What's about a really BIG reprap for mega-sized objects ...

Hi folks,

... what's about building a very big reprap (maybe 2 x 2 x 2 meters) with wood- or better bamboo-sticks for dispensing some centimeter thick trays of construction-foam?

This would be a real eye-catcher at fairs or presentations and this sort of reprap could build really big (but coarse) objects, as for example scanned 3D-data from visitors, really usable furniture or other mega-sized objects ...

With my tripod-approach i can build the mechanics for selfreplicating (=reprapping) further systems with embedding stiff sticks and plates in the foam and use bigger spherical magnets as joints - all other parameters should be adjustable in the software ...

For other foam-types there could be tested different fast hardening 2-component-foams or self mixed organic material (as in my school-chemistry-experiments) - maybe corn-powder and stark too?

Construction-foam is best for use, because it's ready, but with self-mixed 2K-organics i can tune the expansion-volume and the hardenig time.

With a steady tray a normal tube output a filament of 5 millimetres, which expands in some seconds to nearly 2 to 3 centimeters diameter.

My last used construction-foam was nonadhesive in 5 minutes and hard in 2 hours, so it's not suitable for big/heavy objects without inserting sticks and distance-holders.

But for tests with a reduced slice-count it could be used and for high-stacking there should be mixed a special 2K and maybe heated after dispensing for faster hardening too ...

Viktor

Edited 1 time(s). Last edit at 10/03/2007 02:29PM by Viktor Dirks.

I think that this is a fascinating concept. There has been some really interesting work at the "macro" end of the scale - see: [www.isi.edu] for a paper on the subject. I am sure that many of the long-timers on this forum have gone over this many times before... :-)

rmadams Wrote:
-------------------------------------------------------
>
> I am sure that many of the long-timers on this forum
> have gone over this many times before... :-)
>
Yup, about a year ago I wrapped up work on an unfinished prototype I named Godzilla which had a much bigger build volume than current reprap prototypes. That's all back in the blogs from that time.

... for me the Mega-reprap-concept would be very interesting, because here i can show the supreme capabilities of my magnet-joint-hinges in a tripod, which normally is not meant to build bigger than some centimeters in size because of the structural weakness of normally milled solidstate-hinges ...

So i will probaly try it on my own ...

Viktor

VDX Wrote:
-------------------------------------------------------
>
> So i will probaly try it on my own ...
>
By all means. A few things to think about, however, before you invest a lot of money and emotional energy in the effort.

It's going to be pretty easy, at least in theory, to scale up your positioning system. Is the same going to be true of your extruder? A bigger work volume implies that you are going to be extruding at a considerably faster rate. You might want to take a look at the Leister line of "hand held" extruders to get an idea of what you are talking about.

[www.hotairtools.com]

Pay especial attention to the amount of power those bad boys are drawing and how much they weigh. Is your positioning system going to be up to slinging an extruder that heavy around?

Another thing to keep in mind is that you aren't going to be extruding a 0.5 mm thread of plastic any more, so you can expect that your surface resolution is going to suffer.

Finally, for HDPE, at least, you are going to somehow have to compensate for a considerable degree of thermal contraction as you print your object and it cools. That is, unless you plan on enclosing the whole shebang in a heated enclosure kept just a few degrees below the melting point of the plastic. That is a BIG energy issue.

I'm mentioning these things not to discourage you. I REALLY would like to do big extrusions. These are just scaling problems I personally have run into trying to think my way through a workable design.

Hi Forrest,

... in the first post i described how to build thick layers out of construction-foam ...

The outputted 2-components-mixture expands 5 to 10-fold, so a tray of 2 mm thickness (as extruded from the pinhole on the tube) would expand to a finger-thick cylinder.

A normal tube has a weight of merely 500 gramms and the end-volume of the foam may be 15 liters or more!

In school-experiments we mixed a 2-k-foam, which expanded from some centiliters in a cup to a football-sized halfsphere and was hardening in some minutes ...

With a two-way-extruder and a mixer-tube as tip i think, the weight of the toolhead would be less than 200 gramms and the two liquids could be inserted through elastic tubes from above.

