How would you make a mechanical tail?

[Carsten Svendsen]

So, I was thinking about making a dolphin in technic (I've got no experience in "real" LEGO what-so-ever, and it's got no functionality either), and my first concern was how I should make the tail.

For those of you who don't know how a dolphin swims check this youtube video

If I have to build one, I want to make it function like the real one. I was thinking about using two motors for the tail, one motor for the jaws, and two motors for the head. There's no problem in fitting it all, it just depends on the size.

But the tail has left me wondering; I want it to go up and down like the real one, which requires a lot of links. If I have to use one motors for up/down and one motor for left/right, then how would you combine all those links to work fluidly? Is it even possible?

[dhc6twinotter]

Here is a video of the Dinosaur in the LEGO 8484 Control Center II. You may be able to replicate the tail function in your dolphin, although, if I remember correctly, a dolphin's tail moves more up/down instead of side/side. You could just flip the hing points to get that up/down movement. The tail is shown at the start, and again at 3:22.

http://www.youtube.com/watch?v=AytevjfrgkM

The only problem with that setup is that is requires you to have the long flex cables. If you don't have flex cables, I'm sure some other types of mechanical linkage would work as well.

Edited December 6, 2012 by dhc6twinotter

[Alasdair Ryan]

Yeb at Daniel says above flex axles are the best,if you don't have any of those try using axles with 1/2 toggle joints.

[Carsten Svendsen]

The problem is that a dolphins (or any other whales) tail moves from the body, and the fluke will follow later - Just like if you were to pick up and swing an electrician plastic pipe.

With the dinosaur solution - I've also seen a heavily modified LEGO version - and they both consist of motion starting at the tip instead of the base.

Edited December 6, 2012 by Carsten Svendsen

[Alasdair Ryan]

The problem is that a dolphins (or any other whales) tail moves from the body, and the fluke will follow later - Just like if you were to pick up and swing an electrician plastic tube.

With the dinosaur solution (I've also seen a heavily modified LEGO version) and they both consist of motion starting at the tip instead of the base.

Ok in that case what about having the flex axles or the substitute connected to a series of cams driven from a motor,which starts slow but gets faster at the end which would create a wave affect.

[Carsten Svendsen]

I'm not quite sure I understand that idea. Cams that are linked with flex axles I understand, but what with the motor driving it? You want it to increase/decrease in speed constantly? I don't think that's necessary to make the desired effect.

I only have 4 flex cables from a couple of Cybermaster sets, so it's not like I can recreate the theory. I can't use string or rubber bands either as it needs to be flexible sideways and stiff "lenghtways"

[hrontos]

Try to use pneumatic tubes or similar pipes to get desired effect if you do not have enough flex parts. It is probably not necessary to use only flex cables, mechanical axle similar to the one used in piston engine should also generate the wave effect. It can be stronger at the end of the tail to make it swing more.

[Alasdair Ryan]

I'm not quite sure I understand that idea. Cams that are linked with flex axles I understand, but what with the motor driving it? You want it to increase/decrease in speed constantly? I don't think that's necessary to make the desired effect.

I only have 4 flex cables from a couple of Cybermaster sets, so it's not like I can recreate the theory. I can't use string or rubber bands either as it needs to be flexible sideways and stiff "lenghtways"

Not quite,You would need two cams linked together but geared differently.

[Hopey]

Just thinking out loud here...

A not-quite-good-enough setup: Start with a series of hinged parts, with a channel above and below through which a flex cable / pneumatic tube / piece of string can run. Here's a very poorly-designed example to show what I mean (obviously one could do a lot better):

The yellow axles are the hinges, the cable would run through the grey parts. You'd need to put another set on the other side of course, to make it bend the other way.

You'd run a cable through each side and fix it to the tail end. At the body end, you'd have two 180 degree offset cams which pulled each cable in turn.

This would not, of course, have the desired effect; it'd be a simple flexing motion. How to get the wavelike motion?

One option is to cross the cables over to the other side at the midpoint. With this, pulling one cable would produce an S shape, then cycling to the other cable would produce the reverse S, which might look cool, but there's be no directionality to it; the wave wouldn't propagate backwards.

Another alternative is to run a second set of cables that only extend halfway along the tail. Pull these with another set of cams that are 90 degrees ahead of the other two. Maybe construct it so that the full-length cables don't affect the first half, although I'm not sure this is necessary.

If the scribblings I've just done on an envelope are correct, then this should produce a wavelike motion. You might have to play with the synchronisation between the two, and perhap experiment with stepping so that only one set of cams is moving at a time, but I think that'd result in a less fluid motion.

You could theoretically have as many such sections as you like, I guess.

[jorgeopesi]

In my opinion, you're taking all the fun. We try to help in whatever but the most fun is to take a decision and develop it, if it does not work you can ask for advice or try another method but if you start asking, we do participate which is good, but you take yourself development and learning ability. My advice is make a Dolphin as you can, questions to improve it, build more technic MOCs and build another dolphin, the fourth dolphin you build will be perfect . A tail for a dolphin is an awesome project I don´t like animal MOCs yet but I envy it because the mechanic part is a very attractive challenge.

[aeh5040]

I would like to see a technic dolphin!

You might take a look at mahj's Da Vinci flyer - the wings achieve an amazingly fluid-looking motion with only one internal joint:

[jorgeopesi]

Awesome with Shadow of the colossus music I think... Mahj always surprise me .

