Floppy hinges and rotors

[AngrybobH]

I am trying (and nearly succeeding) to build a helicopter with the new robotic parts because of a deep love for these flying contraptions. I have collective and anti-torque(tail rotor with collective) somewhat working decently. Cyclic seems impossible without some form of rotational angle feedback system to control it. So, I went with a tilt rotor type of design. On the 47th (or so) iteration I have scrapped the tilt mechanism and rely solely on reaction wheels for attitude control. I now have something that is nearly able to fly.

I have some issues, though, and I'd like to know if anyone has any ideas for combating them with stock parts. Of course, mods may be necessary to make any of this work. I haven't flown fast or long enough to make a determination if flapping or lead/lag hinges are needed. Anyone care to guess if KSP might inadvertently simulate uneven lifting from coriolis (and other) effect(s)?

Here's my list of issues.

  The large rotor does not make enough power. With a collective pitch of 0 it has a real hard time getting to 300 RPM. Any stress or over pitching the collective (more than ~5 degrees) can stall it. It seems to slow down considerably with any amount of mass attached. putting a rotor on a rotor does not work.

  With rotor blades and collective pitch motors attached in symmetry, the main rotor and tail rotor wobble as if they were unbalanced when getting near 300 RPM.

  You can't put an increment amount into a single keystroke for better fine control.

  Parts pull away from their mounting points when nearing ~350 RPM (see pic, tail rotor in spinning ~400 RPM)

  I scrapped the tilt rotor because all the hinges are floppy. It seems ANY amount of force against them causes a change in angle and the imprecise control of them cannot counteract this quickly enough.

Also, if you are like me and trying to push these parts beyond there apparent design conditions and don't know how to solve it, share your current engineering nightmare. Perhaps something can be learned from our experience with the new parts.

[Geschosskopf]

Well, @Brikoleur has made some great helicopters with the new parts.  You might want to read his excellent tutorial, a link to which is at the bottom of this post.  I myself have, after much bloodshed, gotten reasonably proficient at making autogyros.  Both types of rotorcraft face many of the same issues, although things are more complicated with helicopters.  I would say in general that both myself and Brikoleur have encountered the same problems as you and have found them unsolvable.  So we've worked around them in various ways.

As to your specific issues:

• Yes, hinges are not up to the aerodynamic forces of rotor disks.  The disks go where they want and the hinges can neither resist this nor move them back.  Thus, tilting rotor disks aren't possible.  You have to go with a fixed mast and find some other solution for cyclic control.
• Rotor mass must be minimized to get sufficient RPM, either when powered or autorotating.  So in general, it's better to have fewer blades each consisting of a single control surface.  If you need a wider rotor (to increase the amount of it going fast enough to matter), put stacks of cubic octagonal struts between the rotor and the control surface.  Having a cubic octagonal strut node-attached to the rotor also reduces the amount of "stretch" in the blade, which is more a thing with surface-attached parts like the control surfaces.
• Because tilting the rotor disk isn't possible, cyclic control can be achieved by allowing the control surfaces that comprise the rotor blades to work as normal control surfaces in the pitch and roll axes (but not yaw).  This is rather imprecise and irregular, however, so you need lots of SAS to make the thing easier to fly.   If the craft is small enough,, you can have totally fixed rotor blades (all axes disabled) and just use SAS torque for roll and tail-mounted elevators for pitch.
• Brikoleur has had good success using the smallest servos for collective control.  Node-attached to the rotor, between it and the cubic octagonal struts.
• Tail rotors (or even RCS thrusters) are pretty much impossible to control manually.  They might work with a kOS script tying their output to that of the main rotor but that leaves the issue of making them.  Without mod parts, you can't really make a small enough tail rotor.  If you got the RCS route, probably the best solution is the air jets in @Angel-125's Buffalo mod.  Then you don't have to worry about running out of monopropellant.
• Because of the problems with tail rotors/jets, using counter-rotating main rotors is the most viable option for KSP rotorcraft.  You can either have them coaxial or separated although with autogyros, coaxial is best because only uniform airflow over both rotors keeps them turning at the same speed, which is the whole point of having 2.