With a tripod-structure out of bamboo-sticks and spherical 19-mm-magnets as joints i should be able to move around a weight of some kilogramms without to big displacements because of bending in the supports, so a 2 meter high bamboo-tripod-reprap should be possible and not to expensive, because of the extreme limited mechanics.

I estimate a price of maximal 300 Euros for the (bamboo-) mechanics. The motors, drivers, linear spindles, bearings and most fixing parts are already in my scrapery, so this isn't the problem - i'm on evaluating the conversion from STL-data to tripod-moving, here is the trouble and time-consumption ...

But actually i'm not on the Mega-, but more on the micro-reprap - here i have a 1Watt-Laserdiode with a 50-micron-focus for melting, sintering or cutting plastics and can mix different powders to good dispensable pastes, so my first try is to build a high-accuracy-mechanics and play around a while with micro-fabbing ...

Viktor

I was looking for the same thing then I found the mondo
here's the link: [www.reprapcentral.com]
I can't find any documentation on it though.

I dont think it would be too much trouble to increase the size of a reprap. After all there are large cartesian and delta robots used in industry. As the accuracy doesn't need to be so fine, you could get away with rack and pinion linear drives and servo motors.

The problem as has been suggested would be with extruding. I think the expanding foam idea may be a little unpredictable.

There has been some success with concrete extruding machines for making buildings. Perhaps your expanding foam idea could be refined by using some sort of filler in the foam such as rice or grain. This would increase the wet strength of the material so you could build higher structures and be less dependant on fast curing. It is the agregate in concrete that gives it the strength. An additional bonus would be that the expansion would be more predictable as you would be using less foam in the mix.

Hi Viktor

I really like the idea and using expanding foam sounds like a nice option. I would be interested in helping design / plan / build a machine like that.
Do you know what happens to the foam as you build up the layers? - Would you need to do each layer slow enough so that the foam has expanded before the next layer is added.
Or do you do it all quickly and just keep on adding layers while it is all expanding, then what happens with the Z steps, do they scale as the object grows / expands?

I was thinking you could use 2"x4" timber frame and make a 2M box and a large scale steel wire/pulley system like Forrest has used on the re-darwin to position the extruder.
Then maybe the build-bed (Z axis) can be controlled by a counter balanced system using water to lower and raise the build height ?

What motors...

Another post had a robot arm building chairs, since then I have often thought about how to build a reprap style big machine... the problem I am stuck on is what material to use / extrude - foam could be a nice solution.

Rich.

---

[richrap.blogspot.com]

Hi Rich,

... my initial post is nearly 3 years old, so not really urgent

When i tested with expanding foam, i made this manually and tried to receive uniform tracks, but it's not so easy, as the diameter is decreasing with reducing pressure of the driving gas.

I think you need a 2K setup, where you mix the two components in the nozzle and adjust the expanding factor by percentage of the components.

Another hint is combining the foaming with CNC-milling, where you adjust the height (or contour) of the actual layer ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

VDX Wrote:
-------------------------------------------------------
> Hi Rich,
>
> ... my initial post is nearly 3 years old, so not
> really urgent

- I didn't spot that! - sorry Viktor.

Big RepRap's seem to loose their focus when you carry on using plastic filament and extruders, it's just too slow and costly.

I'm going to think more about the foam, it's not expensive to get a 2part kit and 1Kg seems to make a LOT of foam.


Maybe you can also add fillers (fibreglass strands) in the mix, or just coat the finished part in resin.

---

[richrap.blogspot.com]

This is somthing I am interested in and have thought about for car modelling and large mould making. I also wondered about PU foam as I have used it many times before and have used very quick curing "spray on" which goes off in 15 secs max on a warm day. Which would be perfect as it comes in presurised tanks and is extruded or sprayed though a disposible mixing nozzle.
The trouble as mentioned is the unpredictability of the expansion. Running a very careful smooth line down results in a nobbly fat line.
I came to some conclussions:
You could set up the system so the maximum expansion would always be lower than your desired ammount. that way you wont have any high spots, just lows that can be filled.
You could do the opposite if you wanted to or just live with it and assume an average expansion and work your figures out accordingly.
You could also use a dot matrix style of printing rather than continuous flow. Perhaps that would give more consistency??
The best idea I could think of would be to have a cutter following the print head to ensure each layer was at an exact height ready for the next. Then the cutter runs a cleanup routine after the print to take off the excess as if it was milling the object from a block.
Or you could just use a rasp and sand it down. you can sand PU surprisingly smooth.

