Close

Project log 10: Let's design and calculate a little.

A project log for DIY Mech/Exoskeleton suit.

Mechs are not viable, nor cheap, so I will try to design and build one alone anyway.

fulanodetailFulanoDetail 12/05/2022 at 11:067 Comments
Monday 07:58, 05/12/2022

So, now that I have a 3D sketch/concept, I do want to figure out how everything will need to fit around the fricking mech's body.

Making its body is easy, fitting a human, a power supply, a board computer to control everything and the pressure chambers will be really hard.

I tried to make a really sketchy looking test yesterday on Blender and I only got to fit everything when I scaled it up to around 3 meters, same height as Obsolete's mech (on the wiki it is said they are around 2,55 meters tall).

I also calculated a little bit how much force I would need in the arms in order to lift 1 ton, and basically, I would need to exert a force of 5 tons. So, basically, I'm already needing to multiply the supply costs by 5...

I used this and this mechanical lever calculators, I put 5 cm for the effort arm and 30 cm for the reistant arm on type 3 mechanical lever, obviously, I don't know yet what is the actual size of the final thing, but since it is 2 meters tall, its forearm would easily exceed 30 cm of length.

But if it was the case, it would increase the muscles costs to 400 Reais already... Not as crazy as 6 thousand bucks as the conventional muscles, but expensive nonetheless.

Well, I didn't want to make a exoskeleton, but seeing the costs only increasing and increasing, I may be forced to change the focus of this project to exoskeletons.

After all, as much as I like Mechs, I don't want to just keep going with a project that I think it won't be worth the trouble.

Yes, I know I said that on the project's subtitle: "Mechs aren't viable, nor cheap, so I will try to build one anyway".

But let's just remember that the muscles themselves are the cheap part, not the metal endoskeleton, not the motion capture suit, not the power supply, not the hydraulic pump.

If the suit can't walk without a power supply the size of a house, then this project is useless, don't you think?


And to further the discussion a little more, let's talk a little bit about what I nicknamed (I've never seen this term being used) "human impedance":

The idea of a exoskeleton would be to carry the extra weight the user would need to carry, so their movements and body effort aren't as exhausting.

But if you want an exoskeleton with super human strength, then the exoskeleton doesn't need a human user.
The human user will just get in the way of the exoskeleton because of its squishy meaty fragile body, the Guardian XO, an exoskeleton built to help its user lift 200 pounds (+/- 100 Kg) is this heavy, cluncky piece of equipment that has a lot built into it to just not harm its user.

If you wanted a exoskeleton to give you super human strength and speed and maneuvability, the problems would only attenuate. 

At this point, isn't easier and simpler just take the human out of the "equation"? That's why I say "human impedance".

The more complex it gets, the more you want humans out of the suit and in a safe place, like a cockpit. And worse, instead of the human body sutaining its own weight, now it is almost dead weight on the machine.
But since you're using a exosuit with super human strength and speed, the human strength would be minimal and the suit would need to carry it anyway.

That's why I decided to go with mech, not exoskeleton. But hey, it is my own fault for wanting this thing to lift 1 ton. lol

Edit:

If I wasn't clear enough, let's think this way:

You're not an Olympic athlete, you are sedentary, you never exercise.
And suddenly the spirit of a record braking athlete possesses your body and repeats the same feats they would have done while alive in their well trained body, but with your sedentary body.

What do you think it would happen to your muscles? Your joints? Your bones? Everything would have been destroyed.

Now just replace the part saying "the spirit of an athlete" with "super human exoskeleton". You would have the same results.


Plus, there is another problem that nobody seems to think about exoskeletons: what I call "body on body impact".

So, you're completely covered in metal tubes with sharp edges and no padding at all, you're running and then you stumble and fall.

How much it would hurt?

It would be the same thing as carrying a lot of pieces of metal on your arms and then falling on top of them.

