Clawjet Helicopter Design (low-friction bearings needed)

[Noname117]

(Edit: If you want to see the part about needing the bearings, go look at the more recent posts. March 9th or later).

 

The time was in the middle of last summer. @Jon144 had built his beautiful helicopters while I had nothing to show for it (We've competed with designs in the past and I usually lost). I didn't really like the fact that stock KSP helicopters had to detach their rotors to work, requiring switching crafts to adjust throttle, and set about on the quest to see if a claw based bearing could work (tests in previous versions showed it couldn't, but I wanted to see if anything had changed). Forcing the claw to stand upright did not work, but then the entire rotor assembly tipped upside-down and started spinning like a charm. And thus I had my idea.

 

The Clawjet Rotor

So why did I call them "clawjets?" Simple. The claw is the most important part of the bearing and it uses tip-jets to spin the rotor.

 

 

The Clawjet rotor's main advantage is that the power of the jets can be adjusted mid-flight without having to switch craft and lose control of the helicopter for a few moments. With a properly placed fuel tank its possible to fuel the rotor from fuel stores aboard the main craft, and the blades can clip through the main craft if you wish.

The main disadvantage to it is that it's complex to set up prior to the flight and likes to break at every opportunity. That's the main reason why I never actually built a proper non-experimental helicopter using one to upload to the forums, and decided to just put the whole project on hold.

So how does it work?

Glad you asked and I didn't type out the exact response I wanted you to have in big blue text above this statement!

 

The setup is effectively made up of 3 parts: The claw, the rotor, and the lock (which itself is made up of 3 parts). The rotor is placed upside-down directly above the claw. It is then detached, landing center on the claw. The claw is then set to "free pivot," allowing the rotor to tip upside-down (2 upside-downs makes a right-side up, so this is good). The rotor is then locked in place with the lock. Here's a picture:

In this example the decoupler is between the fuel tank and the claw; its just that the decouper is actually attached to the structural support on the right side of the image and majorly offset over (more than the stock settings allow).

Also notice how the claw is tipped back slightly. This is because the claw can't rotate fully upside-down, it just gets really close to doing so. I think this is the reason why the rotors destroy themselves at every opportunity.

 

So then how does the lock work?

Ok, the lock is divided into 3 parts, which I shall call the "rotor lock," the "detached lock," and the "attached lock." Here's the same picture with the lock parts highlighted instead:

I've named the 3 lock parts in this format because it's descriptive of where they are. The rotor lock is attached to the rotor. The detached lock gets detached from the craft. The attached lock stays attached to the main craft.

The attached lock and the detached lock form your typical bearing setup. The rotor lock winds up in the same place as the attached lock once the rotor has been rotated upside-down by the claw. The detached lock then detaches, locking the rotor lock and attached lock in the same place. Essentially the detached lock holds the rotor lock in place, while the attached lock holds the detached lock in place. The rotor lock is "locked" to the main craft, and thus the rotor shouldn't tip when the helicopter decides to lean.

EDIT: Make sure to lock the pivot on the claw once the rotor is rotated into place. Free the pivot again when you detach the detached lock. This will make sure that the rotor lock locks properly with the detached lock.

This lock setup will work with many types of bearings, and I have indeed experimented with many types of bearings. If you want to get especially creative you could have the rotor lock and attached lock be in different places; as long as the detached lock is locking the rotor in place and allowing it to spin while the attached lock is locking the detached lock in place you should be fine.

 

Success?

Flight was achieved with several prototypes, and at least 1 even landed, although they did all encounter many issues with the rotors breaking off when trying to pull maneuvers, along with the helicopters being practically uncontrollable. Eventually I settled on counter-rotating prop designs as the best option, but even that did not remedy the problem. Eventually I abandoned development because the challenges to overcome were greater than my interest in continuing development.

Anyways, have some pictures:

 

The Future?

The game has gone through a couple of updates since I experimented with these rotor bearings, though I would think they'd still work in the current version. Autostrutting might do interesting things for the clawjets, and perhaps allow them to actually succeed. However, I'm definitely done with them after all the trouble they gave me and a waning interest in KSP. However, now that I'm no longer working on these things, I thought I should probably at least share my progress to the forums.

So I guess that's where you guys come in. Try some things with these clawjet bearings and see if you can get them to work properly, or if the concept is so bad that it's unfeasible.

