Sunday, 27 April 2008

Tetropters

Those who visited the land and sea pages on the site may have wondered why there is no page on the biomechanics of flight, seeing there are pages on 'Walking with...' as well as on 'Swimming with...'. So where is the 'Flying with...' page?

The simple answer is that I am not happy with it yet. There is material on ballonts and on tetrapterate (four-winged) large flyers, and some are in fact shown on the 'air' page. But these organisms are fairly like Terran birds. For a true oddity the tetropters should be considered; in their case, 'oddness' does not reside in them having four wings. After all, Furahan tetrapterates and Terran insects also have two pairs of wings, i.e., four wings, so that isn't really extraordinary. No, the ordinariness resides in the description: 'two pairs of wings', and that says it all: the wings are arranged in pairs. Insects and birds (and Furaha tetrapterayes) all have a body scheme with bilateral symmetry, so their limbs are arranged in pairs.

Not so the tetropters. To be honest, the very first sketches I did of them did show bilateral symmetry. The top animal in the following image shows that primordial tetropter. In fact, their wing movement patterns had already been worked out, and showed a pattern that was exactly the same for each wing. So the wings showed a radial pattern, like a four-pointed star, but the body hadn't kept up. Here is one of those early tetropter designs under attack from a larger tetrapterate.

With their radial wing pattern tetropters were excellent hoverers, and control over wing movements should have allowed omnidirectional movements. But the animals still had a front and aft side. If you have a flight system that allows such tremendous manoeuvrability, why limit it with a limited body design? That is where the concept of complete four-sided symmetry came from. The next sketch illustrates the next logical step in the evolution of the tetroper concept. The top animal has bilateral symmetry, but the bottom one represents a conceptual novelty: the body follows the wings! So the entire animal now shows complete quadriradiate symmetry. By the way, I was taught that you should not mix Latin and Greek roots, and that is explains the switches between 'tetra-' (Greek) and 'quadri-'(Latin). Can't be helped.


I know of no such designs on Earth, although there are animals with five-sided symmetry (starfish, sea urchins etc.). The rest of their body scheme hasn't been worked out in much detail. The design problems are like those of octapods, with eight-sided symmetry. Tetropters too have eyes above as well as below their 'equator', and the mouth parts are all on the ground side. But I haven't done a complete drawing or painting of one yet, so the details remain sketchy - for now.

I found that the flight patterns had much in common with the movement patterns of flying and walking organisms. After all, each leg or wing or flipper moves repetitively, and any gait is no more than a cycle in which the limbs move with a specific set of phase offsets. Although there are an infinite number of ways in which you can do so, only a limited number make much sense. And so tetropters have 'walks', 'trots' and 'paces'. The first pattern shown here is a rather silly one.



I can't show the animated gif here; to see it click here.

Imagine the sphere as the animal's body, seen from below and to the side. The four wings are rotated around their axis to provide lift while going clockwise as well as moving back anticlockwise. While such a pattern does provide lift, it would also result in a net rotation effect on the animal as a whole, which is impractical. To offset that, the rotation forces need to be cancelled by having two wings swing in the opposite direction from the other two. A rough animation of just such an effect follows:


I can't show the animated gif here; to see it click here.

You can see the wings going through one another; in reality they would of course not do so. That's why I need to do a better animation, but it is not as easy to do as it sounds. In the smallest tetropters the wings clap against one another and then to move in the other direction. Many Earth insects use this technique, by the way: the clap their wings together behind their backs, and this apparently generates lift when the wings move away from one another again. In insects, there is just one such 'clap' in each cycle. Tetropters have taken the idea one step further, and the clapping occurs twice in a wing cycle, not just once.

Larger tetropters usually avoid clapping the wings together, so they reverse direction without touching. Whether this is aerodynamicaly better or simply avoids damage to the wings remains to be seen. There is so much to be researched...

10 comments:

  1. Is it possible to come up with a "clap-flap" system with an odd number of wings?

    ReplyDelete
  2. Hi Evan,

    I never expected anyone to comment on a post of over 1 year old.

    Have you seen the later posts on tetropters (June & July 2009)?

    Anyway, the answer is yes, but it will not less efficient that with an even number of wings.

    It is hard to explain without a diagram, but I will try: consider a triopter, i.e. an animal with three radial wings. Remember that every wing moves to and fro, and that it is supposed to clap against a neighbour wing at the extreme positions of its 'to' as well as its 'fro' movement.