And FWIW, here are my much less-informative and rather poorly written notes on autogyros:

 

[Goody1981]

3 hours ago, AngrybobH said:

You can't put an increment amount into a single keystroke for better fine control.

You have made a far more successful helicopter than I.... but I might be able to help with this! Haha...

I believe hitting Caps lock (not TAB as I originally said, thanks @Brikoleur  ) turns on fine control for the WASDQE pitch/yaw/roll

Edited June 7, 2019 by Goody1981

[Guest]

Chiming in: I don't think it's possible to do entirely without reaction wheels specifically because the lack of direct cyclic control; control surfaces reacting on the spinning disk are a poor substitute and they stop working altogether once the RPM goes high enough as they don't react quickly enough (I also think the physics calculations just can't handle them). 

I'm not very far with rotor optimisation research but it's a really interesting field. For one thing, you have to decide what you're optimising for. If you're going for power/efficiency, then what Geschosskopf says is exactly right: minimize the number of parts and go with the biggest disk your chopper can handle.

However if you're going for handling -- which is my #1 priority almost all of the time -- it gets more complicated. Two-blade rotors tend to make for a bit of a bumpy ride, and you'll want to limit RPMs somehow or the rotor will spin so fast it'll make your chopper unresponsive to controls (and possibly eventually shake itself apart). I think I've discovered some general trends and relationships but mostly I do this by trial and error. I try with 2, 3, 4, or 6 blades and pick the best-handling one, then I make the blades longer or shorter to try to find an optimum, then I try again with 2, 3, 4, or 6 blades, and repeat until I'm happy or bored or both. Varying the number and type of control surfaces you use is another axis. Fortunately final tuning -- collective restrictions and control authority -- are pretty simple and you can leave that until the end, you will notice if you've got a manageable rotor by then.

I use six-blade rotors on my flagship Super Kadzook because the ride is so smooooooth, flying that thing is like bathing in warm treacle. I could probably increase lifting power by reducing the number of blades and control surfaces and making them bigger but I haven't really attempted that; instead I just specced my goal (lifting a Rockomax 64) and once I got there I focused on handling.

https://kerbalx.com/Brikoleur/BAK-68-Super-Kadzook-C

As to single main rotor designs, I've made some that are barely controllable but no more than that. I ditched the idea of controllable pitch on the tail rotor, instead I trust the standard KSP yaw control to do its thing, with a fixed pitch otherwise adjusted to more or less the average torque of the main rotor. Then I piled on reaction wheels to paper over the rest. I haven't published the craft because they don't fly all that well, you can get them to lurch in the general direction you want to go as well as go up and down but they are a handful and not much fun to fly. I'm hoping that one of these days I'll have one that works better but am getting increasingly pessimistic. Coax twin rotors are the way to go in KSP, at least until and unless somebody updates the pitch/yaw/roll controls to properly control cyclic on a rotor.

2 hours ago, Goody1981 said:

I believe hitting TAB turns on fine control for the WASDQE pitch/yaw/roll

Caps lock.

Edited June 7, 2019 by Guest

[Goody1981]

46 minutes ago, Brikoleur said:

Caps lock.

Yes that’s the one! I’ll edit my post  (I think I subconsciously knew this which is why I wrote TAB in all caps haha... thanks mate)

Edited June 7, 2019 by Goody1981

[Guest]

All right, here's a single-main-rotor that's somewhat controllable, I got it on the VAB helipad at least.

https://kerbalx.com/Brikoleur/BAK-666-Deathtrap

[Snark]

Moving to Breaking Ground Discussion.

[Geschosskopf]

7 hours ago, Brikoleur said:

I'm not very far with rotor optimisation research but it's a really interesting field. For one thing, you have to decide what you're optimising for. If you're going for power/efficiency, then what Geschosskopf says is exactly right: minimize the number of parts and go with the biggest disk your chopper can handle.

