Note: This instructable is entered into the "Make it Real" challenge - so if you like it, please take a second to vote by clicking the button in the upper right of the page - Thanks!
This Instructable covers one process of converting a digital character model from a virtual model to a "real" world work of art. I will be covering model preparation for rapid prototyping, cleanup of the "3D print", mold making, resin casting, and final painting - I hope I can do it all justice ;)
Background: A while back, I had posted a digital render of a character I'd modeled on a popular CG (Computer Generated) art forum. One of the posters on the forum wrote that when he'd originally seen the image, he thought it was a photo of a "resin kit". At the time, I had no idea what a "resin kit" was, but after a little research I found that resin kits were pretty cool and I thought it'd be fun to make my own - having no idea what I was getting into - lol. I was also inspired by the idea of taking something virtual and making into a "real" object because as an artist who works predominantly in the digital realm, much of my "art" was not physical - it was made of bits and bytes of data - and should electricity ever go away I'd have nothing to show for years of staring into a monitor.
The obvious solution was to turn to Rapid Prototyping technology - but back "in the day" (2005) you had a choice of incredibly-expensive and nice quality, or, somewhat-less-expensive and fairly crude. I figured that crude would be OK since I wanted to learn how to make molds, had more time than money, AND I didn't want to risk messing up a $10K "print" while learning how to make molds (not that I could have paid that anyway). Today, as with all computer related technology, you can get better quality for a lot less money - but the techniques are still relevant.
At the time I started this project, there wasn't a whole lot of information available on making molds and casting resin. Yes, there was plenty of information aimed at industrial casting operations, but not much for the "guy in his garage" - so a lot of what I learned I learned by reading a lot of semi-relevant material and making mistakes. Some of them expensive mistakes. I'm going to try to point out those pitfalls so that hopefully (should you decide to blaze down this trail) you can avoid them.
Through a lot of testing, I found that there is no "right or wrong" way to make molds or cast material - just more-or-less efficient and more-or-less economical. I approached this project like a course in molding/casting and I think I learned more than I ever could have if I'd have paid someone to teach me. My goal was a very high quality polyurethane resin final product and molds that would be capable of small-scale production - and that drove a lot of my decisions (using pressure casting, buying a *real* vacuum pump, more expensive silicone etc). If a person is interested in just a one-off casting, a lot of money could be saved by using a "mother mold" system and cheaper silicone - although block molds are a lot easier for the noob mold maker ;)
Costs: To be blunt, RTV Silicone (Room Temperature Vulcanizing) isn't cheap, and neither is polyurethane resin. Costs can be reduced, again, by using "mother molds" and "slush casting" - but you will exchange your time for the savings (I won't be covering those techniques). The good news is that currently there's a lot of information available on those techniques -an internet search or YouTube search will bring up a veritable cornucopia of information. If you are looking for an even more economical way to mold/cast something, I'd suggest looking into using urethane rubber or Alginate for molds, and some of the super-hard plasters like Hydrocal to cast parts. Anyone who sells mold-making supplies can probably help you make decisions based on your goals and budget.
Materials and Supplies: What you will need depends on the materials you choose and your desired final product. For *this* project, the "core" materials and tools were:
Step 1: Prepare your model for printing
Show All 15 Items
There is a huge variety of software that a person could choose to model with - each with it's own strengths, weaknesses, and workflows. There's some really good free software available which will work as well as some of the expensive packages. For those interested, I'd recommend looking at programs like Wings3D, Sculptris, Blender3D, as well as many listed on this page: Free Modeling Software.
Check with the printing service you plan to use to find out the specifics regarding how they want your model prepared. Since my original model started in a spline-based application, I definitely had to do more cleanup than someone who starts in a polygon-based app, but never-the-less you'll have to make sure your model is very well-structured to avoid getting it rejected by your printing service. In general, your model will be required to be "water tight" and have "thickness" everywhere. While you don't have to make one contiguous mesh, you do at least need intersecting volumes that are completely enclosed. In areas where you might have cloth or thin surfaces - like clothing, drapery - or in my case, feathers - you need to give those parts of the model "thickness" so they can actually be printed. You will also need to make sure that any thin items meet the minimum thickness required by your printing service or the results might be too fragile to be practical. Also keep in mind that many services charge by the volume of material used in the printing process - so taking the time to make your model a hollow shell with some thickness instead of one thick piece can save printing costs - but again - check with your print service to find out what they charge for and how to optimize your model.