One other problem is that if you stop for more than a couple of seconds, the mixed PU will go off in the nozzle and block it. This may be a big problem if you need to traverse to another point.

As for the actual printer itself. I was thinking about using XY hoist Tracking usually used for lifting disabled people out of bed etc. Its very sturdy and will easily take the weight of some large PU A+B tanks along with a Z axis& head of huge proportions.
Its also very smooth and usually has linear bearing in the crossbeam to allow for any deviation in wall parallel.

Just my 2 cents as its something I have already considered.

Edited 1 time(s). Last edit at 06/25/2011 08:53PM by ablainey.

Is it possible to air spray layers of the foam product seperately in aerosol form instead of extruding the premixed foam product? You might get more consistent foam layers that way.

Yes you can spray from up to 16 inches away to get a thin coat. It still has lumps and bumps and is a much wider coverage area. You could probably narrow the spray but I think the problem lies with both the mixing and the fact that random spray is generally globular by nature. (Look at the universe!)

It may be possible to change the mix ratio or add a third part (filler) to reduce the expansion but still get a quick cure.

Hi
Just want to share this with with anyone intrested in building a big extruder [www.factum-arte.com] Anish Kapoor, a sculptor got help from Factum-arte building an extruder for concrete
And look at the rest of there website also.
Enjoy

Pardon for bumpting this old idea, but I think its a really cool thing you are doing.

Have you considered this? Instead of using a system of 'arms' why not replace the arms with 'pullys', and configer it as a delta robot [ something like this [ars.els-cdn.com] ].

edit: This video shows the potential of parallel cable robots [www.youtube.com] .

Obviously with pulleys you can't be as fast and as accurate, but you can scale this to pretty much print entire houses if you really wanted to. You can try improving stability by making the extruder heavier, you could also consider (if given enough money) to also include a spinning metal circle around the extruder as a gyroscope.

Edited 1 time(s). Last edit at 05/29/2012 08:00AM by mofosyne.

... maybe you'll find following posts interesting: [forums.reprap.org]

Go through the atached images and links ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Its really difficult to say...

Ok guys, just another interested partner here.

Till now my team has explored the use of cartesian bots as cheap cnc machines. We've built a 2 x 1.5 meters cnc table powered by a Sanguinololu board: [vimeo.com] to cut foam slices that can later be glued as a stack (no extrusion here).

Right now we're starting a research project to find the right monomers that together with paper paste can be cured quickly with UV or heat sources, so we can built an extruder to replace our milling machine in the cnc table. Our project has an additional need: we need our material to burn easily (I know it is not a common requirement) and not to create nasty (dangerous) gases in the combustion.

The idea of a cutter (ablainey suggests) is something I have seen used in some clay or concrete extruders on youtube.

VDX Wrote:
-------------------------------------------------------
> Hi Rich,
>
> ... my initial post is nearly 3 years old, so not
> really urgent
>
> I think you need a 2K setup, where you mix the two
> components in the nozzle and adjust the expanding
> factor by percentage of the components.

Here is a crowd that uses 2 part expanding foam and two hot air guns to make sure it cures real fast. Not sure what they do to avoid the setting in the mixing tube problem but perhaps they keep the two components very cold and only heat them to fast cure as they leave the nozzle.

There might be some way to use dielectric heating (you read it here first) on the mixed compound as it leaves the nozzle for instant heat instead of waiting for the hot air to turn on and off.

[www.mataerial.com]
[www.gizmag.com]

This was posted on the RepRap facebook page on 25 May 2013

Kalle
--
Johannesburg, South Africa