However if you're going for handling -- which is my #1 priority almost all of the time -- it gets more complicated. Two-blade rotors tend to make for a bit of a bumpy ride, and you'll want to limit RPMs somehow or the rotor will spin so fast it'll make your chopper unresponsive to controls (and possibly eventually shake itself apart). I think I've discovered some general trends and relationships but mostly I do this by trial and error. I try with 2, 3, 4, or 6 blades and pick the best-handling one, then I make the blades longer or shorter to try to find an optimum, then I try again with 2, 3, 4, or 6 blades, and repeat until I'm happy or bored or both. Varying the number and type of control surfaces you use is another axis. Fortunately final tuning -- collective restrictions and control authority -- are pretty simple and you can leave that until the end, you will notice if you've got a manageable rotor by then.

This seems to be a difference between helos and autogyros.  You really want light blades with autogyros because they're just blowing in the wind.  The lighter they are, the sooner they get up to flying RPM, so the shorter your takeoff run.  Autogyros are basically STOL airplanes with rotors instead of fixed wings, so short takeoffs are an essential performance requirement.

But the lightness of the rotors prevents using their constituent control surfaces for cyclic control.  Because there's no motor keeping the rotors turning and the light blades have very little momentum, the increased rotor drag from control surface cyclic deflection VERY rapidly slows the rotors, killing your lift.  Usually on 1 rotor more than the other.  And as you say, once the blades are at high RPM, the cyclic inputs get a bit wonky.  Anyway, the result is usually not only a big loss of total lift but the remaining lift is way imbalanced, so you tumble to your bloody, screaming death.

Thus, for autogyros, I prefer totally fixed rotors and using reaction wheels and airplane-like control surfaces for control.  Which really isn't THAT unrealistic.  There are been a number of real autogyros made that way, and even those with actual cyclic controls still have at least a partial airplane-like tail.  Besides, autogyros are essentially STOL airplanes, dependent on forward motion for lift (they can't hover).  Thus, having airplane-like control surfaces is fitting.

7 hours ago, Brikoleur said:

I use six-blade rotors on my flagship Super Kadzook because the ride is so smooooooth, flying that thing is like bathing in warm treacle. I could probably increase lifting power by reducing the number of blades and control surfaces and making them bigger but I haven't really attempted that; instead I just specced my goal (lifting a Rockomax 64) and once I got there I focused on handling.

https://kerbalx.com/Brikoleur/BAK-68-Super-Kadzook-C

That's a very impressive beast!  I'll be giving her a try this evening.

[AngrybobH]

13 hours ago, Brikoleur said:

handling -- which is my #1 priority almost all of the time

This is the same for me. I try to first fly steady, then be able to pick a direction without crashing, then worry about cargo/load capacity. I am currently on step 2.

13 hours ago, Brikoleur said:

I try with 2, 3, 4, or 6 blades

I have done this to a lesser degree. The heli in the pic I posted originally had 4 but 3 worked much better. This is good advice and I'll tinker a bit more with it. What RPM are your Kadzook rotors turning?(it is a beautiful creation by the way)

14 hours ago, Geschosskopf said:

Tail rotors (or even RCS thrusters) are pretty much impossible to control manually

My tail rotor work wonderfully. I have used 2 designs that worked well but both require an inordinate amount of tinkering. First design used collective pitch tied to yaw (on a joystick for more control) and the second one just has control surfaces that react to yaw (because they are vertical). They both need their starting pitch adjusted to get a near 0 yaw when not giving input. The first design is easy to fine tune trim on the fly but has more vibration and weirdness. The second design is the opposite and must rely on in game trim in flight but is far easier on the pilot in general. A note about both tail rotors, I never alter my RPM limit of my main or tail rotor once I get my 0 yaw point. All control is from collective pitch or reaction wheels. And, yes, that means a change in the main rotor requires all the tinkering to be done again.

