Friday, 7 December 2012

Breaking the mould (literally)

Bad news, everyone- I have been unable to complete my foam latex build-up puppet of Gollum from The Lord of the Rings, which was intended to be built up atop my ball and socket armature. This was not through lack of trying. I did everything I could over the last week to see this project through to completion. I have barely slept trying one technique after another, each time being disappointed as something has gone terribly wrong. The worst part? Every element that has failed me over the last week has worked perfectly for me before, and I cannot fathom the reasons why anything should have been any different this time.

In my plans for this micro project, I researched heavily into Richard Svensson’s brilliant tutorials for building foam latex puppets, whereby he creates a character’s skin in pieces via plaster moulds, which allows the replication of finely sculpted details as opposed to the rather more basic method of painting latex directly over a padded foam puppet body. 

I devised my own simplified variation of this technique- to sculpt a full body character with a solid, permanent medium such as Super Sculpey, to press the model into air drying clay to form imprints that retain almost all of the detail, and then (once dry) to paint latex in layers directly into the simple clay moulds. This way, you get to keep the original sculpt as a display piece, and you still get moulds that can be used repeatedly for the fabrication of puppets.

I will admit, I held my reservations about my technique, unsure as to whether or not it would work as intended. My main concern was whether the liquid latex would fuse with the porous clay inside the moulds, for air-drying clay tends to mush a little if water is added, even after it has set. To determine if my technique was viable, I decided to create a mini test Gollum, mould and skin- and am very pleased to say that it worked fantastically.

This test model was detailed to determine if the air-drying clay was capable of reproducing fine details. I began with a simple wire armature, wrapped in masking tape, basing the design on a small Gollum figure I bought when I was a child. The sculpt was completed quickly in a day, for it was a test and didn’t need to be perfect. Unfortunately, the model was a little burnt this time- I made the mistake of asking my family if they could put the oven on 100 degrees for me. I later was shocked to find mini Gollum baking away at 150 degrees. Luckily, no lasting harm was done, for the model remained intact. He just looks like he has frostbite.

Though I had originally intended to make a full two-part mould for mini Gollum, I only needed to see if the clay could create detailed impressions, and if the latex could successfully be removed from the clay mould later on. For these reasons, I quickly created the impressions of his face and body on one side, to save time, as this could tell me all I needed to know before beginning the main puppet.

With the clay having dried overnight, I began painting uncoloured latex into the mould, waiting twenty minutes between layers. The bottle recommends eight layers for a small mould- for this test I used five. For the final puppet, I shall use more to ensure the latex skin is a good thickness. I found that the latex tends to pool in the recesses and run away from the peaks of the mould- as you would expect, of course, but it is something to consider when applying layers. It is important not to be too hasty and apply too much in one go.

The latex was tricky to peel from the mould, though I expected this having seen it in Richard Svensson’s tutorials. The resulting test skin was exactly what I had hoped for, maintaining all the detail of the sculpt. This was all the proof I needed to proceed with this technique for my final puppet.

Unfortunately, making the puppet turned out to be a nightmare, wherein nothing would go right. Things, for no obvious reason, suddenly no longer worked as they had done successfully in the past, and to make matters worse, they all went wrong in succession. As a result, after several hellish days working non-stop (sleeping a total of nine hours in four days), ultimately I was unable to complete the puppet. Nevertheless, I learnt some lessons regarding the materials I was using (as per my learning objectives) and can use this knowledge to my advantage in the future. Furthermore, I do still have a large amount of successful work to submit (my ball and socket armature was a resounding success) as well as the evidence of techniques tested.

With my mini Gollum test model working exactly as intended, I decided it was time to apply the same technique on a larger scale for the full puppet. I began with a large-scale wire armature, to match my 1:1 printed designs.

The model was padded out with tin foil to conserve materials and reduce baking time, before Super Sculpey was blocked out and refined over the top. 

With the model baked, it was time to add the air drying clay to form imprints of the model. When it came to removing the clay however, I discovered that it had stuck to the model surface. The only way to release was to peel it away as carefully as I could- but even that left large chunks behind. I suspect that this difference in result from my test was due to having opened a new packet of clay. The air-drying clay is much softer and stickier when new, and I did not have time to wait for it to harden somewhat before being applied. It would seem that air-drying clay to be used in this manner is temperamental- whilst it can work very well there are some tricky factors to balance. Ultimately, it seems unreliable for making mould impressions, which is not what you want from a modelling material.

