The community collective ORNITHOPTER challenge

[Klapaucius]

The challenge: build a functioning ornithopter using stock and DLC parts.

 

 

This is a different kind of challenge. It is more like a collective "put our heads together" and see if this is possible.  I've spent a heap of hours over the past day and a half playing around with this with very limited results.

 

 

Here are the challenges and barriers I have discovered so far:

1.  Despite being able to flap like a hummingbird in the SPH, as soon as you put the craft on the runway, physics takes over.  The traverse rates of the hinges and servos put a speed limit on how fast you can flap. I've found a 2.5 second cycle is about as good as it gets.

2. Stability is a huge challenge. Big reaction wheels are heavy, and I don't know how many I actually need to get this thing to hold itself steady.

3.  Duna seems to be a more forgiving test site due to less gravity, despite (and perhaps, ironically, because of) the thinner atmosphere.

4. Increased flapping speed might be achievable with a linkage to a rotor, rather than using the hinge. However, I am not sure how we could increase the wing rotation speed for the upward stroke.

 

Furthermore, I am not convinced that KSP physics supports this. Is my "flight" on Duna due to the downward pressure of the wings on the atmosphere, or is it simply the craft reacting to a lot of torque and bouncing off its gear?

 

A few frustrating workflow things:

1. If you alt-click to duplicate a part, it's motion is not replicated in the controller. Every time you add a set of wings, you need to re-input the keyframes.  (I don't know what you guys call them, but the controller editor appears to be modeled on keyframe editors like in Adobe Premiere and After Effects, so that is what I am going to call them.)

2. There is no way I can see to copy and paste keyframes like you can in the Adobe software.

3. I have not found a way to delete keyframes, rather I have to work around the extra one I accidentally placed unless I want to delete the whole sequence and start over for that part.

 

Below is a useful Youtube video on how birds fly.

Spoiler

B

 

So, the challenge: Build an ornithopter either individually or collectively, or determine conclusively that the KSP physics does not support this type of craft.*

 

* UPDATE: @HebaruSan posted the link below, so it is possible. But how can we make it better? Can we make something viable?

 

 

Edited June 4, 2019 by Klapaucius

[HebaruSan]

Here's the one from the SQUAD announcement post:

On 5/22/2019 at 11:04 AM, St4rdust said:

Seeing the great potential for shenanigans, Bill challenged Jeb and Val to try and "fly like a bird" using new hinges currently under R&D at KSC. Never ones to back down from a challenge, they decided to take him literally. The end result - the Kerbithopter Mk0.

This extremely unsafe contraption uses 2 G-11 hinges per wing to flap like a bird, at the speed of a butterfly, around KSC in a barely controlled but highly entertaining fashion.

Build by @Yargnit

 

 

 

[Klapaucius]

9 minutes ago, HebaruSan said:

Here's the one from the SQUAD announcement post:

 

Cool, well that answers that question

 

I think I have been overthinking this.

Edited June 4, 2019 by Klapaucius

[HebaruSan]

22 minutes ago, Klapaucius said:

Cool, well that answers that question

 

I think I have been overthinking this.

Well it's clearly right on the border line between possible and impossible*. Even that example is just barely staying up and not going very fast. Maybe someone can figure out how to do it better.

 

(*: I see this as a kind of vindication of the tuning for the robotic parts. Something that only sort-of works IRL shouldn't be easy in-game, but it'd be nice if it wasn't completely impossible either.)

[neistridlar]

It is indeed possible. Here is an other variation. It is just barely climbing. 

 

[Klapaucius]

32 minutes ago, HebaruSan said:

Well it's clearly right on the border line between possible and impossible*. Even that example is just barely staying up and not going very fast. Maybe someone can figure out how to do it better.

 

(*: I see this as a kind of vindication of the tuning for the robotic parts. Something that only sort-of works IRL shouldn't be easy in-game, but it'd be nice if it wasn't completely impossible either.)

I'm guessing based on the angle those wings traverse that it was launched from the air somehow, or perhaps off a high platform like @neistridlar did.

26 minutes ago, neistridlar said:

It is indeed possible. Here is an other variation. It is just barely climbing. 

 

In thinking about this, I think this works a lot like a helicopter, just with back and forth vs rotation motion.

 

It's a cool VTOL concept.  

 

I'm still clutching at the bird-like straws.

[doggonemess]

Maybe I can adapt the XAF-163S Thunder Chicken.

 

Actually, probably not, since control surfaces seem to behave oddly on hinges.

[Tapeta]

I was experimenting with ornithopter design before with very poor results, so I was going to move on to something else when I noticed this challenge.

When I saw @neistridlar 's post I was very surprised how nice it worked, and it was eureka moment for me: yes of course use Lifting force not drag.

So I created my own contraption which is heavily inspired by @neistridlar 's . I used well-tried Kerbal concept -use "moar boosters" and I placed servos on top of an alligator hinges so I can tilt them individually and experiment with it. It is still very slow and hard to control.

 

[Klapaucius]

@neistridlar, @Tapeta.  I'd be keen to see how both of these perform on Duna.

[Tapeta]

On 6/7/2019 at 2:03 PM, Klapaucius said:

@neistridlar, @Tapeta.  I'd be keen to see how both of these perform on Duna.

I tried it on Duna, but thought luck it doesn't fly. Those servos are on its limit - I can't put more weight on them (more wing area), If I try they just go completely out of rhythm.

 

 

[Lapis]

I bet it would work on eve, with the thicker atmosphere.

 

[Pds314]

I've been trying a different approach without much success. Hummingbird/dragonfly style flight seemed like it would be harder than simply flapping hard enough to generate negative drag coefficient like a prop.

 

Still haven't managed to do that though.

[SRB]

3 hours ago, Pds314 said:

I've been trying a different approach without much success. Hummingbird/dragonfly style flight seemed like it would be harder than simply flapping hard enough to generate negative drag coefficient like a prop.

 

I’m confused-I thought bird insect-style flight was the only way.

Could you please explain?