Since I wanted a final product that was larger than the available print volume (8"x8"x10" in the case of 3DArttoPart)) - and I planned on making molds - I broke my model up into logical pieces and made sure the all the pieces fit inside the print volume. I did this to make sure that I wouldn't have to pay for *two* prints to get the size I wanted. This actually turned out to be unnecessary although it was an interesting experiment in packaging ;).
When deciding where to break the model up, it's important to keep in mind the mold making process - taking care not to have radical draft angles or entrapment of the mold on the master part. While it's true that just about any shape can be molded, there are small considerations a person can take early on to make life a lot easier when it comes to the mold-making and casting process.
Step 2: Handy Tools to Have on Hand
Show All 8 Items
Some tools that are great to have - but not really required: A pressure/casting vessel (a large paint pot in this case purchased off eBay), a vacuum pump capable of a very high level of vacuum (eBay again), an air compressor (to pressurize the casting chamber), a vacuum chamber for use in de-gassing silicone (mine is a Nalgene 5305-1212), and a cheap toaster oven (that you don't use for food) for heating up clay and post-curing silicone molds (not necessary - just speeds up the process).
So, why de-gas your silicone and pressure cast your resin? Pressure casting yields much better results than "gravity casting" or casting at atmospheric pressure. Molds that have fine details will almost certainly end up with bubbles in the castings - either from air entrapment, surface tension issues, or moisture in the mold (resin+water=foam). Casting under pressure (80PSI in my case) basically "crushes" the bubbles down to almost nothing - or to a size much much smaller than they would have been without it. Of course, this same phenomenon (crushing the bubbles) ALSO works on your molds - so if your mold has bubbles below the inner contact surface, those bubbles will be "crushed" and will produce bumps and bulges in the surface of your casting - so - if you want to pressure cast your resin, you need to have bubble-free molds to go with it.
If you aren't going to pressure-cast parts, you can skip the compressor and pressure pot. If you are going to use a mother-mold system, or, some of the thin-viscosity mold-making silicones (less durable but easier to work with) then you can forgo the vacuum chamber and vacuum pump. What you choose really depends on your end goal and what you're willing to put up with in exchange for your time and money (Isn't that the case with just about everything?)
Step 3: Prepare the Print for molding
Show All 41 Items
The 3d printing service that I chose (3DArttoPart) uses a powder/binder system for printing the model. While this is probably the least expensive of the 3D techniques, it leaves a surface that is less than optimal for pulling molds or painting. Think "100-grit sandpaper" and you have a pretty good idea of what it's like. If you want to spend more money, you could get a much smoother model to work with, and it would be a perfectly legitimate argument to say that the extra money spent would be worth it.
To smooth the surface to the level that I wanted, I quickly realized that sanding the surface of the print wasn't going to work - the surface was too hard and inconsistent - so I had to come up with something else. The technique I settled on was one where I'd flood the surface with 2-3 coats of laquer sanding sealer (light sanding between coats) followed by rattle-can FILLER primer (not sealer) and a series of passes of sanding, chasing details, priming, sanding, chasing details, etc, until I was happy with the surface. Dupli-Color brand Filler-Primer seems to work the best for me - most other brands I tried stay too gummy. Epoxy-based filler primer would be ideal - great build up and super-easy sanding characteristics - but it's a bit expensive ($20/can).
Once your surfaces are smooth, it's time to fill in undercuts and open loops that might "trap" the mold as well as just make parts harder to extract from the molds. You don't have to make the areas "flush" but it's a good idea to minimize them without compromising the design. You'll also want to add small details and details that didn't come through in the print. For example, areas where a bump map is used on your model, you'll have to recreate that detail with clay (epoxy sculpt in my case) or hard wax. This is also the stage where you can cover one part of a mating surface with clay and press it onto it's corresponding part on the model to create a "key" so that when the model is assembled it will fit together perfectly.
