3d printed inflated structures

Replies to This Discussion

Permalink Reply by Ethan Gross on June 18, 2015 at 4:49pm

Hi Fergal,

Really clever! I thought silicone is considered fairly "non-stick", so how do you manage to print on it, even if you're printing in silicone?

Ethan

Permalink Reply by fergal.coulter on June 22, 2015 at 3:45pm

Hi Ethan.

Well indeed silicone is tricky to bond other materials with...luckily it bonds very well with itself (provided it's the same type of catalyst used - you can have tin or platinum based silicones which won't work together) . In this case i use two addition cure platinum based slilcones which bond quite well together.

Permalink Reply by martyn hogg on June 19, 2015 at 6:18am

Great work! I am impressed that you machine looks so clean and is not covered in silicone!

What are the end products used for or are they purely aesthetic?

Are you printing rigid material onto silicone balloons or is the print media also silicone?

Permalink Reply by fergal.coulter on June 22, 2015 at 4:27pm

Haha, yes - I cleaned the machine up for the videos! Luckily silicone peels off very easily - but when you spray it, it really goes everywhere. The fume cupboard that i normally spray in is covered in little flecks, which is a pain to clean!

The material I'm printing here is also a silicone - albeit a much harder one - the balloons are soft - a shore hardness of 00-30, whereas the printed structures are Shore A 73 (or probably harder, as i use ceramic fillers too.) This is really the two ends of the hardness spectrum for silicones.

The purpose of the structures i create is for use with artificial muscles for medical devices - specifically Dielectric Elastomer Actuators (DEA). These actuators function by having a stretchable conductive electrode on either side of a silicone (or other suitable elastomer) membrane. They are effectively flexible capacitors.

When you apply a voltage across the electrodes (a high voltage of 2-3kV) the opposite charges attract each other, compressing and therefore elongating the dielectric membrane. 

An example can be seen here : 

https://www.youtube.com/watch?v=JcRF_y7fWZ0

DEA work more efficiently when they are stretched. This is why I inflate the 'balloons'.

I then want to prevent the pre-strain  from collapsing back on itself when i remove the internal pneumatic pressure. This means trying to create a 'Minimum Energy Structure', which is really a 'tensegrity' - structure combined of struts under compression balanced with a net of continuous tension.  In this case it involves bonding an incompressible but flexible 'frame' on to the stretched (inflated) membranes. 

Here is an example of a dielectric elastomer minimum energy structure created by a research group in Harvard:

https://www.youtube.com/watch?v=nZm36KMVvy8

....So! the extruded 'chiral' structures that I've been printing onto the balloons were me testing the hypothesis of my PhD - that you could use auxetic geometries to partly hold the stretch of a balloon, allowing it to shrink or grow (i.e. actuate) without buckling into strange shapes. 

(if that makes sense!)

Permalink Reply by martyn hogg on June 25, 2015 at 5:35am

Thanks for the explanation, really interesting! I prototype things in silicone at work but we 3d print moulds to inject silicone into. Working with silicone is extremely frustrating as it's so messy and seemingly complex moulds can work whilst seemingly simple moulds can have loads of problems.

There's a company in Sheffield; Picsima that are 3d printing in silicone apparently but I'm not sure how their process works. Also the Carbon3d technology looks really promising for printing elastomers.

Personally I'd be happy with a process that allows mixing, degassing and injecting without covering me and the workshop in silicone!

Permalink Reply by Nik Willmore on June 25, 2015 at 5:47am

I don't bother degassing any more since I set up a paint pressure pot as a 70PSI casting enclosure after wrapping it with a thin bucket heater to maintain it at 160 °F. An insulating ceramic plate via McMaster.com was cut for the bottom, and the threaded feet of the pot allow leveling. It rids bubbles in mixed silicone and then later in powder-filled epoxy castings in that silicone.

Attachments:

• image.jpg, 964 KB

Permalink Reply by martyn hogg on June 26, 2015 at 2:47am

So are you injecting into a mould at 70PSI or are you holding the silicone and mould at 70PSI and letting the silicone gravity feed into the mould?

Permalink Reply by Nik Willmore on June 26, 2015 at 3:05am

The latter. High pressure injection would be scary. The point is, no matter how much air you entrap in the silicone during mixing, the high pressure simply quashes all bubbles, and the heat of 160 °F measured by a dial probe screwed into the pressure pot lid creates an actual internal temperature of about 140 °F that means a silicone like Smooth-On Mold Star 30 is fully cured in a mere 15 minutes. I only have to decrease the temperature for some epoxy castings that might exotherm under heat since they are too large in bulk size and lack heat absorbing filler.

I use a California Air Tools CAT-TL5 Ultra Quiet and Oil-Free 1/2 HP Tankless Air Compressor $190 compressor that is very fast but lacks any pressure cut off, so uses a foot pedal instead:

http://www.amazon.com/gp/product/B008MPH1WK?psc=1&redirect=true...

Gravity is why I mentioned using leveling feet and use of a bullseye level gauge on a flat ceramic insulating shelf ("rigid ceramic insulation" from McMaster.com) in the bottom of the pressure pot.

A mere pressure cooker won't be safe beyond 15 PSI let alone 70 PSI which is why I used a paint pot instead.

Permalink Reply by martyn hogg on June 27, 2015 at 4:39pm

Interesting... I might look into this method then. I've also tried vacuum bagging moulds like they do with carbon + epoxy. That worked quite well but you need holes at every possible air trap on the mould which can make it time consuming to clean the mould afterwards. I wrapped the mould with scotch cloth, then put it in a vacuum bag. Too much vacuum induces degassing bubbling in the mould though.

My concern with gravity feed would be complex moulds would not fill properly but it sounds good for simple geometries... or have you had success with complex geometries?

We often cast into moulds with cores and for parts with very thin wall sections (~1mm)

Usually 40A - 60A silicones.

Permalink Reply by Nik Willmore on June 27, 2015 at 4:52pm

It's not the surface detail that is any problem, down to microscopic scales, since small bubbles just disappear, especially if you first brush the silicone molds with some of the uncured epoxy first. The real problem is large undercuts or channels that retain such large amounts of air that it likely won't just be absorbed into the epoxy, assuming swirling the mold about won't release the air. But realize that the bulk air pressure in the pot will itself act as injection pressure both for the initial silicon mold making and the eventual resin casting in that silicone. I do use quite thin epoxy, which happens to be rather toxic, but is so hard it buffs really well without softening. The heavy metal powder fillers I use also help, tungsten being the heaviest. I'd need a drawing of your application type to better understand what your thin walls are about.

Permalink Reply by David Stasiuk on June 19, 2015 at 6:37am

Looks like an incredibly interesting process...really smart. I'd love to see more about it!

Permalink Reply by fergal.coulter on June 22, 2015 at 4:29pm

There's a little more on my website :

http://www.fergalcoulter.eu

At the moment i'm just finishing up the write up for my PhD thesis, but then i'll be publishing much more about the process. I'll keep you updated! :)

• ‹ Previous
• 1
• 2
• Next ›
• Page