Vacuum casting aluminium parts from 3D print.

[TheBaron]

Earlier this year whilst working on designs for a 1/24th Westland Wasp, my hand was forced into learning how to cast aluminium (using vacuum processes) on a small scale, on account of the part(s) concerned needing to be much stronger than white metal was capable of providing. 

 

On the principle that the problems specific to modellers rarely feature in Youtube tutorials for such processes, it may be useful for others how this process can be adapted to our hobby on a cheap, reliable, but most importantly, safe, basis as home.

 

In relation to scale and detail,  the species of metal-casting videos in closest proximity to our craft tend to be those made by jewellery designers and from which you can form the misleading impression that you need to drop  €2-3K on kilns, furnaces and the like to even get started. Those guys are working with precious metal on a market-volume routine basis so necessarily need consistent, professional gear to run their practice on a daily basis, unlike those of us needing only small volume or occasional metal parts. 

 

Before proceeding any further  I cannot stress the safety factor enough - foundry work (even at this small scale) involves materials and equipment working in excess of 700°C, so as a minimum you require:

1. A well-ventilated outbuilding or outdoor area (with a concrete or stone floor), with the casting area itself well away from any flammable materials.
2. Safety equipment: protective footwear, along with a thick industrial apron and foundry-type gloves like these: 
3. 

Protective goggles are also advised.

 

If you have neighbours right next door as well, in terms of noise pollution/public nuisance, be aware that the furnace emits quite a roar at operating temperature, as well as fumes when the wax resin is being burned out. 

 

Metalworking gear sold online with the word 'jewellery' appended is as a rule wildly expensive for the home user, however, much of what you need can either be made or adapted cheaply yourself with only basic DIY-level skills. For clarity, the following is split into three sections regarding tools, materials, and then the sequence of events involved in producing the parts.

 

Tools

1. Casting flask & base. These flasks are just a stainless steel cylinder and go for silly  money online. I bought a cheap length of  3" wide stainless steel exhaust (for a Mini) from a local car-parts seller for the price of a single flask and cut it down to produce as many flasks as I need:

The rubber base you see it sitting on is one of these, again, an inexpensive item.

 

2. Foundry tools.

The melting crucible is a 1kg carbon graphite job. You can pick these up cheaply enough online from the likes of Amazon however many of them come without a pouring lip which I found made pouring the metal awkward so you can see at the 3 o'clock position on it tbar I ground out my own lip (easy enough using the Dremel). Both the black crucible tongs and flask tongs (yes this latter are really just a big BBQ pair) can also be sourced cheaply online. Ensure your tongs match the size of both your flask and crucible though -  in my case the black tongs work fine with the crucible but don't open wide enough to lift the flask in and out of the furnace, hence the second pair. A large screwdriver is useful for digging the part out of the mould after casting, whilst the tool on the far right is simply an old BBQ spatula folder over in the vice to produce a handy item for scraping slag off the top of the liquid metal whilst it's in the furnace.

 

 

4. Furnace.

You'll find plenty of rather costly gas propane furnaces online for blacksmithing, jewellery etc., and an equal number of  cheap knockoffs. I sought advice about the cheaper alternatives from a sculptor friend with experience working a range of metals: his verdict was that they were too cheaply built to last, plus also noted that on H&S grounds you still had to spend time making them safe to use in terms of treating the thermal blankets from spewing off fibres.

 

As I had already had one of these roofing torches from previous building works, the heat source for the furnace was already taken care of. If you need to source a burner yourself, there's a lot of simple cheap furnace torches of the venturi type available online, or even build your own. Whichever route you follow, you're working with gas and high temperatures, so be careful.

