Large build plate ideas please.

So I'm building a rather large i3. The rough size is three square feet. Needless to say the stock frog plate (i think thats what its called) wont work. So I need some feedback on a good way to mount a build surface to this beast. I need to keep it light weight as I'm still using nema17 motors. I will be printing pla mostly and am not planing on running a heated bed at this point. I'm thinking of using square stock aluminum for the frame and topping with 1/8" plywood or lexan. I guess the only problem I see with that is that the aluminum under frame is whats holding up the flexable plywood or lexan. This would mean that to level the bed I would need to shim the frame where it meets the printed lm8uu holders. Would that work ok?

I'm not a mechanical engineer by trade, but I'm experiencing a multitude of reactions to your plan.

1) This sounds really, really cool.
2) This guy/gal is going to have to go with something beefier than 8mm smooth rods to ensure that axis travel isn't bowed under the weight of those rods, particularly on X.
3) Regardless, that is going to be one heavy bed. I wonder if nema 17s can handle it.
4) Any print that takes advantage of all that build volume is going to take a very, very long time with typical nozzle sizes. Weeks, maybe.
5) If I were to try to build a printer this big, I'd probably use that as an excuse to explore a corexy or delta implementation to move the nozzle around, and let the bed move in Z, if at all.
6) Should this project advance, I want to see videos.

Good luck.

If you are going to make it an "i3" and a 3 square foot build area, that would imply a ~ 8" wide by 4' long bed. You would need an 8' travel to make that work in an i3 format. Even a normal i3 is a bit challenged in terms of Z to Y axis stability. Stretching things to 8' is probably going to make that a significant limit on your design. Sourcing the 3m long rods for that is also going to be a bit exciting. Next layer is that a 4' bed supported in the middle couple of inches is going to be a bit unstable. I think something around 10' or so is probably the shortest you probably can get to work.

If you are going anywhere near this long, dump the round smooth rod. Go with supported linear motion rod. It will droop a lot less than any round rod you can reasonably manage to get.

As you take your belt out to that sort of length it will be "interesting" to keep it from being an accuracy issue. There are a lot of ways to address that, I can't think of any that are cheap.

Sorry I should have been a bit clearer. The width between the y axis smooth rods is a bit less than three feet probsbly more like 2 or 2 1/2 feet wide.

Length of the smooth rods is three feet. So build area will be about 2 foot by 1 1/2 feet.

Not really related to my question but my z smooth rod is also 3 feet long. As for stabillity on the z I'm bolting it to my wall or very large 2x4 frame.

well don't know exactly what you mean but we expanded the i3 aswell
and have this design atm.

prusa i3 expansion

I see you extended the Y axis. My X axis is as wide as your Y axis is long. My Y axis looks to be about the same length as yours. If this dosnt work I figure I probably have enough materials to build two regular size i3's

Here is a picture for scale.



Edited 1 time(s). Last edit at 11/21/2013 08:48AM by hcker2000.

yeah we had the problem that extending our y-axis mad the 8mm rod bend in the middle, also because of the weight of the bed itself, so we are using 12mm rods now instead of 8mm and find this much more stable. therefor you have to look beyond the standard 3dprint shop and look for a lm12uu.

I think your going to get stability problems, if this is your frame or are you going to connect this with your y-axis frame, with a diagonal rod?

I believe that with normal diameter thread rod (or even not normal diameter) you will have quite a bit of swing and sway in your frame. Some of the aluminum extrusion based designs might be a better starting point for a printer with those sort of dimensions. I’ve tried using walls and floors to stabilize things. They seem to move a bit unless it’s the basement floor and a concrete basement wall.

Over that large an area plywood (as a substrate for your bed) is likely to warp. That's why particle board is often used as a well-supported substrate - unfortunately the resin content makes particle board extremely heavy. That's also why plywood used for subflooring is usually tongue-and-groove, 3/4" thick and heavily glued down. Plywood just has a tendency to warp.

I haven't tried doing this myself (constructing a large, lightweight yet totally flat moveable bed) but it seems you'd want something like an aluminum channel structure supporting the right gauge of aluminum plate. It needs to be thicker than sheet but not too thick so that it makes things heavy. The aluminum stock sold (usually in 12" wide rolls) for use in gutter and roof flashing is almost thick enough.

There seems to be a relationship like the squared / cubed law for supporting a flat surface area and the weight needed for the support structure. The larger the area the heavier it gets, geometrically, and this is compounded by needing to have it moveable. All of the large-bed CNC machines I've seen use a moveable gantry with a fixed bed for really large areas, and if you're doing that it opens up another set of possibilities and challenges.

It's an interesting project!

I'm going to try to get a piece of dibond 1/8th inch thick and build a frame of square tube aluminum.

You might want to look into a stressed skin box approach. It's a very common solution to a problem like this.

Are you talking about some thing similar to how they used to do plane body's and wings? If so thats an interesting idea.

Another idea that popped to mind would be using some of the thick pink panel foam. The ridged stuff used for houses. You could easily cover that with aluminum and glue the aluminum to it. Would be light and I dont think it should flex over a two foot span.

You can get most of your strength from the outside of a structure. Stressed skin is how you build an airplane or a race car. Strong thin sheets on the outside tightly bonded to a frame. There are lots of ways to do it. If you dig around you can find people setting pickup trucks on top of towers made from pieces of cardboard .....It's also a useful technique in woodworking when done with thin plywood.

I get the general idea but am having trouble finding any info online about diy methods or the best way to secure the skin.

Really simple approaches:

Plywood on top
Plywood on bottom
Plywood cut in strips in between. (or even better - poplar boards)

Brad nail it up (to keep them all in place while clamping), then clamp it to dry.

Alternate approach:

Aluminum sheet on top
Aluminum sheet on bottom
Aluminum rectangular tube sections in the middle

Carefull (as in small) dabs of hot melt glue purely to hold things while clamping. Epoxy it together and clamp it flat while it dries.

In both cases you have a rim around the outside and some sort of grid inside the “box”. The materials don’t have to be terribly heavy duty. The interconnect between the grid elements can be pretty terrible and it all still works amazingly well. You do need to get all of the grid elements (and rim) well glued to the top and bottom sheets. The main trick is keeping it all together while you set it up and glue it.

You might be interested in the ob1.4 - there are some large builds on utube.

Quote
hcker2000
So I'm building a rather large i3. The rough size is three square feet. Needless to say the stock frog plate (i think thats what its called) wont work. So I need some feedback on a good way to mount a build surface to this beast. I need to keep it light weight as I'm still using nema17 motors. I will be printing pla mostly and am not planing on running a heated bed at this point. I'm thinking of using square stock aluminum for the frame and topping with 1/8" plywood or lexan. I guess the only problem I see with that is that the aluminum under frame is whats holding up the flexable plywood or lexan. This would mean that to level the bed I would need to shim the frame where it meets the printed lm8uu holders. Would that work ok?

Lexan or polycarbonate will expand greatly when heated. Although you don't intend to use a heated bed, the heat from the object being printed will warp the bed a lot. Why not use 3mm (1/8th") aluminium (Australian spelling), it is strong, light and doesn't warp so much.
To make it even stronger and do away with the frame, you could fold the edge over on two (or more) sides to form a channel or box. The folds don't have to be big, just 10 or 12mm would be plenty.
As for leveling, look at the bed leveling system elsewhere or this forum, it is cheap to make and works a treat. No need to have to level your bed any more and that also does away with your problem.

---

[regpye.com.au]
"Experience is the mother of all knowledge." --Leonardo da Vinci