14 hours ago, Geschosskopf said:

put stacks of cubic octagonal struts between the rotor and the control surface

This I will try tonight. It sounds like the thing that is going to make the most improvements for me.

[AngrybobH]

So I played around with some rotor designs and got something I can somewhat fly. I then went for a little extra mass. Things were fine until I exceeded about 45m/s horizontal. Asymmetric lift then got me. I then attempted flapping hinges because that's what real helicopters do for that issue. I also tried lead/lag hinges but I do not recommend that. It felt like I was about to summon the Kraken. Here's the current iteration with flapping hinges.

I guess I need to tune the flapping because horizontal flight caused the opposite roll. I'm thinking differing up and down angles might be the answer.

@Brikoleur How are you getting any power out the rotors? I had to write a MM patch to create a copy of the G-11 rotor with triple the power to get this 17t craft to fly.

[Guest]

4 hours ago, AngrybobH said:

@Brikoleur How are you getting any power out the rotors? I had to write a MM patch to create a copy of the G-11 rotor with triple the power to get this 17t craft to fly.

I don’t do anything in particular to the motors, so I surmise it must be the rotor and that weirdly a single motor driving two contrarotating rotors is way more efficient than driving a single rotor. Kerbal physics.

Did you try with smaller discs?

[FleshJeb]

@Brikoleur Agreed. If anything, the rotors are too powerful. It’s all about blade area. I’m doing great with non-contra-rotating tilt rotor craft. Getting between 50-65kN per blade when using 6 to 8 BigS elevon blades per rotor at around 220 RPM. My crappy 50ton test bed “zeppelin” (no wings) craft got forward speeds above 50m/s with the props set to 75 deg above horizon, and blade pitch of 15.

The big issue is that KSP wings have a much lower lift to drag ratio than real ones. We need about 3x the wing area that we think we would.

@AngrybobH That asymmetric lift problem is a bear. I’m sticking with pairs of rotors (contra or non contra) for that reason. If you do ever need more power, I found you can run a piston with a docking port through the rotors and hook them together coaxially (MH inflatable ports work too). I managed to hook a third rotor to a contra prop assembly, with two sets of blades evenly splitting the power from all 3 rotors.

[AngrybobH]

15 hours ago, Brikoleur said:

Did you try with smaller discs?

I did but it doesn't seem to have an effect, positive or negative. I think next test is a counter rotating rotor with no blades and see if that helps the power.

1 hour ago, FleshJeb said:

The big issue is that KSP wings have a much lower lift to drag ratio than real ones. We need about 3x the wing area that we think we would.

That makes me wonder about FAR. I wonder how that handles helicopters.

1 hour ago, FleshJeb said:

I found you can run a piston with a docking port through the rotors and hook them together coaxially

I'm not sure I understand what you are getting at here. Got a pic or a diagram?

[FleshJeb]

23 hours ago, AngrybobH said:

I'm not sure I understand what you are getting at here. Got a pic or a diagram?

Let’s see if I can draw on a phone with ASCII

| = a fixed point on the craft.

O = a rotor. All have their bases to the left.

= = the closed portion of a piston.

- = the extended portion.

b = a docking port facing right.

d = a docking port offset and flipped so that it’s facing left.

Side view:

     |OO d

|O===---b

Pretend both assemblies are on the same axis. There are blades on the 2nd and 3rd rotors. The 2nd rotor is set to the reverse of the other two.

IIRC, this was showing 780kN on the test bench with two * 6 Big S Elevon 2. (65kN * 12) Now that I know more, I’m sure it can be run with two * 8 blades.

Of note, you can retract the piston after the docking and the ports stay docked. You can try various things with the piston to add tension or compression to the system, and that helps get rid of the wobblies.

I was going to try something similar with struts, but I had a bunch of other pieces in the way, and wanted a solution that wasn’t dependent on line of sight.

[Fireheart318]

Floppy hinges fix - Turn off damping.

I'd assume they made the hinges floppy to allow them to function as some sort of suspension or to keep robotic legs from going nuts. Just set it to zero and it should be fine.