Sheet Metal Design

[baronep]

So, our team in the off season built a project that used custom laser-cut aluminum parts made out of 1/8" sheet. This thickness was good for our design in which we required extra-rigidity and we had no weight requirements. My question is how does someone optimize weight and rigidity. How do you know in a CAD model if your sheetmetal part is going to fail or not (we use Solidworks) and what thickness of sheetmetal is typical on a 148 style robot?

Thanks

PB

[thefro526]

Originally Posted by baronep

So, our team in the off season built a project that used custom laser-cut aluminum parts made out of 1/8" sheet. This thickness was good for our design in which we required extra-rigidity and we had no weight requirements. My question is how does someone optimize weight and rigidity. How do you know in a CAD model if your sheetmetal part is going to fail or not (we use Solidworks) and what thickness of sheetmetal is typical on a 148 style robot?

Thanks

PB

JVN has posted CAD of 148's robots from 2008 to 2011 in CD Media. A quick search will link you to them.

[Akash Rastogi]

Quick search of others asking the same question and a search in CD media will also get you results for your questions.

Here's one that answered a lot of my questions when I started learning about sheetmetal: http://www.chiefdelphi.com/media/photos/35696

[JohnSchneider]

If you remember our robot from last year, we actually used several thicknesses... .125, .060, and I *think* we actually used .250 on something...though it was overkill...Different sizes for different applications!

[roystur44]

Team971 uses .090" 5052 H32 in their designs. Last year we switched from .125" aluminum because we can save up to 27% in weight of our chassis. Also the laser processing time is much faster using a thinner gauge material. 95% of the robot parts are designed from one thickness. The team builds a competition and practice robot so we nest the parts on a sheet to utilize as much material as possible and to decrease our process time on the laser to cut down on costs. Our fab sponsor is happy about that.

It's been an evolution since 2006 and the .250 thk chassis but the team's CAD group is getting real good on how they design the folds, joggles, rivet holes and lighting patterns to maximize strength and chassis stiffness. They are confident so we are thinking of going with .060 5052 this year.

Figuring out if the parts are stiff enough and light comes from experience and iteration of the chassis design. I always suggest to go for lightweight design and make extras then at the end of a competition season do a review and save the info in your robot toolbox for next time.

[PhantomPhyxer]

I am happy to see that most Teams are using Aluminum. It is light an strong when used properly. I would suggest that if you are useing Solidworks. You should go the next step an have one of your mentors teach you Stress Analysis using your computer model. Finite Element Amalysis does not optimize a part but you can iterate on a solution by just changing the thickness of the material and re-running the problem.

When designing think of an "I" Beam. The top and bottom web are very critical in bending. The center web between the top and bottom flanges carry the shear. So, if you keep the flange the same thickness but seperate it by a larger distance then the the moment of inertia is increased and the bending strength increased.

If you ask a more specific question I will try to answer it.

You might also try in using a Ballistic grade of Aluminum like 5083 that is work hardened.

Norman Collier

[AustinSchuh]

Paper is a great way to teach sheet metal, since it is cheap and easy. Design something using 5 thou thick material in Solidworks, make a drawing, print it out, cut it out, and then put it together. Last year we had a sheet metal paper bridge competition and had students CAD up a bridge, build it, and then we had a competition.

[DonRotolo]

Originally Posted by baronep

how does someone optimize weight and rigidity.

Bend the metal for strength and rigidity, put holes in where no strength or rigidity is needed. Use sheet metal in tension where possible.

Easily said. Practice 'optimizing' a sheet of paper, you'll get the idea pretty fast. I can 'fold' a sheet of printer paper so it'll penetrate your skin.

[sanddrag]

Originally Posted by DonRotolo

Easily said. Practice 'optimizing' a sheet of paper, you'll get the idea pretty fast. I can 'fold' a sheet of printer paper so it'll penetrate your skin.

I have made a sheet of paper free-stand 74 inches tall using nothing else more than about 10cm of scotch tape.

For sheet metal design, it REALLY helps to have someone from industry who can help you with specific factors such as different bend parameters, and sizing to fit their exact equipment and dies, etc.

[thefro526]

Originally Posted by sanddrag

For sheet metal design, it REALLY helps to have someone from industry who can help you with specific factors such as different bend parameters, and sizing to fit their exact equipment and dies, etc.

Seconded.

I'd suggest taking a trip to your sheet sponsor at some point and talking with them about their standard K-Factor, Bend Radii, Punch Tools, Preferred Materials, Etc. I have a bunch of information about our sponsor written down so that I can reference it later and I also set my Inventor up so that the default sheet metal parameters match the shops. It saves a lot of headache.

[Jeff Waegelin]

Originally Posted by baronep

So, our team in the off season built a project that used custom laser-cut aluminum parts made out of 1/8" sheet. This thickness was good for our design in which we required extra-rigidity and we had no weight requirements. My question is how does someone optimize weight and rigidity. How do you know in a CAD model if your sheetmetal part is going to fail or not (we use Solidworks) and what thickness of sheetmetal is typical on a 148 style robot?

Thanks

PB

I can answer specifically what is on a 148 robot. All of our sheetmetal is either .125", .090", or .063" 5052 aluminum. On some parts like gears or sprockets, we will use .125" 6061 aluminum. All of these were chosen because it's what our sponsor carries in stock, and these three thicknesses provide a pretty good range of options for FRC robots.

For the most part, how we choose which thickness to use depends on where the part is. Chassis parts are usually .125", as are any other main frame pieces. Secondary frame parts and arms and such usually get .090". End-effectors and anything that experiences minimal load or contact gets .063". Basically, whatever we think is the thinnest we can get away with on a part. If you design your parts smartly, with flanges and support bracing in the right places, you can get away with thinner material just about everywhere.

[joek]

use a lot of ribbing folds, and try to get a mentor who works with sheet metal in industry. 12 ga. is a pretty good thickness for most applications, but thinner stuff works if it's designed right

[mplanchard]

Find a sheet metal vendor for some advice. You also have price to worry about too.

You are in the Dallas area? Go to swugn.net and see when the next local SolidWorks local user group meeting is. You may also get some advice.

When Investigating analysis go to tools, addins and turn on SolidWorks Simulation. You will then see simulation tutorials under Help.

You want to use shell elements. These will provide good results. Also sketch profiles to indicate large concentrations of mass, such as a motor. Then use Insert curve, split curve. Now the area bounded by the sketch can accept a distributed load. Marie

[Curt Henderson]

This topic is right up our teams alley. Check out www.huntingtonrobotics.org for our photos on our past robots.

We build our bots from .040" prepainted aluminum sheet using a technique called Monocoque. We will be releasing a video (most likely after the build season) showing our methods of construction so everyone can gain from our experience. Our robots are traditionally light weight and very durable as seen in videos of past regionals. We form the material to provide the outer shape and then use internal bulkheads to support the outer shell to provide the necessary strength of the members. We use the thin material so we can bend it with hand tools and it is alot less expensive than than the thicker materials and it still provides plenty of strength for what we need to do.

Feel free to send any questions our way, we will try to answer all questions that you may have.

Curt

[tim-tim]

I know what we do is go off of previous years experiences (both our own and other teams) and base the design from those lessons. I can tell you that many teams do not go through the FE Analysis previously mentioned because it is complex and most schools can't support it.

However, JVN will tell you it is about iterative design. If you have the luxury, do some testing. Find out where your design fails and fix it, then go back to the drawing board and make it better. Then go back and find where you can remove material or place a pocket or hole pattern to reduce weight in the design.