Hello,
meanwhile I get along well with Pixhawk 4 mini, Ardupilot and MissionPlanner. I have equipped 5 helicopters with it, 2 of them tandems.
Now I have a few wishes for additional features on the helicopter, if such a thing would be possible.
trimming of the swashplate.
It would be practical to trim the whole swashplate in height (for hovering), similar to -min, -max and -mid.
Especially on a tandem you could easily match both swashplates and compensate differences.
tandem pitch control.
On a real tandem the swashplates make a nodding motion (like on a single helicopter) and move a little bit up or down at the same time. I think -nick- (pitch) becomes a bit more comfortable and softer. And it would be scale-like.
It would be nice if you could adjust the ratio continuously from 0 - 100.
-standby- on the ground .
When the helicopter is on the ground and running, it sometimes happens that the tail wants to turn away (from the heading hold). In tandem the rotors become increasingly interlocked. With -standby- you can prevent this.
I have this with a mixer in the transmitter on the throttle-stick. When it is down, -standby- is active. When I lift it slightly, -standby- goes off.
Could this be programmed, or any other programming with the same effect?
If something like that would work, that would be great!
Hi Holger,
Great to see that you are successfully flying your Tandems!
I think you can accomplish this by adjusting the SERVOX_TRIM for the forward swashplate all in one direction(you would have to account for any servos that are reversed). This would raise or lower the forward swashplate independent from the aft swashplate.
Do you feel that currently you pitch control is too sensitive. In my recent experience with a side-by-side configuration (tandem turned 90deg) I found that the amount of differential collective pitch might be too high and makes the axis difficult to tune because the response is very sensitive. I would have to see your settings and some of your flight logs.
I am not sure how you are using the transmitter to put the tail in “standby”. I have recently been working on something that would fix this issue. I hope to get it into version 4.1.0.
If you are interested, I could give you a copy of it based on the latest version 4.0.5. That way you could provide feedback on how well it works for you.
Hello, Bill,
I thought you would answer! Yes, in the meantime I have the matter quite well under control.I often write reports in the German ROTOR-Helicopter-Magazin, also a lot about tandems. This year I also wrote about flight controllers in helicopters, not many people do that in our country yet. Felix also liked it very much. Now I am supposed to write another report, more about the practice. So your activities are included. The easier and better you can use the flight controller in a normal helicopter, the more it is accepted.
swashplate trimming
with ServoX_trim I can only trim the individual servos, not the 3 servos together at the same time ?
Tandem
We currently have only one system for tandems, the BAVARIAN DEMON 3SX, and they all fly it.But it can only control 2 servos per rotor. So I have already tried this with a 3rd servo. I controlled it directly from the transmitter only from the throttle. When -nick- steering (pitch) the 3rd servo did not move. So the swashplate did a nodding movement at the same time. That was very good to fly.
In my current tandem-project with PIX I controll the 3rd servo in the same way (switchable from MOTOR 3/6 to RC THROUGH). the first flights are coming now. I will report.
My first tandem electronic, also for 3 servos (10 years ago), was able to adjust the nodding movement from 0 - 100%. And you could trim both swashplates separately in altitude.
PIXHAWK / ARDU is now the only system with 3 servos.You should take advantage of this. And as I said - it would be original scale-like. And it would convince the ROTOR-readers even more.
Standby
I have STANDBY on a switch once. So I can test and adjust the mechanical setup at any time without heading-hold.
Additionally I mixed the switch with the throttle stick in the transmitter. When Throttle down = STANDBY on. If I raise Throttle slightly = STANBY off. So nothing happens on the ground and when I lift up I automatically have heading-hold.
I would like to have a look at your new file. You can send it to me: mediapico@gmx.de
After thinking about this, trimming the height of the swashplate can be done by adjusting the H_COL_MIN, H_COL_MAX, and H_COL_MID and H_COL2_MIN, H_COL2_MAX, and H_COL2_MID relative to each other. So determine the values of H_COL_MIN, H_COL_MAX, and H_COL_MID and H_COL2_MIN, H_COL2_MAX, and H_COL2_MID that give the collective pitch values you desire for the forward and aft swashplates. After that, adjust the aft (#2) swashplate up or down to trim the aircraft in the pitch axis. You want to change H_COL2_MIN, H_COL2_MAX, and H_COL2_MID by the same amount of PWM to trim the entire swashplate up or down.
