Go Down

Topic: Need help lightening my robot's sturcture (Read 2191 times) previous topic - next topic

ttkoshi

Hello all. So I'm hoping to find some advice or guidance with my current project. I'm building a hexapod that's gone through many iterations, and I'm working on the final one. Previously, the chassis has been made of plastic and vynil, this time I'm cutting the body out of 1/8" aluminum (unless this seems thin).

Attached is an image of the main body. You can see the 6 points where servos will be attached. My question is this: how do i know where I can remove material without weakening the body. I'm just a hobbyist, and don't have any engineering training, so just trying to stick to what looks right. I have Autodesk Inventor if that helps, but I'm a beginner with it. I intend to use 6061 Aluminum as my material. Any info or direction would be immensely appreciated. Thanks!

-TTKoshi

Far-seeker

Ok, some addition information is required if you expect any useful advice. 

First, about how heavy where your previous versions of this robot?  Second, you need some dimensions to go with the image, at least the maximum length and width to give a sense of scale.  Third, you noted where the servos are attached now include the other major components like electronics, sensors, batteries, and anything else with appreciable weight.  Fourth, was the previous design defective in specific areas or are you changing the frame for other reasons?

ttkoshi

Apologies for not adding the info:

The dimensions of the design are 11" by 8.5" (216x280mm approx). The reason for the change in design is that my previous version was laser cut 1/8" acrylic, which ended up cracking on me (also due to a poor design on my part which caused a weak point). I don't have any background in materials engineering (I'm actually returning to school in August to begin studying robotics), so I'm new to most of this. As for the weights, I can provide those when I get home and have a chance to measure the robot parts. Also, I've attached an image of my previous design, with the indicated weak point. Thanks!

PaulS

The weight of the parts is important, but so is the design of the plate that the servos are attached to. If you look at the 2nd picture, you can see that the slot and the sharp notch where the part failed are the problem areas.

Your new design is better, but still needs improvement. At a minimum, there should be no tangent discontinuities in the edge shape. Since the load from the servos needs to transfer to the main portion of the plate, no slots are allowed in the arms.

After cutting the basic shape out, be sure to debur all edges. Tangent discontinuities are where loads concentrate and cracks start.
The art of getting good answers lies in asking good questions.

oric_dan

I'm sure 1/8" aluminum will be many times more robust than acrylic of the
same thickness, but I also suspect 2 sheets of aluminum that size are much
heavier than you might imagine.

There is a serious tradeoff between (a) weight, (b) servo size [torque],
(c) leg length, and (d) batteries [ie, weight vs energy] when building walkers.
I think most people learn to solve these matters by building several different
walkers, and experimenting a lot.

For starters, I might not make the chassis cutout too fancy, but rather
experiment some with items (a) - (d) to get some feel for what's involved
first.

Lynxmotion.com is probably the most successful seller of walking robots, and
you might look at their website to get some ideas. They have a lot of useful
info in regards successful combinations of (a) - (d).


ttkoshi

So pictures should help. Here's my robot in its iterations:

Mackie Mk1 - The original, made with hand-cut poly plastic, ultra light but very flimsy, wobbled like a drunk.


Mackie Mk2 - Adjusted spacing issues, first time using a laser cutting service to produce a body.


Mackie Mk3 - Refined body, tried to reduce weight but ended up weakening the body. This is the current body, which is pretty stable, but delicate.


Another image - I had built a cheap gantry for the robot to keep it stationary while I tried to figure out how to make it walk.


And lastly, the current cad version of my design. I'm learning about keeping fillets instead of sharp corners to relieve stress on the joints, and still trying to decide where to remove more material


So there you go, my baby in progress. And as for the name, Mackie MSK-6S is the name of the first BattleMech in the Battletech universe, which is what began my love affair with robots. Many thanks for all the help!

Go Up