting about the TCurve, this is how I fly my helis so naturally this is how I setup the new model for the drone. My radio maintains channels between flight modes so this would not be the radio causing it, but I bet its the RSC mode. Let me play round with that, as always thanks!

My main concern was to know if it should fire in stabilize mode or not (sounds like it should), I wasnt sure if acro was a lift-off mode and stabilize wasnt (although that would seem odd to me) since stable should be the smoothest flight mode so take-offs should be fine in it. I figured acro and stable would both be take-off flight modes so this was why I reached out before attempting t fly (to prevent any newb issues like this one). Let me try it with modding the RSC modes.

(Bill Geyer) #33

I also noticed that all of your RC Min and Max values were 1000 and 2000 respectively which makes me think that you haven’t calibrated your radio. Also you should not have any mixes or special settings in your RC transmitter. the only special mixing should be done is to get multiple flight modes with one or two switches. Don’t use any dual rates or expo for any channel. this is all done in the pixhawk. trims should also NOT be set in the transmitter. I basically use an airplane model in my transmitter and set the throttle hold switch to channel 8

Yes you can takeoff and land in any of the manual modes (stabilize and acro). Loiter, Althold, and poshold all use vertical rate command and altitude hold for the collective input. Control of longitudinal and lateral depends on the flight mode. Yaw in all of these modes is rate command with heading hold. You can also take off and land in loiter, althold and poshold.

(Josh) #34

I was tweaking the min/max on my own to test the radio/throttle results, but yes, it was RC radio calibrated successfully. No mixing is done in the radio (except for the CPPM swashplate type of 120 degree which seems to be correct), I tried straight swash but that didnt work right. 120 seems to be correct and working with this model. No other mixing is done or enabled (my switches are directly tied to channel 7/8 but not done through any mixing, this is standard 1to1 mapping).

(Bill Geyer) #35

you should NOT be doing any mixing in the RC transmitter. i recommend using an airplane config in the RC transmitter. the pixhawk does everything

(Josh) #36

hmm, interesting, so for the swash, nothing is needed at all and everything is handled through the Pix? Really interesting and good to know, I am going to test this and see if it fixes all the little weird issues. I can try a plane model then go through the calibration and config again.

(Bill Geyer) #37

So I saw that you had modified the SERVO parameters. is that what you meant by calibrating it manually? if it is, then you’re doing it wrong. the RC parameters are what should be used to set the output range of the transmitter PWM signals. Doing an RC calibration through the set up screen in mission planner or QGC, will automatically set the RC values for your transmitter.

Yes. just the calibration of the RC transmitter using MP or QGC. Turn off any ICE throttle cut off features.

(Josh) #38

Nope, not what I meant, I meant I actually go through the calibration in GC, or MP, I actually go through the steps to hit all endpoints and go through the steps (hold left, hold right, hold mid, ETC), the settings you are seeing were to overcome a pitching issue with one of my new servos reacting oddly, but the param file you are seeing , is older, I removed the servo settings (but everything else remained the same).

(Josh) #39

Thanks Bill, I made a new model for Airplane, and will go through the radio calibration, my ICE is for my 700, sorry, wrong heli, no ICE in my drone, the drone uses a cheaper ESC (not my Talon HV). I will go through the radio calibration using the plane model and follow-up. Maybe this is why all my inputs were screwed up originally (as I was always using a heli model - 120 CPPM).

(Bill Geyer) #40

I am talking about the feature in RC transmitters that lets you turn off the engine. I forget what they call it

(Josh) #41

Throttle cut, idle-up, I use throttle cut for my safety toggle to keep the massive 700 blades spinning until I am ready, then once ready, I remove the throttle-cut and spin the blades until mid-stick, then I switch to idle-up, this sets my governed speed so at knife’s edge its equal at any position. I still have throttle management until I hit mid-stick then its equal across the board, positive or negative --> 2200RPMs no matter what the throttle stick is set to in IDLE-UP. I have equal RPM in both inverted and normal flight (11 degrees neg, 11 degrees positive) with a governed RPM of @2200RPM (DFC).

(Bill Geyer) #42

Ok sorry this is an airplane thing. I’m not talking about the throttle hold, again that is all done through channel 8. on your RC transmitter, when you choose the airplane model, they typically have a feature for ICE that you can set to cut fuel to the engine. I want you to turn off that feature in your RC transmitter. It will mess with the collective range (throttle stick) if you were to accidentally flip that switch.

my big point here is that your transmitter is nothing other than joysticks and a few switches to the pixhawk. you don’t want to do any special programming that will mess with the signals that the pixhawk gets.

(Josh) #43

Bill I cant thank you enough! Your advice for the Plane model was the best advice so far! You have no idea how I struggled to set and write the servo params to get them to work respectfully in the heli model to using an airplane model and not needing to do ANYTHING! None of that is needed in the plane mode. Removing ALL mixing was the trick, it even calibrated 100% without any extra swipes of the joystick, it just worked first time. I now see why I had so many issues, it was the damn 120 mixing of the swash throwing off all my radio readings in the PX4! My drone is ready to fly, it was ALL the DAMN MIXING!!!

Just so you know, I am writing up a step by step tutorial for others who wish to use PX4 with a CP drone (stingray, VL380, or HK) and will site your help. I am writing out all the things to do and not do to get this working across the board. I am still new to PX4 (pixhawk in general) but obviously not to RC or helis so my struggle has been converting the heli knowledge to CP drones. You have been a huge help in this area, I have all issues sorted now, including throttle, flight modes, servos, tweaks, ETC. THANK YOU! The AP mode saved the day and made all the sense in the world.

I am so used to just fixing issues, that I missed the underlying root cause (CP 120 sawsh) so my solution was to tweak the servos, none of that is needed. This is the type of things I will note in the user guides for others so they dont struggle as hard as I did. WOW am I shaking my head at myself, total engineering oversight on my part, SHEESH! I cannot believe I missed such an easy and obvious issue as mixing in the radio!! Wow I am embarrassed at myself to be honest! Than you so much Bill, I will get her in the air today or tomorrow and send a video or at least a follow-up but it is definitely far far far better now than it was before when I had CPPM swashplate configured in my heli model, but using AP model, no issues at all, the PX4 mixes everything beautifully!