It is the same thing as belly flop on a pool of Lego's, but the Lego's are around your entire body falling with you.


Well, first, let's start downloading a low poly rigged human, so I can actually try to make them fit into the mech.

I found this rigged 3D model for free on Free3D.

I will make them 2 meters tall, because if a 2 meters tall human can fit into a cockpit, anyone can. lol

By the way, you just need to go to "Pose Mode" on blender in order to move these sticks/armatures just like you would do with an object (pressing "G"), for some reason there is a myriad of different ways of doing that.

I will try to make something like a "piggy-back cockpit" on the mech, just like in Obsolete.

You see, that's why I'm stoopid: it just crossed my mind that I should had put the pilot around the mech and then modelled the cockpit instead of doing it on the air...

And obviously, it doesn't fit.

The mech has more or less 2 meters (6 feet) of height, and the human inside has 2 meters of height too, so you can have a perspective on how stupidly big the cockpit is compared to the mech itself.

It is like a giant goofy backpack.

Without the cockpit.

CockpitOkay, this looks a little less goofy, still sucks anyway.

Well, I unfortunately had to stay with 2,55 meters tall, or else it just gets goofy looking and akward.

Well, the "only" problem is that now the forearm is 55cm long instead of the "worse" scenario I wrote at the beginning of this Project Log. Of course, I can always shorten them a little bit (I think).

Accordingly to the second lever calculator, if I had a 1 ton (9806 Newtons) load at the tip of a 55cm long forearm, and I had a bundle of muscles at 10cm distance from the fulcrum point, making it a type 2 lever with a load arm of 45cm, I would "merely" need to apply 4499kg (44127 Newtons) of force to reach equilibrium at 90º.

Yes, equilibrium is not necessarily the entire force necessary to lift this weight, if the forearm never reached the 90º degree angle, it would still be able to carry it, not necessarily lift it.

I know that probably for smarter people than me this would seem obvious, but not for me, lololol.

Also, that's why you can't lift a heavy object (like a 20kg package) at the tip of your arms, but you can easily carry it when you support this weight at your shoulder or at your belly.

Another thing I didn't stop to think about it, but I will have 10 or more McKibben muscles on every place that has a "balloon" on the 3D model, will everything fit in there without reach that "too muscular" problem?

Example:




Also, I totally forgot to talk about: the power consumption, this post was meant to talk about the amount of hydraulic fluid, pressure and flow this contraption would required, but I totally forgot, lol


Also² I forgot to talk about the "walking chair" concept, where the cockpit would be the entire torso. I think the idea is cool, but I believe the cockpit would restrain the walking movement and it wouldn't be able to even climb some stairs.

The probable solution would be to either space the legs out like in the Obsolete-like mech or invert the cockpit, but then it would be too reliant on cameras and/or mirrors.

Other solution would be to use chicken-like legs (that I heard are more efficient), but I don't know how intuitive would be to control a chicken mech with your body movement.

Not very comfy.

I tested on vertical too.


Edit:

Normally I would create a new Project Log, since I'm adding this edit 2 days later, but I don't think it will matter much in this case.

I tried other designs, like the walking chair, and I didn't like the results that much...

This does look like a Gurren from the anime Gurren Lagann (as shown bellow) and it does look cool as heck.

The reason I didn't like it is because the legs are too far apart from the center o gravity (the middle of its body), maybe it won't be a big of a deal with a small mech like this, but if you're counting on carrying something around 1 ton, it will definitely affect its balance.

In this attempt I tried exactly to solve that, but the arms are too far apart. I don't even think these would be able to reach the front (where the knees would be).


Edit²:

Tried a new design, I kinda liked this one (I didn't fix the arm distance, I just made this as fast as possible).


Edit³:

I've made these new designs, following the advice of dekutree64 on the comments (I think it was them, I don't remember quite well).

Where the torso should be solid to reduce moving parts and it could house the electronics, power supply etc.