 

There'd been an empty Calor gas cylinder knocking around in the garden ever since we moved in about two decades ago; I'd never thrown it out thinking 'it'll come in handy one day' and sure enough, this turned out to be its hour of glory, being transformed into a furnace using Brian Oltrogge's impeccablly concise beer keg method:

 

 

Not needing such a large furnace as Brian did, I cut the middle 25% out of my  cylinder's original height, whilst the handles were simply recyled from an old airbrush compressor:

Not needing such a heavy amount of heat as a larger furnace to get up to operating temp. (typically 732°C for flask burnout / approx 700°C for melting the aluminium in the crucible.*) I also replaced the original nozzle of my propane torch with a smaller bore version and made sure that the diameter of the tuyere opening matched it during construction so that it would fit comfortably into the opening like so:

For the refractory lining of both furnace body and lid I followed Lionel Oliver II's homebrew mixture to the letter and it works superbly, as long as your initial warming and firing process to cure it follows Brian Oltrogge's curing steps of air, heat lamp, then steadily increased the burner temp over several hours, which he outlines in his tutorial videos.

 

Vacuum chamber/table

Unluckily my defunct airbrush compressor was a diaphragm rather than piston version, otherwise I would have had a go at converting it to a vacuum pump along the lines used by Tariq Curtis. That said, Vevor sell a decent, inexpensive pump/chamber combination that provided plenty of scope for modification along the lines required here. It didn't have to be pretty, just work consistently, and was modified thus:

The chamber comes from the manufacturer with the connection and gauge mounted onto the lid, so as I needed to be able to swap the lid out for a casting table (which I made from that circular steel plate you can see hiding behind the pump), the side of the chamber was drilled out and the connection repositioned to there, with the original hole in the lid being sealed by that highly professional and sophisticated method of sticking a 50p piece over it using JB Weld:

That then gave me the means to both de-gas the liquid investment inside the chamber when casting the 3d printed pattern, and then by swapping the lids over, use it outside as a vacuum casting table. By no means is that combination an original idea on my part (it's a well known and widely used process with some beautifully-engineered equipment produced by others) but has the virtue of being quick and cheap to produce in this instance. If buying a similar pump/chamber, go for a pump with the higher flow rate as you will need that extra speed at the metal pouring stage.

 

Casting

1. The 3D print and investment mould.

 

The initial casting step proved the most expensive part of the process as, in experimenting with both resin and homemade casting mixtures, you pretty soon find out that there is a reason that specialist materials exist. You gets what you pay for!

 

After satisfying myself that tens of thousands of people around the world weren't lying about the fact that standard resin doesn't cook off when subjected to extreme heat but just turns to ash:

 - I realized that it was necessary to buy one of those printing resins especially formulated for metal casting. After going square-eyed from reading brand-comparisons (as well as bug-eyed at some of the prices!), I settled on Siraya's well-known 'Tech Cast Purple' as a compromise between quality and cost. It's been around a while, is well spoken of, and gave excellent print results first time out:

Did I say compromise? Actually there's no compromise here in quality terms, the surface finish being on par with anything I've got from Elegoo's standard grey: the only issue I found in regular use was the occasional apparition:

This particular blend of resin does require a more elaborate cleaning and curing cycle post-print, however if you're prepared to pay a bit more then I undertand the newer Blue version simplifies this step.

 

Prior to obtaining the vacuum gear and a custom investment powder I explored some of the cheaper options for investment casting first, having several goes at using a 50/50 mix of Plaster of Paris and silica sand and just the effects of gravity to get the metal into the mould. Many people have routinely gotten great results from such inexpensive materials but for producing larger chunkier parts than what I required here, so it was no surprise when I was unable to get the proccess to produce more than a 70% complete print on such small parts. Remarkable however that it yielded that much!

 

A brief reading up on the chemical composition of investment materials rapidly convinced me that for 3d printed materials at this scale, the use of a professionally-produced powder was necessary. Digging around online I was able to find a supplier in Spain who was able to ship a box of Ransom & Randolph's 'Plasticast' blend, specifically formulated for using with 3d printed materials, at a reasonable rate:

It has a lovely creamy texture and performs consistently well, plus you get to sing Mud Mud, Glorious Mud whilst degassing it:

When it comes to the actual mixing and casting process, George on Makerspace gives a far better demonstration of this visually than I can ever hope to achieve in written for. Essentially follow everything he does here and you will get good results. Hat tip to you George. 

