[MOC] Kamov KA-32 Coaxial helicopter + free instructions

[sandtrooper]

Thank you very much for offering the instructions for building this! Hope to build this in the future.

[Blakbird]

I'll start my review of the functions by reviewing the collective mechanism. In any helicopter, the collective lever "collectively" changes the pitch of all the rotor blades which has the effect of increasing or decreasing lift. In a coaxial helicopter, the pitch of the blades on both the upper and lower rotors must be adjusted together.

The orange lever sits between the seats. The blades have a neutral pitch (zero lift) when the lever is down. As the lever is lifted, it moves the orange mechanism. At the ends of the mechanism are a pair of 12 tooth double bevel gears. These gears lift the entire yellow collective gimbal assembly as shown. The blue parts of the lower rotor head rotate counter clockwise, and the red parts of the upper rotor head rotate clockwise. The dark colored portions (dark red and dark blue) translate up and down with the gimbal assembly while the other portions remain fixed. As the swashplates go up and down but the rotor heads remain in place, the links change the pitch of the blades. You can see the pitch changing from neutral to positive. The torque links between the swashplates and the rotor heads cause them to rotate together.

Here is a closer view of the collective motion without the rotor rotation making it easier to see the blade pitch change.

Feel free to ask any questions about the function. Since I'm sure there will be questions about how I made the animations, suffice it to say that this is the most work I've ever put into such a project. I derived all the equations of motion for each moving part by hand as a function of time, then used POV-Ray to animate one complete revolution. In some later posts, I'll discuss some of the more difficult math involved for anyone odd enough (like me) to be interested.

[Edwin Korstanje]

Great MOC, great technical solutions, great building instructions and now this moving renders.

Can only say wow guys :wub: :thumbup: , what a work and effort.

[Brickthus]

Well done Sheepo and Blakbird. This is good to see.

It must use a lot of the motor power in the friction of the turntables. Could more motors be added to increase the speed? I have used train motors before.

Looks like the blade roots are substantial enough to allow more speed without inducing a blade-off event!

Mark

[Blakbird]

Well done Sheepo and Blakbird. This is good to see.

It must use a lot of the motor power in the friction of the turntables. Could more motors be added to increase the speed? I have used train motors before.

Looks like the blade roots are substantial enough to allow more speed without inducing a blade-off event!

The rechargeable battery allows speed control. At full speed, it is already way too fast for proper stability, so I would not advise adding any more speed.

I found that there is great variation in the turntable friction. The upper swashplate barely worked in the first version I built. Then I changed to a different turntable and now it works great. I tried lubricating them but it didn't help much. The thrust load is low.

[Blakbird]

Next I'll present the power system that drives the rotor rotation.

Both rotors are powered by a single XL motor. The yellow and blue rotor shafts are concentric and counter rotating. The weight of the blue rotor is supported by the lower turntable. Rotational power is passed from the lower turntable to the rotor head via the vertical 1x8 plates shown. Rotational power is passed from the rotor head to the swashplates via the torque links (scissors mechanism). The weight of the yellow rotor is supported by a stop bracket at the lower end around the 16 tooth drive gears. The yellow swashplate is suspended from the rotor head vertically by 3 links and rotational power is passed through torque links.

This closeup of the drive system shows how the counter rotation is accomplished. The orange drive gear powers the green and red axles. The green axle drives the ring gear of the blue differential. This diff passes through the turntable and is anchored to the upper half. The red axle drives the yellow rotor. The reason for using two gears is that the entire yellow axle translates up and down with the rudder pedals. The spacing of the red and yellow gears allows them to remain engaged during translational motion.

An XL motor may seem excessive for doing nothing but rotating a shaft, but there are good reasons for it. Firstly, it has a low speed which eliminates the need for a lot of gear reduction. Most importantly, the turntables have a lot of friction so it requires considerable torque to rotate them. The new turntables would probably work better. Since neither the internal or external turntable gears are used, the new turntables could probably be dropped in with no modifications.

[Blakbird]

The next function is the anti-torque pedals. On most helicopters these would control the pitch of the tail rotor, but on a coaxial helicopter they must control torque on the main rotors.

This animation shows the system. A pair of interlinked pedals (green) on the right and left side of the cockpit work in tandem. As one pedal pushes forward, the other comes back via a set of 8 tooth gears. A crank then drives the orange link which connects to an L-shaped crank at the aft end. A tie rod connects this crank to an axle with a 12 tooth double bevel gear. Movement of the gear lifts the purple rack assembly up and down. There are stops in the structure (not shown) to limit the motion. The down stop also supports the weight. A pair of yellow 16 tooth spur gears inside the purple bracket attach to the rotor shaft and lift the whole thing up and down. This results in the entire upper rotor head translating up and down. Since the upper swashplate is fixed, this changes the pitch of the blades.

This closeup shows the pitch change in more detail. Note that the angle goes both positive and negative. This whole system is nearly, but not quite, like a real coaxial helicopter. The change of the upper rotor blades' pitch would indeed produce a torque differential and therefore result in yaw motion, but it would also result in a major change in lift which would have to be compensated by the collective. On a real coax, the anti-torque pedals simultaneously increase the pitch on one rotor while decreasing it an equal amount on the other. This still results in yaw, but the overall lift is held constant.

