Support material speculation

The origin of this thread is a comment made by Forrest Higgs on the peristaltic pump thread.
He mentioned that his next concern now would be the making of a support material delivery system.

I think this issue deserves a differentiated thread so that everybody can give their best ideas on the subject. Here are mine:

I was thinking about waxes, that need to be molten and extruded. You could maybe do the same with gelatin sols. But walking home from work just now i have thought about something a lot easier and very reliable:

What about a water or glycerin based paste? something along the consistency of toothpaste?
It doesn't need to be heated up, it has enough structural strength to support the polymer deposits on top, it has high density assuring that the polymer won't sink in. It can take the heat of thermoplasts. You can make it as inert as you need.... and most importantly, it stays put: It doesn't stretch or shrink.

The recipes are many: Glycerin/water and flour or powder sugar, Glycerin/water and sand-dust... any type of solid, inert filler in a suspension medium

yup, if you read through the blog, thats exactly what Adrian has been experimenting with lately. what you're basically describing is equivalent to wood glue. we still havent settled on a decent material yet, but there are quite a few ones that seem to be pretty decent.

Ehemm... indeed.
Polyfilla, i heard that name a lot in the forum. Basically it's plaster cracks filler paste, right?

Man... there's just an incredible amount of amazing info in those blogs!
It would be so nice if we could index and categorize each entry to be able to browse and retrieve it.

Right. We call it "spackling compound" as a generic term in the US.

***It would be so nice if we could index and categorize each entry to be able to browse and retrieve it.***

You can get at most of them simply by using Google with the search term plus "reprap". That's what I do.

i do the exact same thing =). hopefully simon will get me access to the wiki/forums servers soon. one of the things i'd love to do is to integrate the google site search with our stuff.. for example the wiki search is hidden on the lower left and gives terrible results. googles search however, is rock solid.

/end threadjack

yaaa, the wiki, wait 'till i get my hands on that one!
I have so much news on the UV set polymers that is begging for a place to document it.
I think I found a REALLY cheap solution... waiting for the products to arrive so that i can start mixing!

Actually you know what? I think that it would be an advantage if the support material never completely sets. That way, it would be far easier to remove, even in labyrinthically intricate parts, where some support material remains stuck in somewhere. just dip it in water and it all dissolves away. I guess you could disrupt the drying or setting capacity of the mix by adding some percentages of liquid soap to it.
This way the mix is easier to deposit too, and doesn't inadvertently set in your syringes or dispensers... just an idea

You know...you could just build your fabber in a chest freezer, then use water as the support material... :>

problem: hot plastic melts ice :-/

Granted, not the most universal solution. However...won't it melt wax as well?

Are there very many, safe and available, materials that freeze at a considerably lower available energy, than they melt? If not, I think it'd be difficult to implement the filler and the main material using heat. You might make it using wax and plastic, but wouldn't the plastic soften about the same time the wax melted out? And if not, wouldn't the plastic melt the wax as it was being laid down, same as the water?

And yes, I know most materials change states, to a degree, like I asked. You can cool by boiling something, for instance, or every air conditioner invented before the peltier junction would be worthless.

Perhaps a material that is particularly susceptible to a given solvent...syrup, for instance.

Although, I was being partially facetious.

I just had an idea.

Use the same material for support as you're using for everything else...then paint it with cooking spray, or some other lubricant.

For hollow objects, this wouldn't work well, as the filler would persist in being too large to exit any holes, but for other things, it might.

However, it'd be nice to be able to lay the oil down the sides. In this manner, you could "stack" supports laterally, as well as vertically, meaning you could pull the first one out, then shake the second one out the hole left by the first.

P.S. This is what I get for thinking about these things while hungry and facing the kitchen. :>

P.P.S. Be aware of the flashpoint of whatever lubricant might be used.

When I was working for a company that manufactured hydraulic valves, we once had a rapid prototype made for a casting core. I do not know _how_ they manufactured this core, but do remember that it was all one type of plastic. What they did to create the support material, was using very small, thin (i.e. 0,5 mm) "spikes". They could easily be broken of, which left you with the actual part. When using FDM, it would mean that you would lay down a small series of "dots" where the support needs to be, and on each layer add "dots", stacking them up into tiny, supporting spikes (can't think of a better word to describe this). When the object is finished, you can easily remove them.

Now, this might not really work for holes, but it could do the trick when there is a larger hollow space beneath the object to be printed. It would require a certain level of accuracy, and the advantage is that you can use the same extruder and material, which might speed up the manufacturing process and can probably be solved in the software part of the reprap machine.

Joost

i see... so you use a 1 'pixel' column of your normal thermoplast as the support? that just might work. they would either break off very easily, or be cleaned up with fingernail clippers in a jiffy.

Was the casting core rigid or elastic? I could see the sprue technique working quite well with something that was rigid, but I can see problems with it if it were elastic. Namely stretching and tearing in spots other than the connection.

Also, how viscous is the gel, when being applied? If it is sufficiently viscous, some overhang may be created without a support material. This would probably require the head track on two additional axes, however, to point the nozzle at the built up surface.

Something else to consider. Build the part in a box. If there is a continuous container around the part, then some powder might work. Sand, for instance, or perhaps something finer that wouldn't interfere with the finished product if it was embedded. Talcum powder, for instance.

As for the box. Build it while you build the part. You'd only need it if there was overhang on the outside of the item, or if there was a hole that ran laterally to the working surface.

I don't think I've seen anyone trying to make spikes like that yet. I know there are some early tests about thickness of walls etc. The trick is going to be cutting off the flow of the thermoplast. If we drag it along as we've seen in many pictures it might not be able to turn on and off and hit a spot repeatably. Maybe someone can design a cutter? to snip the flow or another mechanism for causing a "dot" to be laid down. But at least there could be a hollowish structure built with thin walls that don't connect to the main structure securely. If the intent is a cavity it would maybe require a little cleaning, and if just a reduction in material it would remain hidden inside.

I've been able to print dots with HDPE. They are about 2-3 mm in diameter, though. The response time of the Mk 1 extruder doesn't let me get any smaller than that.

How about doing the printing in Polypropylene (PP) and using PLA as Support. PP is chemically very stable, so maybe we can find a solvent for PLA, that is non toxic and does not attack PP

Not many chemicals bother PP, so you will be able to remove the PLA .
But PP will warp really bad, much more than ABS.

---

[richrap.blogspot.com]