    Let's have two of the three wings clapping against one another in their cycle. Exactly half a cycle later, both these wings will be at their other extreme position, where they should clap against another wing. The problem is that there is only one wing left, and it cannot be in two places at once. So you will miss one 'clap'. Apart from providing less lift, the process will provide uneven amounts of lift around the body axis, causing the poor triopter to wobble in the air.

    The same holds for pentopters, heptopters, etc.

    ReplyDelete
  3. Yeah, I've read them all (and intently! :) ), but I posted on this one because it had the most relevant material for my comment (I'll quit if it's a problem).

    The reason I asked is because in my own exobiology project I'm working with tripedal critters, and while I have a smattering of bilaterally symmetrical flyers, I want to have radial flyers as well. I've put together some diagrams of the "clap-flap" method, and I found a solution to each wing not having another to clap against at each extreme of its motion: wings go slower in one direction than the other, and their phase difference allows them to meet up at each end. But it would still have the "wobbling" lift problem that you explained.

    I've been looking at your "silly" method as well, and as far as my own creature is concerned I was thinking of developing its tentacled mouth-parts into flight surfaces that inversely correspond to the wing's beat direction. this would counteract the net rotation, wouldn't it?

    ReplyDelete
  4. As for having the wings vary in speed to allow the clap effect to take place, that is ingenious. What it probably also means is that the variations in speed lead to variations in lift, unless these can be compensated by varying the angle of the attack of the wing. If any variations in lift remain, the question arises whether the gain of having a clap effect outweighs the reduced lift of a lower wing speed. All in all, it sound like exactly the kind of messy optimization problem that biological evolution comes up with, and is good at solving.

    You can indeed solve the rotation problem by counteracting control surfaces. I do not know whether your designs allow two sets of wings, if if it does, the rotation problem can be solved by having the two sets rotate in different directions.

    Can your designs be seen anywhere on the web?

    ReplyDelete
  5. I'd never thought of adjusting angle of attack to compensate for the low-speed lift loss. That probably means that "clap-flap" could probably work for some of the radial flyers.

    As far as my "rotopters" are concerned, another set of wings could be developed from their mouth parts, as I mentioned, but they wouldn't have the same surface area as the more traditionally developed wings would. Could the size difference be compensated for with increased speed/angle of attack as well?

    And while my radial flyers are not yet ready, I have other stuff up at the following site:

    http://temp.evanblacknet.officelive.com/Nereushome.aspx

    It's all still under construction, so keep that in mind. My work can also be found in the Speculative Evolution forums.

    ReplyDelete
  6. 'Kay I hope I'm not bugging you with all of my comments, but I've put up a rough animation of how I see my "rotopters" working:

    http://temp.evanblacknet.officelive.com/images/RotopterA.gif

    Again, this is just to give a general idea of how the creatures could fly.

    ReplyDelete
  7. The end of that link is /RotopterA.gif

    ReplyDelete
  8. Vey nice! I do not know to which extent the counter rotating nether winglets negate the rotation effect of the much larger upper wings, but it should help. Unless there is something vastly superior in the air in your world, this could work.

    How did you produce the animation, by the way?

    If your rotopters are the direct fictional descendants of my tetropters, then I wouldn't mind posting the animation on my blog (but in that case it would be nice to increase the frame rate of the animation).

    ReplyDelete
  9. There are bilateral flyers on Nereus, but they're relatively new to the scene so they've only put this order of life into sharp decline instead of completely wiping it out yet.

    The model was made in Google SketchUp, frame by frame, then compiled in Adobe Imageready, so I could increase the frame rate, but it would require constructing the new frames by hand and incorporating them into the animation. If I had a better program I'd probably use that, but anything else costs money. Mostly I just do these to give myself a rough idea of how creatures would move, so it hasn't been worth purchasing anything yet.

    But rotopters are an example of convergent evolution to Furahan life, unfortunately, so it doesn't have too much relevance to your blog (which is inspiring and helpful, by the way).

    ReplyDelete
  10. Sketchup! I've played with it but would not have guessed it could be used for things like this.

    I appreciate you making rough animations for yourself just to get an idea; I do that all the time. Still, it's a pity about the frame rate.

    ReplyDelete

Please leave a message if you find any of this of interest.