1/8th aluminum Y plate

Thanks, Im less worried about strength than:
“is there something in the design that requires that the Y plate is thicker.”

Based on Modulus of Elasticity and Bending Moments, I think that 1/8th 6061 should be equal or better as far as stiffness than 1/2 in plywood or hardwood this case. Did a quick part stress analysis and that also looks to be same or better.

Now the hardest part, deciding what colour to print parts in…

I don’t think there is anything in the design that requires thicker plates. All the parts that attach to the Y plates reference off the inside face of the plate, so thickness shouldn’t be a factor. I THINK you’ll be fine.

Ryan doesn’t like supports, and he doesn’t like knowing how thick flat parts are going to be. It shows in his design of the mp3dp especially. So thinner, but just as strong should be fine. Worst case, some of the screws may stick out a bit farther.

1 Like

worst case scenario you just print some spacers. :wink:

Thinner is fine but I tested Stainless plates and they bent way too much. I can’t imagine aluminum working better I would cut some wood or another set to back those up.

1 Like

I’ve seen figures for MOE for Aluminum is around 70GPa and for plywood around 10GPa. That would mean that the overall elasticity is about 7x more.

I can’t quite figure out the original is a 1/4" or 1/2" plate [*]. Going from 1/4" to 1/8" is a 2x change, and leads to a 2^3 = 8x reduction in moment of inertial. That would mean that the overall change is negligible. This provides some theoretical support for @81VR6’s very nice rule of thumb.

But going from 1/2" to 1/8" means a change of 4x cross-sectional thickness, and thus a 4^3 = 64 difference in moment of inertia, for a total reduction of stiffness of about 1/8th.

If an 1/8" balsa wood plate were used the additional mass would be negligible and this would easily increase moment of inertia well past that of a plywood or MDF plate.

That being said, I’m shocked that the lever arm is long enough to twist an 1/8" SS plate. I guess I haven’t looked that carefully at that part of the build yet.

[*] v1engineering specs the original y-plates as 1/2" MDF or plywood.

I’ll cut a backup set out of wood as my first part :slight_smile:

Since the School I am a FRC mentor at (go Team 4160) is closed due to the virus, I’ll be without access to a commercial CNC for a few weeks. So I’ll get to see if these are solid enough to cut wood.

Just got a notification that my kit shipped, and I’m 1/5th done printing parts…
Wood pile shrunk nicely building my table, and a trip to the metal store tomorrow to pick up my order will complete the last bit’s I need.

You can use the first wood you cut to back the Al plates. :wink:

The internet suggests the MOE of MDF is ~2.5GPa, so you likely currently have something which is around half the original part stiffness. That’s not bad, and considering that these parts are not optimized to kingdom come there’s a good chance that you’re within the range of acceptability, depending on the part you’re trying to make.

BTW, if you machined another Al part and made an Al -- foam -- Al sandwich you’ve have an incredibly stiff plate. You can use any blue/pink/green rigid insulation foam you might have sitting around. (Just make sure you don’t crush the foam with screw loads).

P.S. I’m in a similar boat, my hardware kit should arrive tomorrow and I’ll start checking local metal suppliers for tubes. So excited!

Worst sandwich ever. :stuck_out_tongue:


TLDR; if you can machine 1/16", then a composite aluminum foam endplate will be just as light, but 7-30x stiffer.

Haha, it certainly does’t sound appetizing, but if we can resist eating our handiwork a brief analysis shows that a sandwich with foam is quite promising in terms of rigidity:

The bending moment of inertia of a rectangle is cubic with the thickness. Let’s define the original plate’s thickness as 8, as in 8/16ths. What happens if we replace the plate by a 1/16 Al -- 8/16 foam -- 1/16 Al sandwich?

Ignoring some scalar constants, the original inertia is 8^3 = 512. The new inertia is (8+1+1)^3 - 8^3 = 488. So pretty much the same!

But, Al is 7-30x stiffer (MOE) than MDF. So the overall stiffness leaps by an order of magnitude, and, because MDF is ~.7g/cc and Al is ~2.7g/cc for a 4x difference in density, at the same time overall weight doesn’t increase because we’ve replaced 1/2" wood by 1/16 + 1/16 = 1/8" Al.

But what if we were happy with the original stiffness but instead wanted to reduce weight? Could we try 1/32", aka 0.5/16ths?

(8+0.5+0.5)^3 - 8^3 = 217. So the inertia drops in half, but that’s okay because the enhanced Al stiffness (MOE) still makes up for it. And our weight dropped by half.

How far can we take this? Well, the math tells us that 1/128", aka 0.125/16ths, is the sweetest spot: (8+0.125+0.125)^3 - 8^3 = 50. So a 7-30x stiffness (MOE) multiplier puts us squarely back around the original stiffness. And at only 1/16th the weight. Win!

Of course, at that point we’re abusing the mathematical theory and neglecting the engineering reality, which is a) how do we intend to machine a part out of what is effectively 7 layers of heavy duty aluminum foil and b) have we utterly compromised the part’s strength and robustness?


I understand the theory in practice. I’ve been on top of what’s essentially a 1/16" tinplate 1" foam sandwich used for a warehouse-sized freezer.

1 Like

@ Kenn, In San Diego, “Industrial Metal Supply Co.” is the place to go for any kind of metal including the required tubes, at reasonable prices (cheaper than other places here). They have other locations in the South West…

1 Like

Not to get too far into the weeds, but at some point the ‘x-axis’ compressive strength starts to matter as the parts will deform the surface and try to pull the bolts through. Now… if you had localized stiffer bond plates… that would be one hell of an optimized part!

1 Like

I miss that place…soooooo much! best metal shop I have ever been in, by far.

1 Like

Or just fender washers?

1 Like

Well just for everyone’s benefit I buggered mine up by not putting the horizontal bearings in when I put it all together.

So have cut out 2 out of 10mm ply and put 5 layers of glass/epoxy on one side.

But then thought more about it all.
So am going to put a pile of glass/epoxy on a piece of plywood and when it dries, cut 2 Y pieces out of it. Save me all the work.

I cut 2 out of 10mm marine ply and epoxied glass cloth from boat building days onto the outside but now looking at it I see fiddly work ahead to trim. I did not think straight. So will keep them as a reserve.

So dunno what the modulus of whatever is of this will be but it won’t be doing much bending I am sure of that.

Stay tuned. I just figure if I have to pull it apart and redo then I might as well make it as flex free as possible.

Your man in the colonies.

Maybe hit it with an oscillating multi-tool? The one we have at the sailing pavilion melts through composites like butter.

Also, AFAICT so long as hole placement is correct what’s going on at the edges is really unimportant.

Regarding the stiffness… yeah, it’s gonna be stiff! A nice layer of FG or CF placed on a thick board is gold. Best would be if you could do both sides, but that’s tricky without a vacuum pump or a large glass plate and a bunch of lead.

I know it’s a funny question but if you’ve got a dust extractor, why not fiberglass first and THEN cut the parts out?

In a general sense, glass and resin are expensive, and they wear through tools quickly. Although for a roughly rectangular plate I think that’s a great option.

@Robertblackburn, correct me if I’m wrong, but that’s what I understand you had said you were going to do with the new piece, right?

Yes I glassed the ply, got estlecam to do 2 copies with the XY set at the very middle of the epoxied glass and away we went. Managed to get it done and it is perfect

Then used the old mdf one as a template to drill out the stepper holes.

Have reassembled and will give it a run tonight.