From here, I did not know entirely how to proceed. For a while, I was concerned I had no time to trial a different method on the fly. Never being one to give up easy, I decided to try my best at following Richard Svensson’s method of building up a mould container and using plaster to form the mould. The difference was that my model was sculpted solid and could not be picked out once the plaster was hard, though considering Gollum has no overhangs, I believed he would work rather simply with a two part front and back mould. Just like my press mould for my second micro project, I coated the surface of the model with a thin layer of Vaseline to help release it once set.

With Gollum in t-position, and his arms out wide, I did not wish to use a rectangular tray to make the mould, for I would be using an excessive waste amount of plaster. For this reason, I chose to make my own container using foamcore and my glue gun, a tried and tested method again borrowed from micro project 2. The sides were constructed to loosely fit the form of the model.

With the side positions marked out, I began to seal Gollum to the base with Plasticine, bulked out with foil, to meet his mid point (where the seam would be). This first mould would be the front, with a second back mould to follow.

Sealing Gollum took a lot longer than I had expected, and by the time it was over I was very pressed for time. I mixed the plaster, poured it into the mould, and discovered then that I did not have enough to fill two moulds- despite having four 1kg packets, which evidently do not go very far at all.

At this point, I had to come to terms with the fact that the puppet was not going to be finished, for I had exhausted all of my means to make my mould for the latex skin. Even so, I could still use what mould I had made to demonstrate my technique, and I still had the original sculpt to hand in- well, only just.

Once again, to reiterate, I was using techniques that I myself have used previously, with successful results, and felt sure that they would work this time. This was not the case, as it seemed things were only getting worse. When taking the mould out, I found that Gollum, for whatever reason, was sealed fast inside. This time, instead of acting as a lubricant, I can only suggest the Vaseline acted more like an adhesive- sealing everything airtight was the idea, but things ended up pretty much vacuum sealed to the plaster.

It took three hours to carefully chisel away (literally) the plaster to reveal my model unscathed. The image above shows the parts of the mould that released whole, and were able to be salvaged. They are essentially useless, apart from proving the ironic point that the impressions made in its surface were spot on. A week’s worth of non-stop work, only to end up having to destroy it to step backward several stages. This was incredibly upsetting, mostly for the many instances where things should, in my opinion, have worked without problem. I guess that’s always the way when you’re running tight to a deadline.

Overall, I can say that I have learnt a lot of what not to do for making character moulds. The plaster will work great providing the model can be released. You also need a lot more plaster than you would expect! Vaseline is not a good idea, and I have yet to find an alternative or any advice on the matter, though I certainly will do so. Air drying clay is a bad idea it seems for anything large- it seems far better suited to small models.

In a perfect world, I do believe that if I had more time, I could make the plaster mould work. With more time to plan the shape of the mould, I could have conserved space and filled it properly, ensured everything was fully sealed, and used something like ‘Ease Release’ to prevent the model sticking. I do think this technique warrants further investigation, and I shall look into it in my own time.


  1. sorry to hear the bad news Michael. the gollum sculpt looks awesome, its a shame it didnt work that would have been great to see it.

    i dont know if youll want to change your materials but i now make my two part moulds from liquid plastic. its alot easier to mix and guess how much youll need to use. no vaseline is needed just ease release and it works a treat. just a thought.

    all the best, Mark.

  2. Thanks for the reply Mark,

    I'm very pleased with the sculpt, I managed to get it all done in a couple of days! I was hoping to take some proper photos of it, but I didn't have time unfortunately before handing in my project work. I will be getting everything back in a month or so though, and I intend to paint the model properly and add finishing touches then to go on display.

    I'm very open to changing materials to whatever works best! Thanks very much for the advice, I shall look into liquid plastic and hopefully try again to make a mould sometime!

  3. You've presented the topic in a quite thorough yet concise manner, that is genuinely useful when someone wants to get the facts without spending too a lot time searching the online world and sifting out the noise to discover the answers to their questions.

    I'll definitely come back here to see your finish product.

    best colleges for animation

    1. Hi Jake,

      Thanks for the comment- I'm pleased you like my blog! One of my aims has been to try and cycle the things I learn back into the animation community to help others in similar situations, so it's great to hear that you found this post useful!

      I'm back from my Christmas holidays now too, so expect more posts soon as I begin development of my final piece!