Larger RepRap machine?

Hello.

I'm building a new RepRap machine and was wondering: can you make bigger Rep Rap? So it can print objects larger than 8x8x8 inch...

On a glance, it doesn't look too hard, but maybe there are some software constraints . Did anyone try?

Ivan

On the cartesian robot I'd be worried about the sagging of the steel bars. You can use thicker bars or a different shape to fix this.

Hi,

Large format, multi-tool is what I'm thinking as well.

I'm starting with a 2'X2'x2' table testbed with all the drive gear transferable to an 8X4x2(bed) floor-mounted sheet material cutter (laser and router).

It's a bit ambitious and very early days at the moment.

I'm getting sidetracked by the software solution I'm after which I'll Post in detail, elsewhere for advice, look up phatsprinter software...

So, what I have got already is some very nice approx. 4x4 aluminum track, that comes in 6' or 12' lengths, (pic attached) available fromABS online in the states for $210 for 12'pole with 2x 2"x6' punched steel racks (as in 'and pinion') pre-attached.

If your in Europe contact me and I'll beat the american prices...(this is not an add, I just bought a lot, a big lot...

Actually a VERY BIG lot (2400 feet or so)

So have a look at the pic, as the basis of an affordable Gantry Crane style RepStrap.

The Poles are Designed to lift 500lds at 15 feet per miniute vertically to 48'.

Horizontally the can handle 3x the load of non-profiled 4x4 ali Box section at about the same cost...

I don't know if there's any point using the 120:1 gearbox that runs of a 750W drill I'll have to think on that...

Edited 1 time(s). Last edit at 05/07/2009 06:13PM by john@drivenupthewall.

Sorry forgot the pic....

---

talk soon

john

The real problem is that if you change all 3 dimensions by a factor of 2, the printing time goes up by a factor of 8 at the same resolution. On the other hand, if you want to keep the print speed constant, you'd need to use 6mm filament at double the extrusion speed. This means 8 times the extruder power (which is already a limiting factor in many ways). Alternatively, use 4 3mm extruders in parallel at double the speed (hard!).

Anyway, you'd need to move 8 times the extruder weight around at double the speed. This means really beefy (expensive) motors with ditto drivers. Given the long printing times we have now, I'm really not sure wether the current technology really scales up. Maybe print *very* sparsely filled structures...

Another problem related to a larger print area is that of layer bonding. With the smaller print area it is easy (or unnecessary) to keep the previous layer warm enough so that the new layer can bond firmly to it. Again this is a print time issue on a small object the layers print quickly enough that the previous layer retains enough residual heat (with perhaps some help from a lamp or hot air blower) that the heat of the new layer allows the two layers to melt together seamlessly. As the print gets bigger and layers take longer to print the first part of the layer tends to cool to a point where the new layer simply hardens on top of the old layer rather than partially remelting it and fusing the 2 layers together. The 3 possible fixes for this are:
1: Print faster- see GeertB above for an excellent summery of the problems with this idea;
2: maintain the print area at an elevated temp- Probably means boxing the whole unit(expense, bother) and running the motors, glides etc in a high temp environment(generally bad) and lots of problems with temp control(top too hot bottom too cold, large heavy objects sag at temps that are OK for small light ones);
3: reheat the work area as you print- expensive,slow, with problems related to differential expansion/warping as one end is heated while the other is cooling.

It seems likely that larger areas are going to present major challenges for thermoplastic fused filament modeling for a while. Controlled deposition of epoxies might be a better approach to this problem, or perhaps LOM modeling or, of course, subtractive techniques.

Now I'm no expert,in fact a complete novice at this (I only started building my repstrap last month), but I, too, have been wondering about scaling up and had thought about the problem of keeping a larger workpiece warm.

Although I was thinking of slightly smaller than 2 foot x 2 foot x 2 foot, I was thinking a cheap and readily available heating method in the States/Europe would be either the heating space of a breadmaker or a slow cooker. Probably the slow cooker, as the heating isn't so extreme and it can be kept on for a long time - about as long as you need. Heating comes from the bottom/sides, there is already an insulated box around the element, and I'm sure you could arrange a "lid" holding the extruder to slide across the opening, thereby maintaining temperature.

Just a thought.

---

Hazel

[copycatbot.blogspot.com]

The heating itself is easy, the rest (especially cooling those parts that shouldn't be too hot) is what complicates stuff. However, the speed issues are the real deal breaker for scaling things up.

Instead of designing larger cubes, it seems to make more sense to increase just one dimension, so it becomes possible to make long skinny objects.

-Geert

the thinking for starting with the 2 foot cube is to allow for 'on machine' build-up of composite material components.

the reasons are:

to reduce cleanup support webbing or the need for form boxes in high sided elements

to allow for bonded in elements i.e magnets, coils, electronics, etc

I'm designing a DCPM Axial hub motor similar in concept to ones you can cast using of resin,

so I'm looking into whether RepRap extruded materials can replace the resign or at least be used to make the moulds?

I think my biggest issue may be material shrinkage?

comments welcomed

Are there any software constraints? Will it work if I just make it bigger (not taking problems noted above into account), or I need to code something in the software?

Hi Ivan,

... look here: [forums.reprap.org] - maybe some more ideas?

Viktor

Heh, I could just imagine: building a smaller RepRap (current standard size) inside the giant one.

---

-------------------------------------------------------------------------------------------------------

"The whole is more than the sum of its parts." - Aristotle

[www.cyrozap.com]

Cyrozap Wrote:
-------------------------------------------------------
> Heh, I could just imagine: building a smaller
> RepRap (current standard size) inside the giant
> one.

I was thinking the total opposite: Whats the smallest size for a darwin to still operate and print parts for its bigger counterpart.

A larger machine need not mean larger objects.

You could use it to produce more objects under remote supervision
before human hands are required.
Or print parts onto larger, existing objects.
Or mill a larger object (with a fan blowing the dust off the work-surface)
and then print things onto it.

Looks like it's all just software, calibration and sensors.

Scenario 1:
"Remote operation for extended time with 2 webcams (extruder-head
and overview) and a kill-switch"
Who cares if it runs all week, every time you can supervise
the next part of a series to be build from a webcam with a
remote kill-switch?
Next weekend it will be done (or something happened that requires
a human) and you go down to the workshop to collect, refill, clean
and set things up for the next week.
...optionally with multiple tool-heads to extrude, mill or grip+move
parts on the larger work-surface partly via remote manual control.

So, larger machines have other uses then just building larger objects
via extrusion.

---

-------------------------------------------
* homeprototype free 3d design repository
* Blog
* Google+

Hi just seen your post.

Its a year old how much progress have you made?
I want to build a machine to make moulds or patterns for composite manafacture.

you said you had some good extrusion,

If your machine is working could you build some parts for me to short cut building a big machine no point in reinventing the wheel!

Oh and where are you based?