I will have to think of how to implement a trim feature in the code but this should work for now.
Based on my experience with full scale tandem helicopters, the longitudinal cyclic or pitch nodding of the swashplate is not used as much for pitch control but more for maintaining a level fuselage throughout the airspeed envelope. I would be happy to put it in but my vote would be to have it be scheduled with forward airspeed. I have another project that may benefit from this but for a side by side configuration. So I will see and let you know.
What board do you want me to provide my firmware that will help with the heading hold on the ground? Pixhawk 4 mini?
Currently I have H3 - 120 degrees.
but you can determine the positions of the servos freely, I already had 90 degrees.
Swashplate trimming
The way you described it, I do it too. But it would be much easier if you could trim the swashplate(s) with only one value (= nice to have…).
If you don’t use the modes with ALTHOLD, you could also do it with the throttle center trim in the transmitter, of course only with the single helicopter.
I did not quite understand what you mean, also in the translation.
In a technical description for a real tandem (CH47, CH 46…and older) I saw that it is controlled like this.
And as I said, I have already tried it successfully.
But if you work on it in another project, maybe it will work here, too. let me know.
It explains what the purpose of the swashplate tilting is in the pitch axis. On page 4, it discusses how the swashplate tilt is used to keep the fuselage at the same pitch attitude while varying forward airspeed. On page 6, it discusses the use of differential collective pitch to control the pitch axis.
I can add the feature that you requested but I don’t feel it makes the aircraft behave like the actual CH-47.
When I have time I will add the trimming feature. I think this would be useful for all dual heli types.
Hello Bill,
the document is great, now I have understood it
I repeat: DCP (Differential Cyclic pitch) is for the inclination of the fuselage - as we had it before
the nodding motion of the swashplates is for increasing the speed while keeping the pitch of the fuselage constant.
This would be good in the model helicopter too. Since we can only control this together, my suggestion comes very close. And it would also be true to the original.
And maybe later you would have the possibility (by a separate control of the nod function) to roll only on the rear wheels, as seen on flying days.
Hello, Bill,
i think i have the ultimate programming idea for the tandem in the air and on the ground
normal
the pitch-control as suggested by me = mixture of DCP and nodding swashplate.
fast flight
Using a switch with an RCx_Option (let’s call it “speed”) the following happens: The longitudinal inclination of the fuselage is limited to “0” or only a very small inclination. When steering -nick- (pitch), the swashplates only make a nodding motion, no more DCP.
Alternatively, this function could be activated (soft) via the airspeed.
rolling on the rear wheels
Using a switch with an RCx_Option (let’s call it “wheel”) the following happens:
DCP and CP of the rear swashplate are deactivated. If you now push the throttle-stick upwards, the fuselage lifts in front and stands only on the rear wheels. About 20 - 30 degrees. This pitch is held by the FC. if you now steer nick(pitch), both or only one swashplate will make a nodding movement. this will push/pull the helicopter forward or backward and it will roll forward or backward on the rear wheels.
You could also set it up so that the helicopter automatically lifts into the 20 - 30 degree position when you turn the switch.
Of course this function may only work on the ground with throttle minimum.
Maybe the front swashplate would have to be corrected a little bit in the nick direction, because otherwise it will pull backwards the heli at 20-30 degrees from the fuselage.
Of course everything sounds complicated for the programming, especially the rolling.
but maybe no. 1 and 2 will work.
just to be clear, DCP is Differential Collective Pitch and is currently used in ardupilot to control the pitch motion of the helicopter for a tandem.
this would be very unique feature to tandems. I want to ask a friend who flew H-46’s to see how they did the “rolling on the rear wheels” maneuver that you describe. I can’t make any promises but I will try to work on these when I have some time.
