[douglasco]

How do you calculate what shape and size holes to use? My particular application is like an engine hoist, where it will have a load on one end, be supported on the other end and have a cantilever support close to that end.

[keebler303]

You are asking for a nontrivial calculation. An application like this will most probably fail due to shear or buckling. I would not trust either of your attached drawings in a standard engine hoist application. say you are using a rectangular tube, to top and bottom of the tube are carrying most of the load. The sides keep the top an bottom spaced out and transfer shear loads. A good estimate could be had by investigating the shear stress and column buckling loads of the spaces between the holes.

Do you need the weight savings or just for fun? If you need the lightest possible structure, some calculations will be involved. If its just to look cool (and gain a few HP ) Then just err on the side of caution and leave plenty of meat. You can get a pretty good idea of what will work by looking at telescoping boom lifts and the like.

I might be able to help further if you provide some details but there is no simple number or rule of thumb that I'm aware of.

Good Luck
Matt

[Geof]

..... You can get a pretty good idea of what will work by looking at telescoping boom lifts and the like.....

And if you look at examples you will see the holes, whatever shape they are do not extend over more than about 40% of the depth of the beam.

In the DIY world doing this kind of weight saving is maybe ill-advised; unless you are a DIYer with Professional Engineer certification.

[douglasco]

Hey Thanks! Here is a little more info:

Its going to be a telehandler kind of boom with three stages. 7ft collapsed, and whatever i can get out of it extended. For now i'm using standard steel channel for the top and bottom of the beams and the sides are going to be welded sheet . I'm going to try to get away with .120 and .100" sheet on the sides(with the holes cutout). I think this design really puts the material where it needs to be(in the corners). I'm going to try using the lightest channel first which has a web thickness of .18-.20". Might have to go thicker if thats not enough to hold the flanges square under load. The outer stage is 20"x6"(so 6"channel), middle 19ish"x5" and inner 18ish"x4" with UHMW in between. Hoping that I can build it tight enough that when not fully extended the stages can lend strength to each other over the collapsed section. Furthermore around the cutouts i'm going to weld a 1/4-3/8th thick ring(band) to effectively make a flange around the perimeter of the cutout. The flange will be close to .5" tall(just shy of rubbing the adjacent stage). I'm going to try to take the cutouts all the way to the channel(16,15,14" diameters) then the bands can be welded where they touch the channel. Slightly more then 40% . Its really more of a prototype and one-off then for industrial use. But i'm hoping to lift alot of weight with it(lots of testing first). And yes, aesthetics play an important role in the design choice, thus the round big holes. What do you think, any other ways to make it work. I could weld x's out of 1/2" bar stock in between the circles on the backside, but then i'm getting an awful lot of HAZ and i'm not sure that "triangulation" would be in the right place.

If it is successful i'll be rebuilding it out of either 4140 CM or Alum(or something that can be heat-treated).....Only problem is i'm not aware of any chromo or 7075 channel available. But i'll worry about that later:withstupi

[keebler303]

Sounds scary to me.

Homemade equipment lifting heavy things scares me, unless I built it
I would recommend going without holes for the first cut. Then you can hire an engineer once you want to optimize it. You aren't gaining much to have the holes and it makes it sketchy, especially for a homebrew. I would think aesthetics are never as important as safety, and theres no reason to take strength out of it.

If you continue the project on your own, I would proceed with caution and do at least some basic engineering calculations before getting yourself in trouble.
You might start at:
www.engineersedge.com

Good Luck
Matt

[douglasco]

Are you for hire? I can do lots of testing without putting anything or one at risk.

[keebler303]

Possibly, I have an ME degree. PM me if you're interested in me taking a look.

Matt

[RICHARD ZASTROW]

Too many crane collapses in the news of late. I expect to see a few PE's getting sued out of existance.

Personally, I deal with things firmly attached to the ground, like metal cutting machines. No overhead lifting or anything that flies. lol

Dick Z

[razzor7]

Depending on how technical you are willing to get, McDonnel Douglas published a free design guide on grid-stiffened structures. It's got all the equations you'd need to do a first-order approximation of design stresses. I could dig it up if that seems to be what you're after. Grid-stiffened structures tend to benefit from directional materials like carbon fiber and not so much metals. It looks like you're trying to do with metals what many aerospace companies do with composites.

As far as software goes, Solidworks has an optimization module in their 2008 version called OptimizationWorks that does exactly what you describe. I'm not sure whether it's topology- or sensitivity-based analysis, but it basically removes material where it's not needed and keeps material where stress exists. OptimizationWorks is not up to the level of OptiStruct or VisualDOC, but it might be up your alley if you're comfortable with CAD/CAE tools. Looks like it's much easier to use, too. I have not had the chance to use it yet, but we should be getting SW 2008 in September. I intend to give it a shot then.