Model preparation is important - remember that the surface quality of your castings will never be better than that of the "master" object - so take your time in the preparation steps.
Step 4: Adding Detail
Show All 10 Items
As I mentioned in the last step, adding details is something you'll need to do unless your model contained them all. Where I had used bump maps to "fake" detail in the CG rendered version, I had to recreate that detail using clay (a note here - when I say "clay" I'm referring to Epoxy Sculpt).
Now, a problem arises: How do you make a really thin (1/32") layer of Epoxy Sculpt (clay) that is *consistent* in thickness across it's surface? Trying to do that manually (slathering it around until it's a uniform thickness) is way beyond my skill level or patience - so I came up with a pretty simple trick: Roll the clay out like a pie-crust and then stick it on the model. The problem then becomes "How do I keep the clay from sticking to my rolling pin? (piece of PVC pipe, in my case) I found that if I put the clay between two pieces of wax paper, I could roll it out, peel off one side, and use the other side as an applicator - it worked really well. One little trick to keep in mind: Once applied, do NOT immediately start sculpting and adding details - the clay will be too soft and will turn into mush pretty quickly. What you want to do is wait about 25-45 minutes until the clay is starting to get firm, and then start adding the details. Coat the clay with a thin layer of petroleum jelly so that any tools you use will slide easily and not stick to the clay - and go ahead and add detail as needed.
Step 5: Casting the Arms & Making Interim Cast Parts
Show All 38 Items
I decided to make interim castings of the arms for 3 main reasons:
Preliminary molds are also a great way to "get your feet wet" in regards to making molds - the outcome isn't super-critical and you can make mistakes without it feeling like it's the end of the world.
Step 6: Molding the Torso
Show All 36 Items
The torso of this model was a challenge - it's big and it's littered with places to trap air - places where I didn't want to put vents because I'd later have to cut those parts off and risk making a mess of some of the curves of the model. I thought about it a while and realized that if I rocked the mold when it was about 2/3 full to allow any trapped air out, then followed that by filling the mold completely, I could probably get away with one sprue placed in an area that wouldn't be so critical shape-wise during assembly. The sprue ended up poking out one side of the mold (usually, the sprue runs along the seam of the mold) - which is OK because it doesn't "lock" the casting in place. Sometimes you have to do things a little differently ;)
A few notes on silicone:
Liquid Silicone varies in viscosity from pretty thin, to VERY thick. If you donít have vacuum equipment, youíre going to want to use as thin a material as you can that still meets your needs. Viscosity of the liquid silicone is measure in unit called poise and indicated by a number followed by "cps" (centipoise). The higher the number, the thicker the material, and the harder to degas - so keep that in mind when deciding what you want to use. Once again, take advantage of the experts at your supplier of choice - they deal with this stuff all day. ;) Less viscous silicones will mix easier, pour easier, trap fewer and smaller bubbles and degas on their own more readily. The downside to them (generally) is that they don't make as durable a mold.
The "hardness" of cured silicone is measured on a durometer and represented on the "Shore A" scale - so usually looks something like "Shore 20A" where the "20" represents the hardness of the cured material. The higher the number, the harder the silicone will cure. A mold made from silicone that has a high Shore A rating is more difficult to deform accidentally when clamping and tends to be a bit friendlier to work with once it's been made. If, however, I were going to make a "mother mold" type mold, I'd probably use one of the softer silicones in the 30A and below range. More on hardness here.
For this project, I used a product from Shin-Etsu (KE-1300T) - a clear silicone that cured to a Shore-A hardness of 40 (about as hard as a running shoe sole). This stuff is THICK - think "cold honey" and you have an idea - just mixing in the catalyst was a workout by itself - lol. As you can see from some of my "mistakes" - failing to degas this silicone adequately will result in poor-quality molds - but then proper degassing will yield very solid molds that can take a lot of abuse.