I kinda like how it looks, but like they said, it still needs a little bit of mobility on the torso (also, ignore the arm clipping through the cockpit, and the too protuberant shoulder platforms).

The first from left to right is the 2,55 meters tall version, the second one is the 1,98 meters tall and the third is using the first cockpit design I've mad (that is terrible).


Edit

Two on the right are the Dekutree idea: 150mm diameter tubes as torso/support for the arms.
On the left is the same thing, but with a container in front of it, it would contain electronics and the hydraulic pump and whatsoever.

Edit ⁵:

I tried this bucket looking design, lol.
And it is the 2 meters tall mech, by the way, not the 2,55 meters tall.

(and yes, I'm trying to find how to calculate a McKibben muscle, ok? It is just hard y-y)

Discussions

FulanoDetail wrote 12/07/2022 at 22:32 point

@dekutree64 I see, that's a really good idea!
Thanks for the advice, I will sure try it later :)

  Are you sure? yes | no

kelvinA wrote 12/07/2022 at 19:31 point

"Also Squared". I've never heard that phrase but it just rolls off the tounge much better than "also also" and looks cleaner and shorter in text than "additionally".

  Are you sure? yes | no

FulanoDetail wrote 12/07/2022 at 19:51 point

Hehe, thanks.
And now that you said it, I think I'm saying "also" too many times :)

  Are you sure? yes | no

dekutree64 wrote 12/06/2022 at 23:00 point

lol, that piggy back looks hilarious. I think the best bet will be a walking chair with arms. I think the anime Gurren Lagan has one like that, although I've never actually watched it. The Obsolete style looks good too.

As for falling, you'll obviously want to design it so there are no sharp edges that you're likely to bump into, and probably something akin to a roll cage like that Obsolete mech so the pilot can't be crushed under the weight of the machine. But it will still hurt.

An alternative would be to use 4 or more spider style legs, allowing the cockpit to be closer to the ground (less distance to fall), and largely eliminating the possibility of falling in the first place.

Something to keep in mind is that there are millions of one-armed, tank treaded mechs already in service, aka excavators. For full size, you might be able to repurpose the hydraulic system from one.

  Are you sure? yes | no

FulanoDetail wrote 12/07/2022 at 01:01 point

About the cockpit, I was thinking of doing that, and fill it with impact absorption padding (like those neat "non-newtonian fluids" you can find on youtube.

About the hydraulic system, I don't know very well yet if these will be good for this porpuse, because deal if really high pressures and consume a lot of energy (and are heavy).

  Are you sure? yes | no

dekutree64 wrote 12/07/2022 at 20:26 point

Nice work on the mockups! The last one (narrow spaced legs, wide spaced shoulders) is what I had in mind, but seeing it this way, the human legs obstruct the mech legs unless you put them far enough forward to obstruct the mech arms... Smart design by the Obsolete artists, having the human legs stick forward to keep them out of the way (they do still obstruct a lot of the arm range, but I think you could get the mech hands together either above or below the human legs).

But being in such close contact with the upper torso, I think you'd want to make it fully rigid for safety. Just one spine twist actuator between the cockpit and pelvis. That would also allow the torso to be as narrow as possible (just a tube) so the human legs don't have to be spread so wide to reach around it. That may make it possible to get the mech legs into a full squat (knees above the hips) without hitting the human legs, so you can get in and out without a ladder.

  Are you sure? yes | no

dekutree64 wrote 12/10/2022 at 03:19 point

Shrink that cylinder to 150mm diameter and see how it looks. That's what I had in mind for the Obsolete style. Not a container for anything, just the minimum structure to support the arms.

There's not much room for electronics elsewhere though... just that sort of head-like lump that the arms attach to. I suppose you could make that a bit bigger without causing any problems. You could put some stuff on the underside of the seat too, but it would be better to have anything heavy be toward the front to counterbalance the weight of the pilot.

P.S. I am male.

  Are you sure? yes | no