Advice - First Sheet Metal Frame

[Luckryan]

Our team has always used the kit chassis so this is our first basic design for a sheet metal frame. We don't have anyone on the team who has designed one of these before, so we've just looked at other teams' designs for ideas. I'm sure there are lots of things we could get better, though. Are there any problems or fundamental flaws? Any advice would be fantastic.

.090" Aluminum 6061
32 x 23.5
There are no holes to mount the drive train yet but these will come later.

[Chris is me]

How is your sheet metal being manufactured? Punched, lasered, waterjetted? This determines a lot about whether or not your design makes sense for your resources.

Also, 6061 does not bend too easily - 5052 is a better material for bent parts.

[Luckryan]

We're in contact with a machine shop who say they will laser it for us. And we could make it out of either kind of aluminum - thanks for the tip.

[msimon785]

Originally Posted by Luckryan

Our team has always used the kit chassis so this is our first basic design for a sheet metal frame. We don't have anyone on the team who has designed one of these before, so we've just looked at other teams' designs for ideas. I'm sure there are lots of things we could get better, though. Are there any problems or fundamental flaws? Any advice would be fantastic.

.090" Aluminum 6061
32 x 23.5
There are no holes to mount the drive train yet but these will come later.

I see no glaring problems with the frame structurally. Of course you haven't implemented holes for shafts or anything yet, so comments are tentative. Your pocketing seems somewhat arbitrary, which is okay. However, depending on how you are incorporating it into an overall robot design, a simplified lightening pattern should suffice.

As Chris said, 6061 does not bend. We made this mistake last year and had to remake 100% of our sheet metal parts. It is a comparatively brittle alloy and will crack far sooner than bend. 5052 on the other hand bends quite easily and serves sheet metal applications far better.

Make sure to CAD and model the entire rolling chassis. Explode your assembly and make sure your fabricator can make all the bends. You will not be able to bend 6016 T6 use the 5052. We typically use .080 for most of the parts and carefully use some .125 for strength. Keep your total number of individual parts as few as possible. All of your fastener locations and fastener choices should be pre engineered. Don't go crazy and cheese out to much, this is your foundation, but keep close track of weight. Understand that fabricating dimensional tolerances are not as tight as machining tolerances, you won't be able to bend to +- .010. It is important that your assembly is square and stays that way. Planning and drawing are important, much can go bad very fast. We have built several fabricated chassis, each one better than the last.

[apalrd]

6061 T6 DOES bend, just not easily at 0.090" thickness.

Our entire chassis last year was folded, riveted 6061 T6 at 0.063" thick, bent on our 36" manual brake and a nearby 54" manual brake.

We made a few gearbox plates out of 0.090 6061. They're hard to bend, would not recommend.

It's all about strength to weight and ease of manufacturing.

Also, when you get down to really thin stuff, we generally stop getting crazy with the pocketing to save machine time and because it really saves hardly anything. We engineer our pocketing of 0.250" plate a lot more than 0.050" folded assemblies. A 2" square of 0.050" sheet weights ~8.5 grams, which is almost nothing.

[tim-tim]

Yes, 5052 is TYPICALLY easier to bend than 6061, however, this depends on the state of the material. I work with sheet metal on a daily basis. Our primary alloy of choice is 6061-T6. And yes ~75% of our parts are bent.

This being said, 6061 does require a larger bend radius. What size bend radius are you using in your model? To bend .090, you should bend with a .188 radius. This will just about garuantee that there will be no cracking in the parts.

I honestly recommend getting with your manufacturer and see what they recommend.

S/N: I can bend 6061-O at a much tighter radius, but the strength is reduced.

[tim-tim]

One other thing to double check is your "k" factor for your bends in your modeling software. The aluminum actually stretches as you bend it. This will affect your flat pattern and tolerances of the formed (bent) pieces.

A quick way to double check is through simple bend deduction or bend allowance calculations. Again, check with your manufacturer and make sure everything is good.

[Nuttyman54]

It's a great start! I assume you probably have some plans for a belly pan for electronics and the like, but you should make sure it's at least somewhat structural. The flanges on the front and back rails are good, but they won't hold up to abuse * without some stiffener in the form of a structural belly pan, or additional center supports. Bumpers will help some, but you don't want to rely on those.

*These pictures are of the damage to the reinforced repair of a 2x2x0.125" aluminum angle, since the original 1/8" sheet chassis couldn't hold up.

[Akash Rastogi]

Originally Posted by tim-tim

One other thing to double check is your "k" factor for your bends in your modeling software. The aluminum actually stretches as you bend it. This will affect your flat pattern and tolerances of the formed (bent) pieces.

A quick way to double check is through simple bend deduction or bend allowance calculations. Again, check with your manufacturer and make sure everything is good.

Going to quote this for emphasis.

Talk to your shop about what material you'll be using, what they want you to use, and anything else they need from you in your CAD.

[trilogy2826]

Though it may sound excessive, our CNC bend operator prefers the bend annotation dimension on the print down to 3 places past 0 (ie: 1.203").

My personal preference, especially when running structural components with sheet is to radius every single corner. Not only does it look better, but according to multiple FEA analyses on our previous frames can decrease stress concentrations and facilitate a long term stable solution.