[SNIPE]

Hi,

You have a vertical pulley and a horizontal pulley and have string on them then have motors in the lower body also with pulleys on. You use the motors to change the X and Y axis of the tail or have a set program for the motors this does it by reeling in string which pulls the flexible tail linkage.

Be sure to use the clutch gears so you don't break anything apart say if the tail flexes too much because of the motor being left on.

The hard part is getting the actual tail to flex but at least you can have the string go through the pin and pulley holes / cross-holes.

Good luck

Edited December 7, 2012 by SNIPE

[Carsten Svendsen]

Another alternative is to run a second set of cables that only extend halfway along the tail. Pull these with another set of cams that are 90 degrees ahead of the other two. Maybe construct it so that the full-length cables don't affect the first half, although I'm not sure this is necessary.

If the scribblings I've just done on an envelope are correct, then this should produce a wavelike motion. You might have to play with the synchronisation between the two, and perhap experiment with stepping so that only one set of cams is moving at a time, but I think that'd result in a less fluid motion.

You could theoretically have as many such sections as you like, I guess.

What you describes there seems to fit with the video of the da vinci flyer mechanism. This might just work with enough links! I just have to experiment a little first.

Hi,

You have a vertical pulley and a horizontal pulley and have string on them then have motors in the lower body also with pulleys on. You use the motors to change the X and Y axis of the tail or have a set program for the motors this does it by reeling in string which pulls the flexible tail linkage.

Be sure to use the clutch gears so you don't break anything apart say if the tail flexes too much because of the motor being left on.

The hard part is getting the actual tail to flex but at least you can have the string go through the pin and pulley holes / cross-holes.

Good luck

I already mentioned above in my 2nd post that this will not be correct. It might work, but it's the wrong end that moves first.

Edited December 7, 2012 by Carsten Svendsen

[SNIPE]

It can be stiff sideways just don't use a horizontal pulley and only use connectors that let the body rotate vertically.

Edited December 7, 2012 by SNIPE

[F0NIX]

Carsten Svendsen, what size are you aiming for in the dolphin?

I'm thinking out an idea for an mechanism in my head right now, but it will be a bit bulky and requires a lot of gears an liftarms...

[Carsten Svendsen]

It doesn't matter what size it is, I was figuring perhaps 0.5 - 1 meter in lenght.

The important part is that it is as functional as possible

Edited December 7, 2012 by Carsten Svendsen

[Carsten Svendsen]

This is what I imagined it should be like:

As you can see, the motion is going up and down in synchronization, and it's straight in the middle at all times.

Now, what I don't understand is that the "tail" seems to bend more downwards than upwards. I have no idea why, but I'd like the motion to be identical in both up and down movements.

At the end of the "tail", there should be made a mechanism like in the "Da vinci Flyer" posted above, to simulate the fluke.

The next step would be to make this motion possible both like above, and sideways. This could be done with flex axles instead of liftarms, but the possibly difficult part would be to assemble it all in one giant gearbox.

[hrontos]

Now, what I don't understand is that the "tail" seems to bend more downwards than upwards. I have no idea why, but I'd like the motion to be identical in both up and down movements.

Nice smooth movement.

More down than up can be caused by the play of the gears in the combination with the gravity or it is a slightly not precise alignment of the gears. When tail is straight, each gear has to be in the topmost or bottommost position. It's hard to judge from the video. When you will make it a real doplhin with this movement, people will be so impressed they will hardly notice that slight difference. If it is caused by design of the solution, make it go more up than down, since gravity will compensate it on a real model.

[allanp]

looking good. To get an even more life like movement, try having the cranks out of phase, so that the middle and front sections moves before the end section.

[Carsten Svendsen]

Nice smooth movement.

More down than up can be caused by the play of the gears in the combination with the gravity or it is a slightly not precise alignment of the gears. When tail is straight, each gear has to be in the topmost or bottommost position. It's hard to judge from the video. When you will make it a real doplhin with this movement, people will be so impressed they will hardly notice that slight difference. If it is caused by design of the solution, make it go more up than down, since gravity will compensate it on a real model.

I don't think gravity is the case here, more likely play in the gears as you say. Now that I think about it though, a dolphins tail does go more downwards than upwards.

looking good. To get an even more life like movement, try having the cranks out of phase, so that the middle and front sections moves before the end section.

I just tried that, and you're right, it does look more lifelike.

However the tail won't be straight in the middle, but that could probably be solved by the weight of the tail whenever it's done.

[F0NIX]

The motion you are after is like a "wave motion" that starts at the root and spreads out to the tip of the tail...

Tried a few things here. I found one method but it would not work on vertical motions, only on horizontal motions like on a fish. Because I used clutch-gears that spread the motion from the root and out to the tip with one clutch gear on each joint, and the joint has restricted movement.

I will try to make a short movie of it later...

[robuko]

Carsten, I think the motion you are aiming for was achieved very nicely in the neck of the Giant Zrrk zoid, using a system of cams. Worth a look.

[Carsten Svendsen]

A search turned up no results. Would it be too much to ask for you to post a link?

[robuko]

Here we have the up and down positions of the neck. The head stays parallel throughout the motion. The whole neck pivots around a vertical spindle at the base. Hope this might be some use

Edited December 11, 2012 by robuko