 

And since it's the spacecraft exchange I better post a download link

Below is the download for all of the experimental helicopters and rotor setups I have. All of my research into these rotors is in there, so hopefully you can go from where I left off.

https://www.dropbox.com/sh/sjsjqxn4yh5xv9o/AACOVuObf-mRivggylhRmwsTa?dl=0

Edited March 10, 2017 by Noname117

[klond]

3 hours ago, Noname117 said:

Flight was achieved with several prototypes, and at least 1 even landed

 Average day in KSP .

Thanks for sharing.  I think this is worth looking into.

[THE NARWHAL]

Nice idea, i'll do some testing and make a few prototypes, and get back to you on what i find.

[Azimech]

Nice. Maybe the concept is useful for other things.

 

[Noname117]

46 minutes ago, Azimech said:

Nice. Maybe the concept is useful for other things.

 

Probably quite hard to be useful for much else. It would be hard to precisely aim anything with it, and the claw must start out pointing upwards and the rotor do a full 180.

It's basically only useful for spinning jet powered things, and even then only really things in which the claw can be pointed up on. Helicopters are perfect for it, just about everything else would require large amounts of engineering to get it to work.

[Majorjim!]

If this can be spun with jets pushing on parts internally this will be awesome! 

 Great find man, I'm sure folks here will get to grips with it quick. 

[Noname117]

1 hour ago, Majorjim! said:

If this can be spun with jets pushing on parts internally this will be awesome! 

 Great find man, I'm sure folks here will get to grips with it quick. 

Likely very hard to do, as I don't think jet engines can push on stuff which is part of the same craft as it, but you can still "hide" the jet engines within the hull of the craft without any real danger

[Majorjim!]

37 minutes ago, Noname117 said:

Likely very hard to do, as I don't think jet engines can push on stuff which is part of the same craft as it, but you can still "hide" the jet engines within the hull of the craft without any real danger

Lol yeah my bad, im just so used to choppers being two craft! 

 :-)

Edited February 4, 2017 by Majorjim!

[klond]

2 hours ago, Majorjim! said:

I'm sure folks here will get to grips

Claw.. grips.. Shame tryin to sneak one past us like that

 I took a whole engine, bearings and all, with reaction wheels and tried to use the claw as a limited swivel.  Really sad - it's almost uncontrollable, like gyroscopically weird AF.  Would have been neat.

Note: It was still 2 craft here.  'Haven't been able to do the flip thing like the original post says.

 

 

[Noname117]

8 minutes ago, klond said:

Claw.. grips.. Shame tryin to sneak one past us like that

 I took a whole engine, bearings and all, with reaction wheels and tried to use the claw as a limited swivel.  Really sad - it's almost uncontrollable, like gyroscopically weird AF.  Would have been neat.

Note: It was still 2 craft here.  'Haven't been able to do the flip thing like the original post says.

 

 

Take a look at the dropbox  folder I provided. It is full of the prototypes I built and with a little experimentation you should be able to reverse-engineer them.

[Azimech]

1 hour ago, klond said:

Claw.. grips.. Shame tryin to sneak one past us like that

 I took a whole engine, bearings and all, with reaction wheels and tried to use the claw as a limited swivel.  Really sad - it's almost uncontrollable, like gyroscopically weird AF.  Would have been neat.

Note: It was still 2 craft here.  'Haven't been able to do the flip thing like the original post says.

 

You probably need more distance for the gyros from the center. I'm guessing the forces from the rotor are greater than the gyros can compensate. Besides that, the center of the gyro torque is offset the center of rotor thrust. Try a gyro quad configuration, offset 4 meters so all forces go through the center. Might be an idea to raise or lower the center of the gyro forces above the center of thrust or below the pivot as an experiment. Also the mass of the gyros on long arms will tend to stabilize a bit due to inertia. Lowering blade angle might help too.

 

Although I like these experiments I'm working in haste to finish my manual. I want to upload my newest helicopters today, exactly two years after my first turboshaft helicopter.

Edited February 4, 2017 by Azimech

[The_Rocketeer]

Too bad, @Noname117 I think the claw has been patched, and will no longer invert. It looks like it tips to no more than 45 degrees from it's original alignment. I'm also not getting it to spin - seems like there's a lot more resistance to vertical-axis rotation than the other 2 axes, which could have something to do with Klond's strange torque feedback. There may still be some potential here for bearing design, but the claw has never been intuitive (or safe) for uses other-than-directed.
 

[Noname117]

Ah dangit, these were quite promising and its a shame to see that gone

[The_Rocketeer]

Agreed, I was looking forward to integrating the claw into a propellor plane.