From the Shin-Etsu Data Sheet:
To eliminate voids within the rubber before cure, air entrapped during the mixing cycle must be removed. To accomplish this, place the mixture under a vacuum of 28-29 inches of vacuum. As full vacuum is applied, the material will "froth" and expand about 4 times its original volume (lies! - Expansion can be MUCH more ~ Joe), crest, and eventually recede back to its original level. (A film coating along the container sides should be evident above the original volume.) The deairation cycle is complete approximately 10 minutes after the frothing ceases. Should the container size used be inadequate for deairing the mixture, the vacuum may be broken during operation to reduce the bubble formation. To remove air without using a vacuum system, place the mixture in a freezer overnight. The silicone will stay liquid but will not react to cure. The air bubbles will slowly rise to the surface.
I never tried the freezer method so I can't vouch for it's efficacy. What I CAN vouch for is that with degassed silicone, the few bubbles introduced during a pour would rise right to the top and with inadequately degassed silicone, the bubbles would just "hang" in suspension and not move - I have no idea why.
Be sure to leave plenty of "headroom" in your mold barriers if you plan to degas your silicone in the mold - it wonít foam up as much as it did if you degas it before you pour it, but it *will* foam - so be sure to babysit it while you degas it to prevent it from foaming out of the mold (don't ask me how I know.....) After experimenting with both methods, I found that I preferred to degas the silicone in a larger container and pour it as opposed to doing it in the mold.
When degassing silicone, donít leave it unattended - even for a few seconds. It WILL expand to many times itís liquid volume - at least 3 to 5 times and probably more. Additionally, you need an almost PERFECT vacuum to degas effectively - which means a high-quality vacuum pump - an HVAC evacuation pump won't cut it.
It takes about 10 minutes to de-gas silicone - sometimes more, sometimes less - you'll know you're done when it stops foaming under vacuum. The beauty of translucent silicone is that you can see the bubbles. The difference between *degassing* and just *foaming up* a little (vacuum-wise) is very small. If your vacuum pump and chamber canít "boil" water, youíre not going to be able to effectively degas very thick silicone, and only partially degas the thin stuff. Also, in the process of degassing, you can pull a vacuum, wait for the bubbling to slow down - then POP the vacuum to burst surface bubbles - then pull a vacuum, wait for foaming to slow down, then POP the vacuum again - do this about 3 times, about 3 minutes apart.
(The ZIP file at the end of this step is a video of the de-gassing process).
TIP: I found that putting the poured mold in direct sunlight really sped up the curing of the silicone (which can take several hours) and also had the added benefit of making bubbles rise much more quickly as well.
It's worth mentioning again - but detailing the clay edge - where it intersects the model - is very, very important. This will become the parting line on you model. Itís also critical to clean all excess clay off of the model - otherwise it will become part of the mold, and subsequently every casting you make. I use a "Lecron" carver and other dental laboratory tools - they seem to work the best for cleaning up this area of the model, and they're pretty inexpensive. Itís also important to make the "ledge" around the model as smooth as possible - this will help make sure that your molds are easier to clean, and seal more effectively.
Silicone doesn't stick to much of anything - other than itself and some porous surfaces - so you need to make sure that when you're pouring your mold halves you use an adequate parting compound when pouring silicone to silicone (you don't need parting compound when pouring silicone against clay - except on the mold barrier). An inexpensive mixture I found that worked very well is Naphtha and Petroleum Jelly. Mix about 1 part petroleum jelly with about 2 parts of naphtha (solvent) and paint it on with a brush. The naphtha will "flash" off leaving a really fine layer of petroleum jelly. Remember to coat the inside walls of your mold barrier when you're pouring your silicone! While it won't be impossible to get the cardboard separated from the silicone, the PITA factor will make you remember to do it the next time ;)
Just as with silicone, casting resins are rated for viscosity and hardness - as well as pot life, gel time, and how long until fully cured. They are rated on the same viscosity (cps) scale as silicone, but they are generally rated on the Shore D hardness scale as opposed to the Shore A scale. I chose harder resin for the small and thin parts - especially the sword - because softer resin has a tendency to droop over time. The harder resin is more brittle so it can be trickier to de-mold parts cast with it - but on some parts (thin, non-supported parts) it make sense to use it.
