Glove Box for bubble free casting??

[Anvils Hammer]

So Ive been messing about with vacuum casting for the first time today.

It occured to me pretty quickly that air bubbles can still remain in the parts because there is air in the molds when you fill them with resin.

We are all sqeezing the molds and pressuring to get the remaining air bubbles out which have been trapped by the resin.

In Rapid Prototyping of PU components, they mix and pour the resin IN the vacuum chamber. so the mold is already completely empty before the resin in poured in.

This means when the mold is removed from the vacuum its already 99.9% full of resin, and atmospheric pressure will crush any remaining air..

So I did some research and it turns out what I need is a glove box!

This is a vacuum chamber with a rubber glove which I can then stick my hand into.

I can put molds on a tray, mix the resin in a cup, and put it all in the box, stick the lid on, hit the vacuum, this empties the molds of air, and degas's the resin.

Then I stick my hand in the glove, pour the resin into the molds..

I can then immediatly take the tray out and start pouring the next tray..

I think this would be easier and faster than pressure casting, and also should theoretically elimiate ALL air bubbles.

Glove Boxes are normally for scientific/bio med/electronics industrys, and cost thousands of pounds.. Im going to build my own!!

What do you guys think of my plan? Thoughts of more experinced casters please!

AH

[mangozac]

The conclusion makes sense I suppose, but building a glove box? The gloves would have to be some really heavy duty rubber to stop from expanding to the volume of the chamber under vacuum. Personally I'd try to go for a more automated approach: some kind of robotic mixing/pouring mechanism that can be sat atop a mould.

My first port of call for an automated mixer would be looking at stepper motor controller peristaltic pumps. These can be used to pump an exact volume of fluid (work out the volume based on the density for weight measured products). So you would have two cups (one for each resin part) each with its own peristaltic pump. Have them pour into a fancy shaped silicone or teflon coated funnel which mixes the two parts nicely as they pour....

What do you think? What are your electromechanical skills like?

[Anvils Hammer]

My electo/mech skils are certainly up to scratch, and I have friends who know more about it. I could do a mechanical solution.

each mold plate will have a number of molds, say 10.. and they can be any shape and size, this means the device pouring the resin has to move up and down different amounts, pouring different ammounts of resin in to each mold, I would need a joy stick on the outside capable of controling pour spout in both axis, as well as being able to pour very precise ammounts of resin..


Basically Id need a robotic limb with the same degree of control as say.. my hand..

A glove is the easier solution..
It will need to be resistant to the vacuum.. there are a few companies making glove boxes and the gloves are availalbe separatly, so Im going to look into purchasing the glove, then building the box..

The glove and seal between glove and box are the only really complicated parts, the rest is plumbing..

AH

[mangozac]

[Anvils Hammer]

sorry i meant up and down as in "across"

yeah you are right, its doable and would make a fun project.. but im trying to do this for a living not a hobby so "quickest" is king.

I challenge you to make a machine that can pour resin more accurately and faster than I can by hand.. Ive been filling molds most days for 9 months...

Dexterity will certainly be an issue, but all I need to do is grab a cup and rotate... simple right..

AH

[blind pig]

[Eltnot]

Don't forget that you'll need some way of looking into a glove box. It should be quite simple, but just something else you need to consider.

[mangozac]

[Lane]

You do not need to pour separate cups then mix. There are nozzles that will connect to two inputs and mix internally, though I suspect you need a third 'purge' input to prevent curing in the nozzle.

Robotics in the vacuum chamber has two problems. Lubricants exposed to vacuum. Lack of air cooling under vacuum.

The primary problem I see with a vacuum glove box is size. It has to be large enough to get that stiff glove in and/ or the robotics. Remember 'pounds per square inch' multiplied by a lot of surface area means a lot of force acting on the structure. Add the probable large flat surfaces and you need a lot stronger material. Also consider the volume in your pump requirements.

[Anvils Hammer]

True.. and automated process which puts exactly the correct ammount of resin in a load of molds, all under vacuum would be awesome and very time saving! but in the short term the development cost/time outweights the benefits.

When volume of production increases.. I will start playing around with Lego Mindstorms prototypes..

Lane you are quite right, and Ive considered those various points.

The key issues are- can the vacuum pump evacuate the air in a reasonable time frame.. and can the box withstand the pressure.

The answer to both problems is to make the box as small as practically possible. Im hoping for around 15L internal volume, My pump needs to empty that in less than a minute to be usefull.. As after I mix the resin Ive got about 2-3 minutes to pour it.

The construction will most likely be 10mm polycarbonate, overlaped and multi layered so tha the vacuum actually makes the box stronger as it sucks it in. Ive got the rough design in my head..

Lots of thinking and research to do...

AH

[mangozac]

[Anvils Hammer]

Absolutly, of course I will be testing the theory first!

I plan to rig up a balloon under a cup.. so that the vacuum causes the balloon to expand, tipping the cup over, or somthing like that..

Once Ive established it works, THEN ill start building the proper box!!

AH

[blind pig]

Has anyone seen the new (well, new to me in Australia) Araldite Self Mixing Nozzle?

http://www.selleys.com.au/adhesives/hou ... elf-mixing

It might solve the problem of mixing liquids in an enclosed container, if not *how* to do the process...

[mangozac]

My understanding is that's how most mixing heads are designed - with some kind of helical piece which mixes the two parts as they are forced through. Of course, cleanup is a pain though...

[Anvils Hammer]

Testing done!!

not perfect results but interesting..

The theory- vacuum first to remove all the air, then pour the resin, which will totally fill the cavities as there is no air in there.

The experiment-

Good old lego-



The mold on left with a splash sheild. The resin is poured into the cup and remains upright while the plate is placed in the pot and the vacuum applied for a minute or so.

A sharp tap on the side of the pot caused the cup to tip over-



..sending the resin flowing in to the theoretically empty cavities.

The results- some small air bubbles..

each part had one or two small air bubbles, which were proportionate to the size of the part.

Bubbles should in theory, be impossible. the only explanation is that the vacuum is not good enough.

I dont currently have a gauge (getting one soon!) so I dont actually know what kind of vacuum im getting, but I do know that when I close the valve on the pot, so that the pump is only pulling on the air hose, the noise the pump makes changes noticably.

This suggests to me that the pump is pulling a higher vacuum when only attached to the hose, which means my pot is leaking slightly, it would only have to be a tiny leak to stop a "full" vacuum from forming.

My pump is rated to an ultimate vacuum of 15 microns, which means in theory less than 0.01% of the air should remain in the chamber.

If my pot had that little air in it.. there would be no visible bubbles, which means a leak of some kind..

The next step is to purchase a good quality vacuum gauge and see what kind of performance Im actually getting from my set up.

I will keep you guys updated with the project.

The aim is have a glove box with a pump or two pumps set up, that will evacuate it down to a true 15 microns, allowing true bubble free pouring.

AH