Frame tubing wall thickness

[sanddrag]

Recently I am finding myself constantly trying to push the limits of robot design and get away with less material in critical areas. So, this is another one of my "how thin is too thin" threads.

Traditionally we have used 1/8" wall 6063 aluminum box tubing for our frame rails. But now I'm wondering if we can get away with .090 wall 6063 (2.5"x2.5") for the main frame side rails.

The way we like to play, the robot is often subjected to heavy contact. So, can the .090 wall 6063 2.5"x2.5" handle it? There would be a 1.5" wide slot centered in each of the four sides. The wheel would stick through the top and bottom slots and pillow blocks would reside/slide in the side slots. Sort of like this. I'm thinking where the slots are, it would not be strong enough. Since the .090 wall thickness would most likely be achieved by surfacing down .125 wall tube, what if we just left it .125 in the slotted areas? Would it be strong enough then? Or would we need to add a doubler plate?

[Veselin Kolev]

If you can actually find a 2.5x2.5 with a .090 wall, awesome

anyway... I think that what you are trying to do is quite possible, and be able to take pretty large hits. The thing you want to avoid is having long beams that are unsupported, where they can easily crush inwards. I think that if you place a small welded aluminum "peg" every 8" or so, just like a square hole in the 2.5x2.5 with another box beam welded in there, maybe a 1x1x1/16 wall, perpendicular to the 2.5x2.5, you would have a light and very strong frame. Kindof like ghetto internal trussing.

The one thing that worries me is that the 2.5x2.5 is really wide. If someone rams you really hard, it could crush upon itself, even if it has a crossbeam right behind it. If you internally brace it somehow, like the ghetto trussing mentioned above, your idea of using really thin walled stuff would work very reliably.

Oh, and I like your idea of the wheels in the beam. Very hot.

[Swampdude]

From our 2004 experience I personally won't go under 1/8" exterior structure (actually 3/16 minimum for me, but I don't use box tube). It's true you could grid a well placed backing that would take the hits. But you'd need at least the amount of material you're saving by dropping the wall thickness. This is just an opinion. I'm not doing any analysis to prove it. But I've watched .090 cave in like a beer can. Then if you open up the walls like in the photo at an unsupported mid span, you're definitely creating a hot spot.
If you have CNC or water jet support, I'd recommend you keep the thicker wall and take truss shapes out of the tube for lightening.
Another thing that bothers me about what you're doing is using this tube for your drive train features that require good alignment. If it starts deflecting the axles or gear teeth (if you integrate the transmission like you did last season) would become misaligned. The last thing you want to do in this game is compromise your drive train.

[Holtzman]

Sanddrag;

The photo which you attached is of the prototype drive train constructed by 1114 in the preseason. The same concept was used for the competition robots of Teams 1114, 1503, and 1680 this year known as the NiagaraFirst Triplets. I've included much better pictures of the competition drive train in the assembly phase.

Based on our experiences using exactly the type of drive train you intend to construct, I would recommend against using a wall thickness less than 1/8(which is what we used). None of the triplets had any major problems with the drive train(the most common being burnt FPs), although in the future, I think we would leave the outside face of the box beam solid, and only lighten the inside faces.

[sanddrag]

I would leave it 1/8" in the areas where the cutouts for wheels and pillow blocks are, but still, is that enough? As for the rest of it, should I leave it solid .090 or .125 with holes/cutouts? I'm thinking solid .090 is better.

Anyway, as for the critical area like where the wheel goes, it is possible I could start with .25 wall tubing and take it down to .1875 wall in that area, and then .090 or .125 on the rest. But that is a lot of material waste.

You are all probably thinking "but you aren't going to get the same outside dimension. Anyway, I know, and I don't really care. All I care about is the rough outside diemnsion of 2.5" sq and making it less than 1 lb per foot.

[Dick Linn]

In 2004, we used 1" x 1" x 1/8 " wall square tubing, which is a lot lighter than that big box stuff. It held up just fine. It was held together using small gusset plates and 10-32 socket head cap screws.