A word of warning: Resin gets very hot, very quickly, when it's curing. Don't underestimate how hot it can get (180+ F is typical) - so be careful when working around it. Additionally, molds and any containers used for mixing will get hot as well. The resin generally won't heat up until it starts to gel, but always keep in mind how much time you have before it starts to "kick" - you don't want to be holding on to a plastic mixing cup full of resin when it starts to go cloudy ;)
To follow up on the comment that the resin will heat up the mold - realize that a warm mold - or hot weather - will cause your resin to kick even more quickly. It's a good idea to let molds cool between castings, and be aware that hot weather will shorten your work window.
When preparing a mold for casting, you'll want to make sure it's cool, clean, and DRY (water + resin = foam). Use a good quality silicone-based, spray-on parting compound to help ease the removal of your finished part. Parting compound also reduces the surface tension between the mold and resin (fewer bubbles and better penetration into detail areas) and extends the life of your mold. You will probably want to apply the parting compound before each casting - but you'll have to be the judge of that (the resin will come out without it - but it will reduce the life of the mold).
A big mistake: At one point, I didn't make any castings for a few weeks and when I went back to make some more castings I failed to shake up the A and B parts of the resin before measuring/mixing them. There was no apparent settling or stratification in the containers, so it didn't strike me that they needed to be shaken. Mixing these parts together resulted in resin that didn't cure fully in the mold - leaving behind a mess that was *amazingly* hard to clean up. I came very close to tossing the mold and making a new one. So the lesson to walk away with is to **always shake up your resin components - even if it's only been a day or two since you made a casting**.
The process of casting:
Go ahead and fill the cups to their respective marks.
In a well-ventilated area, while wearing safety goggles, gloves, and old clothes, you will do the next steps in 3 minutes or less (assuming a 3-minute gel time):
At this point, you can walk away for 20-30 minutes and take a break. When you are ready, release the pressure valve on the pressure pot and bleed off the pressure slowly - this will be pretty loud, so you might want to wear ear plugs. Once the pressure has bled off, remove the lid, remove your mold and demold your part. Let both the part and the mold cool someplace undisturbed. Thin parts should be supported while fully curing to prevent them from sagging. Don't use the mold again until it's cooled off - or you might have resin curing in it MUCH faster than you anticipated. When the mold is cool, clean it, inspect it, and get it ready to do it all over again ;)
De-gassing Process.zip5 MB
Step 7: Molding the Legs
Show All 45 Items
Not a lot of commentary on these steps - pretty much the same as the torso but for the legs. Any additional details will be in the captions.....
Step 8: Wings, Skull, Sword....
Show All 15 Items
The skull, wings, and sword were probably some of the trickier parts to cast - either complex shapes (skull) or thin. I had to make some additional parts to mount the wings to the skull since the CG version isn't burdened by inconveniences like "gravity" where things actually need to be attached to keep from falling off .... sheesh.
Step 9: A Few Details
Small parts were clustered together in the molds to make casting them more efficient.
Step 10: Sculpture Assembly - Pre and Post-Painting
Show All 36 Items
The general steps in assembly and painting are:
Step 11: Painting
Show All 33 Items
These images cover most of the painting process. Not a lot to say other than I use a combination of acrylics (mostly) and some laquer (mostly as a sealer and "dull coat") Application techniques span the range - airbrush to kitchen sponges to colored pencil - whatever works :)
Step 12: Final Results
Show All 9 Items
Well, it was a LOT of work, but I learned a ton doing it. Thanks for taking the time to read through this beast - I hope you enjoyed it and I hope you can find something of use in it.
Step 13: Pressure Casting vs. Gravity Casting
Show All 8 Items
Pressure casting versus gravity casting - pros, cons, and a few problems that were encountered....
Step 14: Molding the Head (the wrong way)
Show All 7 Items
This was an attempt to make a monolithic block mold for the head. This did NOT work very well because not only was demolding the part like opening a bear trap (the silicone was too hard for this style of mold), it made a harsh parting line along the back of the head that would be very difficult to clean up. The final mold is split front-to-back with a parting line along the seam on the side....