Given the limited range of motion, I feel the claw free pivot might help cut down on part count for piston-car valve timing or such, but these are quite massive mechanisms anyway.

[klond]

4 hours ago, The_Rocketeer said:

it tips to no more than 45 degrees from it's original alignment

 I wonder you put a claw in a claw would you get 90 degrees movement for engines you want to pivot?  Image now a giant snake of grabbers and reaction wheels.

I also noticed it makes a really good pendulum hinge, like nearly frictionless.

[Azimech]

14 minutes ago, klond said:

 I wonder you put a claw in a claw would you get 90 degrees movement for engines you want to pivot?  Image now a giant snake of grabbers and reaction wheels.

I also noticed it makes a really good pendulum hinge, like nearly frictionless.

A typical klond project I assume: make a chain out of them.

;-)

[Jon144]

Yeah. Noname referenced me here. 

I did my own tests on this concept after he brought it to me. My results were nearly identical to klond. I found that a claw could never be turned perfectly upside down so the rotation would become unstable at about any speed. It's so sad because the design is essentially friction-less. I was able to use them on a modified Kinchook but the strange wobble eventually shook it apart. 

So if they did get patched now and don't turn beyond 45 degrees it's not a total loss because after days of testing I couldn't find any significant practical use as an engine bearing over conventional designs so I let Noname continue the work. The main problem was having to invert the rotors before flight which means it was hard to design anything that could actually start on the runway to begin with.

Edited February 7, 2017 by Jon144

[klond]

10 hours ago, Jon144 said:

The main problem was having to invert the rotors

 I want to recognize that @Noname117 had some really creative ideas to get those props to flip.  Offsetting+gravity, inverting and thrusting up, dual rotors, etc.  It's worth downloading some of the craft files to see the effort to make this bug work in his/her favor.

[Noname117]

28 minutes ago, klond said:

 I want to recognize that @Noname117 had some really creative ideas to get those props to flip.  Offsetting+gravity, inverting and thrusting up, dual rotors, etc.  It's worth downloading some of the craft files to see the effort to make this bug work in his/her favor.

And with that all being said a few designs did fly for short periods of time, with I think one being controllable and stable enough for extended flight and landing. Still a shame that they fixed the bug...

At the same time, I'm wondering if stacking 2 claws on top of each other to get a 90 degree tilt would work. I remember odd things happening with just a 45 degree tilt, but I'm wondering if it's possible to cancel out some of those odd things by stacking 2 claws. Or it may just make the problem worse.

Edited February 7, 2017 by Noname117

[Noname117]

Bearings needed!

So I'm experimenting with a new clawjet design right now, and I have achieved (relatively slow) rotation. I've tried out Jon144's bearing, but it's not working (do note the problem may be with the design and not the bearing).
 

So I want to ask the good people of the forums, what's the smallest bearing with the least friction with the lightest spinning part you can create which is still stable enough to use with helicopter blades (excluding tip-jets)? I need to try out different bearing designs here to figure out if the problem lies with the claw or the bearings.

 

[EpicSpaceTroll139]

Basic Turboshaft

E-50A Triton

This rotor and heli demonstrate the virtually frictionless, reasonably durable, and exceptionally lightweight antenna-solar panel bearing, which was originally based off Majorjim's antenna-solar bearing. These bearings are used in pairs, with each consisting of a disk of static solar panels (attached to the heli) with a small hole in the middle in which a communitron 16 antenna (attached to the rotor) spins freely. 

In the E-50 one of the bearings can be seen to be replaced with an experimental lower part count bearing which uses 4 octagonal struts instead of a disk of solar panels.

Edited March 10, 2017 by EpicSpaceTroll139

[Azimech]

Please make sure you don't use Kerbal Joint Reinforcement as it negatively impacts the maximum angular velocity of bearings. Reasons unknown. KJR is wonderful, just not for stock bearings.

[Majorjim!]

Well, the smallest possible bearing/hinge is my thermometer design. All small hinges you see are based on my original design. 

Having said that, looking at the craft you posted I think you want something a little bigger.  

 

[Noname117]

I'm looking for something where the spinning part is lightweight with very low friction. The new design is not the ones above and functions differently

[EpicSpaceTroll139]

10 minutes ago, Noname117 said:

I'm looking for something where the spinning part is lightweight with very low friction. The new design is not the ones above and functions differently

The antenna-thermo and antenna-solar bearings mentioned by @Majorjim! and I above are very lightweight and have virtually zero friction.

Or are you looking for other bearings that keep the rotor as part of the ship?