The only thing left to animate is the cyclic which will be the most difficult. No one has replied to this thread for a while so I'm not sure anyone is still interested. If not, I won't spend the time on it.

[brunojj1]

The only thing left to animate is the cyclic which will be the most difficult. No one has replied to this thread for a while so I'm not sure anyone is still interested. If not, I won't spend the time on it.

Thank you very much for your work and beautiful animations! It helps to understand more the mechanisms in the real machine and who could explain it better than a aerospace engineer ;-). Very well designed from Sheepo, big respect!

[Cumulonimbus]

No one has replied to this thread for a while so I'm not sure anyone is still interested. If not, I won't spend the time on it.

Thank you for another enlightening episode Blakbird, please keep going, I really appreciate the explanations. I'm familiar with the mechanics of a single rotor helicopter, but I never took time to figure out the working principles of a coaxial helicopter. My respect for Sheepo and his creation(s) only increases.

[jzlego]

The only thing left to animate is the cyclic which will be the most difficult. No one has replied to this thread for a while so I'm not sure anyone is still interested. If not, I won't spend the time on it.

Please do keep going Blakbird. I'm going to show your incredible animations to my brother who doesn't "get" Lego, but is very interested in helicopters. Then we can both have our minds blown together.

As a side-note, I'm not very knowledgeable about mechanics like most here, but it was threads like this that got me very excited about collecting Technic. Buying official sets and learning some concepts is great, but then having access to the parts to build the MOC's from the community here and learn even more complex concepts, that is awesome. It helps me appreciate the value of having a Technic parts collection. It is very cool how the community here is so dedicated to building accurate models, and then showing the functions of the models and teaching others how they work. I'm a huge fan of the work you guys do, and hope to start building the MOC's from guys in this community soon.

Edited September 16, 2015 by jzlego

[Jay Psi]

No one has replied to this thread for a while so I'm not sure anyone is still interested. If not, I won't spend the time on it.

Not true BB, I'm definitely interested. But coaxial helicopters isn't something I (or most others I would guess) know much about, so I can't add anything useful to your excellent explanations.

But if you need any encouragement to continue, here it is

Edit - It would seem my opinion is relatively widely shared...

Edited September 16, 2015 by Jay Psi

[Rishab N]

Agreed. This is fascinating please continue. I just don't think I can add to the discussion

[filsawgood]

Pictures are fascinating! Very cool!

[LEGO Train 12 Volts]

I can't handle the complexity of this mechanism ...but the result looks epic!

[keroro]

Awesome...!!

[Boulderer]

One of the best threads I've seen on EB!

This, for me, is what Technic is all about ... demonstrating and aiding explanation of the functions/workings of real life mechanical systems.

Keep up the great work

[nerdsforprez]

This is outstanding ya'all. thanks for sharing.

[Lighti]

keep it up blakbird. the helicopter with flex system (8856) was the biggest set I had when I was a child, and I loved it. Since I restarted building technic I wanted to build some cool helicopter, but I do lack the skills until now. Ultimate goal would have been one with counter rotating, so I kept saving links to threads with all kind of helicopter MOCs.

Now sheepo comes with full instructions of a fully functional one.

Although I consider the kamov itself "not so pretty", your animations of how that thing works convinced me even more that I have to build that thing (and give it another "body" if my building skills ever improve)

I always show other people cool stuff build out of lego technic. Mostly reactions are "yeah cool, you can do a lot in Lego", but when showing this and then your animations, with differently colored moving parts and everything shown very detailed everyone was "WOW thats really complex stuff"

and as a side note, flying a real helicopter is easier than I thought,

I had the possibility to fly an hour in one made for training with all controls doubled, and an instructor on the other seat

And after 10 mins he placed his hands on his knees and said "you seem to have a feeling for it"

Probably playing countless hours of "Comanche" in my youth helped a little ;)

[TinkerBrick]

I finished building the excellent MOC last night. First many Thanks to Sheepo and Blakbird for creating and sharing instructions for the amazing machine.

If you think about building it - be warned - it is enonomous. Once you add the fuselage and the rotor blades - a normal desk might be to small.

I had to alter mine a bit since i was missing those blue panels. So i replaced blue by black and a bit DBG. I was also missing a LiPo-Battery so i used the standard AAA BB and an IR-Receiver to control the rotor speed. The Kamov weighs quite a lot, the super hard Unimog shocks are struggeling and the main gear bends under the weight.

Here a few pictures of my version:

There are a few more pictures in my BrickSafe.

To sum it up: it is a great MOC. The Instructions are great (and free). The rotor head with its swash plates is ingenious.

Edited January 1, 2016 by TinkerBrick

[Blakbird]

Someone has recently brought this old topic to my attention again, and I've realized that I never finished writing about the animations.  (Also some of the links had since broken which I've now fixed.)  I did all this animation in POV-Ray which has very limited animation abilities.  You need to make all frame changes a function of the variable "clock" and then increment that variable.  You can't rotate around an arbitrary axis, only about the origin.  So each group of moving parts needs to be put into a submodel with the rotation axis at the origin, then the equations of motion for each need to be derived and entered manually into the code.  Definitely the brute force way to do this kind of work.

