Clear Resin Casting

I’m doing some research for a lighting project my firm wants to develop further. We’ve built a few prototypes and an A-level finished product which was installed, but we’re looking for ways to improve the fabrication process. Just some overhead information: We’re a 5-person studio with in-house fabrication and we’re hoping to increase furniture production to give us some more work.

We’re making a chandelier and the main component is a clear resin-cast ingot (looks sort of like a blimp) which is quite large (80"x12"x9") and weighs around 160lbs. It is cast in multiple layers. First is a base clear layer, then a pigmented layer, then a milled layer, and I think another clear layer on top of the milling. This creates several different refractive effects and the result is quite beautiful but there are some issues to be addressed for which I’m hoping someone might have a good resource or wisdom to share…

Here’s the important part…

It’s being suspended by cables near the ends and because the load is between those points it looks like it has creeped due to gravity. What is the timespan that we would expect to see noticeable creep in a fully cured resin cast? It could be that whoever finished it by hand did this (I wasn’t working here when the prototype was made).

What is a good material for a reusable mold in this kind of process? We’re going to make it ourselves. We were thinking UHMW, MDF, fiberglass…

Please let me know if there’s any other information that would help you help me.


not sure what you mean by “creep” yss. Is the cable actually starting to cut through the resin?

Polyester casting resin shrinks a lot with shrink rates as high as 6%. Lower shrink rates can be obtained by using fillers (to displace the resin) but then you don’t have a transparent part anymore. You are aware that there is a difference between “regular” laminating resins and casting resins … yes? Far too much information to cut&paste here, but if you google “casting resin shrink rates” you will find a lot of information to read through.

Mold wise, it doesn’t seem to matter, in my experience, what the mold is, polyester/fiberglass, plaster, RTV silicone tooling, etc… the casting material shrinks so fast that the material at the mold surface is still uncured when the interior of the part is done. This is especially true if the thickness of the part varies greatly… like say, a teardrop shape (blimp?). A flat slab would cure more uniformly (although I’d be really surprised if it stayed “flat”)

If you can imagine this… as the material shrinks away from the surface of the mold, tacky strands of gooey resin form strings that stretch between the still uncured surface of the part, and the mold. The net effect is something akin to spider webs on the surface of the part… a major pain the axx if what you want is an optically clear part. Lots of hand sanding is required to polish it out. But the effect can be stunning.

Somewhat like the apparently random surfaces Italla developed for their Ultima Thule line of glassware (although the glass process is far more controllable).

My understanding is that all plastics will creep over time, sag due to gravity or other stresses. I researched it all a few years ago and have forgotten most of it, but I think you’ll have to deal with the creep in some way, pre-cambering, hiding with curvature, additional wire support in the middle of the span.

Here is a thesis on PMMA creep.

The material creep is something that you are just going to have to put up with or design for. Nxakt is right, there are many ways to ways to plan for creep with the geometry of the part. You might try using a few different resins to see which one has the least amount of creep. This would be best accomplished by casting some test strips and then coming up with a loading scenario that helps accelerate the process. Maybe you could support the strips by their ends and put a weight in the middle. Check them after a few days to see which one has deflected the least. I assume you are using a polyester or epoxy resin? I would also try some of the urethane plastics you can find at these sources: and I have used the Alumilite clear on a few prototypes before. It turns out great, but it needs to be degassed and pressure-cast. I think Smoothon has some slow-cure products that don’t need these added procedures.

As far as the mold is concerned, if you are looking to make a bunch of these, then I would go with a fiberglass/ gelcoat mold. They are expensive and take a long time to make, but they will last for hundreds of castings. Also, it is a hard mold, so draft needs to be taken into consideration. Silicone or urethane molds will work if you only want to make a handful. At the size you are talking about, a soft mold would need a support structure, usually fiberglass. The big thing to remember with molding is that the surface finish of the plug is the surface finish of the finished part. If you polish your plug to a mirror finish, your parts will come out with a mirror finish. If you have scratches on your plug, your parts will have scratches. Spending the extra time to get the plug perfect can save hundreds of hours of finishing time later.

Thanks for the helpful comments guys!

Soak: I think we have been using Smooth-on Crystal Clear 206, but your testing idea is a great suggestion. I’ll get on that right away.

Keep us posted yass. I’m curious as to how your parts will come out. I never had any luck with polyester leaving a “mold perfect” surface.

And post a few pics of your project when you have time.