Holger,
I spoke with my friend. She said that they did not normally taxi with the nose wheel off the ground. Because the shaft of the rotor hubs on the forward and aft rotors were tilted forward, taxiing would be difficult. This was possibly due to the aircraft wanting to accelerate forward with the shaft tilt built into the rotor system. They actually had to use a setting in the cockpit that put aft longitudinal cyclic only on the aft rotor system. Then the would use lateral cyclic stick to steer the aircraft. They did not use the pedals for steering on the ground. The only time they performed what you describe above where the nose wheel is lifted off of the ground during taxiing was to taxi backwards.
In your case, I guess we could implement a taxi mode or feature. I suppose the longitudinal cyclic could be used to tilt the swashplates forward and then it would require the pilot to pull aft on the stick which would pull the nose off of the ground.
How do you steer the aircraft on the ground with the current software? I would think a combination of lateral cyclic and pedals in the same direction would cause only the fwd swashplate to tilt laterally and keep the aft swashplate level and could steer the nose of the aircraft. But I guess if you don’t have the rotor shafts tilted forward on your model then you may not be able to increase the aircraft speed on the ground.
Hello, Bill,
yes, that would be a very special feature for a tandem.
I think I understood the description of your friend.
Rolling only on the rear wheels is not a normal procedure, it is only sometimes shown as a highlight on flying days.
I have only seen it on the Viedos on the CH47, not on the CH46. (but it should be possible there as well).
Normal rolling on the ground: the CH46 has only one wheel in the front, that should be controllable.
On the CH47, the two rear wheels are steerable. I have implemented this in my big Chinook. In the software I did nothing else, I steer it with “side left/right”.
If you can / want to program something there - gladly.
but the other, 1st and maybe 2nd suggestion would be much more important. If you could at least implement the 1st suggestion soon, I could include it in the current report.
In Germany are more and more people flying the CH46 LABRADOR with 2,20 m rotor and 22kg weight. They are also often on flying days, where tandems are always something special.
So your efforts would not only be for me, I think this could be interesting for many pilots. Adding a 3rd servo would not be a big problem.
Here I wait for better weather to test your addition.
Its talking about preset transmission tilt not cyclic inputs here I think. Meaning the axis of the forward rotor is 9 degrees and the axis of the rear is 4. On models the front and rear axis are almost always the same to simplify the mechanic of them. Normally the whole mechanical structure on a model has some amount of tilt in relation to the body of aircraft. Lots of heli both manned and models have a nose up look of the body at neutral attitude so that they are more aerodynamic when the rotor disk is in what ever forward flight angle they decide makes most sense for cruise.
Tilting the swash plates forward to the rotor shaft axis will still create a forward roll torque on the airframe and cause the heli to pitch continuously until the rotor disk and shaft axis are inline again.
I would guess the different transmission angles on a ch47 are for body clearance because as I understand it the net thrust angle relation to the fuselage angle is ultimately all thats going to affect the orientation is fast forward flight. So the options on the chinook would be both rotors are 6.5 degree tilt and loos clearance for both over the center of the fuselage or do what they did and set the front and rear so they have a net angle of 6.5 but both are tilted away from each other to gain clearance possible the amount dictated by how much they expected the rotors to flap or something. If you look at other tandems they build in the center clearance other ways with body shape.
This counter to each other front and rear swash tilt I guess could be setup on a model for realistic look. I cant think of why this on its own would give any performance though. Fore aft stability might be worse since there would affectively be less dihedral but I also don’t know how much that really affect helis at all.
The main thing I can think of that having different fore aft tilting swash would do is change pitching speed and authority. Since different parts of the rotor have a lot different leverage on the airframe for a tandem where the forces are applied might make huge differences in how fast and easy they pitch. The differential collective pitch of the two rotors means part of the differential force of both is overlapping in the center. With swash tiling all of the lifting force could be moved all the way to the forward ark or all the way to the aft ark giving more authority for the same amount of drag on the rotors. I don’t actually know just analyzing based on what I understand.