Here is the code I came up with to do the collective animation:

//helicopter crap
#declare spin=0;
#declare lev=((-cos(clock)+1)/2)*45;
#declare lev2=lev-15; 
#declare levy=40*(cos(lev2*0.01745329252)-0.9659258263);
#declare levz=-40*(sin(lev2*0.01745329252)+0.2588190451);
#declare lev3=lev*2;
#declare lift=-lev3*12*8/360;
#declare bl1=57.29577951*asin(lift/40);
#declare lev4=57.29577951*asin(lift/160);
#declare quad1d=57.775+lift;
#declare quad1e=(1600-400-pow(quad1d,2)-3600)/(-40);
#declare quad1f=-quad1d/20;
#declare quad1a=pow(quad1f,2)+1;
#declare quad1b=2*quad1e*quad1f;
#declare quad1c=pow(quad1e,2)-3600;
#declare quad1y1=(-quad1b-sqrt(pow(quad1b,2)-4*quad1a*quad1c))/(2*quad1a);
#declare quad1y2=quad1d-quad1y1;
#declare alpha8=-(50.09205-57.29577951*acos(quad1y2/40));
#declare alpha9=57.64167-57.29577951*acos(quad1y1/60);
#declare quad2d=48.048-lift;
#declare quad2e=(1600-918.9386-pow(quad2d,2)-1600)/60.628;
#declare quad2f=quad2d/30.314;
#declare quad2a=pow(quad2f,2)+1;
#declare quad2b=2*quad2e*quad2f;
#declare quad2c=pow(quad2e,2)-1600;
#declare quad2y1=(-quad2b+sqrt(pow(quad2b,2)-4*quad2a*quad2c))/(2*quad2a);
#declare quad2y2=quad2d-quad2y1;
#if (quad2d<66.0972)
#declare alpha11=12.5051-57.29577951*acos(quad2y1/40);
#else
#declare alpha11=12.5051+57.29577951*acos(quad2y1/40);
#end
#declare alpha6=-(77.00163-57.29577951*acos(quad2y2/40));
#declare quad3d=60.168+lift;
#declare quad3e=(3600-429.401-pow(quad3d,2)-6400)/(-41.444);
#declare quad3f=-quad3d/20.722;
#declare quad3a=pow(quad3f,2)+1;
#declare quad3b=2*quad3e*quad3f;
#declare quad3c=pow(quad3e,2)-6400;
#declare quad3y1=(-quad3b-sqrt(pow(quad3b,2)-4*quad3a*quad3c))/(2*quad3a);
#declare quad3y2=quad3d-quad3y1;
#declare alpha7=66.72592-57.29577951*acos(quad3y1/80);
#declare alpha10=-(61.57813-57.29577951*acos(quad3y2/60));

Then here are the rotations and translations I applied to each group using the above defined variables:

//lever
rotate <-lev,0,0> 
//link1
translate <0,levy,levz>
//link2
rotate <-lev3,0,0>
//link3
rotate <-lev3,0,0>
//link4
rotate <lev3,0,0>
//link5
rotate <-lev4,0,0>
//link6
rotate <0,0,alpha6>
//link7
rotate <alpha7,0,0>
//link8
rotate <alpha8,0,0>
//link9
rotate <alpha9,0,0>
//link10
rotate <alpha10,0,0>
//link11
rotate <0,0,alpha11>
//collective
translate <0,lift,0> 
//rotor1
translate <0,lift,0> 
#if (spin=1)
rotate <0,-clock*57.29577951,0>
#end 
//rotor2
translate <0,lift,0> 
#if (spin=1)
rotate <0,clock*57.29577951,0>
#end 
//blade1
rotate <bl1,0,0> 
//blade2
rotate <0,0,-bl1> 
//fixed1
#if (spin=1)
rotate <0,-clock*57.29577951,0>
#end 
//fixed2
#if (spin=1)
rotate <0,clock*57.29577951,0>
#end 

To see if actually worked, you need to run the animation with enough frames to see the motion and then go back and try to troubleshoot any errors.  There were a lot of them.  In the end I rendered from 0 to 2π radians in 40 increments to create the final animation.

I never did get around to doing the cyclic animation.  It was going to be several times more complicated than collective and the next shiny project came up and distracted me.  These days, the whole thing could probably be done much more easily with one of the kinematic tools now available.

[opaudo]

Hallo Sheepo,

I can not download the instructions from your site "speepo´s garage", can you help me.

Thankyou very much,

best regards,

Udo

[2GodBDGlory]

3 hours ago, opaudo said:

Hallo Sheepo,

I can not download the instructions from your site "speepo´s garage", can you help me.

Thankyou very much,

best regards,

Udo

I ran into that same problem last year, but if I'm not mistaken, you should be able to find them for free on Rebrickable.

[opaudo]

4 hours ago, 2GodBDGlory said:

I ran into that same problem last year, but if I'm not mistaken, you should be able to find them for free on Rebrickable.

Thankyou, good clue

best regards

Udo