 

One crucial difference is that he uses a kiln to fire his flask whereas as I use the gas furnace:

This obviously lacks the precision of a digitally-controlled thermal environment so aside from a handheld thermal temperature gun (which will only give you an approximate understanding of the internal temperature due to confusing reads off the hot gases emerging at the top) you have to work old-school and judge the required temperature by the colour of the steel flask. This is not as hard as it sounds once you get your eye in.

 

One issue I balked at when humming and aahing about buying an expensive investment powder initially was the multi-hour temperature-control cycle the manufacturers recommend for baking your flask. No way would this be practicable or affordable on a gas furnace. I was heartened though to read in Tim McCreight's Practical Casting (a superb handbook to have) that there was a 'quick and dirty' 3hr bake you could use instead (essentially the same routine followed  by George in his video). Certainly this has worked fine in my case, despite the difference in equipment.

 

As to the actual metal pour, this is the setup I use out back, with plenty of leg room between furnace and vacuum table in order to avoid tripping over anything:

The black cylinder is an old airbrush tank cut down and insulated (so that I can stick the baked flask under to keep hot whilst melting the metal down) whilst that silicone gasket on top of the vacuum table is just a cheap piece of catering sheet cut out with a hole in the middle for the flask to sit over. 

 

A fresh pour:

Surprisingly the vacuum setup exerts a stronger pull with hot rather than cold items, I've no idea why. 

 

You want to avoid the flask shedding too much heat whilst out in the open on top of the table as the chamber evacuates and you fetch the crucible of metal over, so I bung a couple of large rocks into the chamber to drastically reduce the volume of air the pump needs to remove.

 

The subsequent cooling and cleanup cycle again follow what you see George doing in his video, whilst the net result of the above has remained pretty consistent from cast to cast. Here are copies of the parts in question - the forks which hold the main rotor blades on the Wasp - neither resin, plastic, or white metal capable providing the required tensile strength to hold the blades without making these fork elements grossly oversized:

These are all still in their raw state out of the mould so the next task of course is to begin cutting them off the sprue and give them a sand and polish - hopefully I can start that over the next few evenings.

 

Is this work going to make any jewellery designers feel threatened? 

 

Hardly.

 

But it is capable of reproducing features down to about 0.15mm in size on parts that need to be stronger than more commonplace modelling processes can provide. And that's all that was needed in this instance.

 

I'll post up a photo of the forks in the finished condition in a few days.

 

Kind regards,

Tony

 

*This Vietnamese foundry produces the best guide I found anywhere online about the temperature ranges for pouring aluminium, in terms of balancing strength with surface detail. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[Austin 7 model]

Thanks for this Tony. It must have taken you ages to put together. I have had some success with cast parts in aluminium, and just as many failures, but I haven't tried vacuum casting usin ivestment. I will bookmark this page for future reference.

Thanks again, Lee

[TheBaron]

13 minutes ago, Austin 7 model said:

Thanks for this Tony. It must have taken you ages to put together. I have had some success with cast parts in aluminium, and just as many failures, but I haven't tried vacuum casting usin ivestment. I will bookmark this page for future reference.

Thanks again, Lee

Glad it was of some encouragement Lee!

 

After the first flush of success as it were, over time I came to find the amount of work it took constantly tending the above arrangement to get 'good' i.e., usable for modelling purposes overly time-consuming, so took the plunge late last summer and invested (no pun intended ) in a proper burnout kiln for the studio. Now it just works away in the background whilst I'm doing other things and rarely a fail.

 

I got mine from Tuffnell Glass over in Yorkshire; Martin there is a very nice guy to deal with.

 

Best,

Tony

[Austin 7 model]

Thanks for the extra info Tony. I would like to get back into casting again, if I can find the time. I have 3D printed the patterns and moulds to make a small single cylinder petrol engine, bore size 30mm. Most of the parts would be sand cast, but the name plate would look much nicer if investment cast. The smallest letter is 1.5mm high, a letter "o", the middle is .5mm It might be easier to drill that out.

 

Lee