Some thoughts on resin warping & shrinkage

Tom Madden <tgmadden@...>

Can't offer much in the way of solutions to this very interesting
set of problems, but if you'd care to consider some possibly useful
insights, read on….. (Most of you know a lot of this already, but
it's good to get it all down in one place.)

As Pierre says, styrene and cured urethane resin are two different
materials. They have different physical properties, and they respond
to heat in entirely different ways.

Styrene is a thermoplastic material; it can be re-melted and formed
into new shapes many times over. The heat necessary to melt styrene
is applied externally, and styrene solidifies when it cools. Any
shrinkage is due entirely to its coefficient of thermal expansion.
If the injection mold is held at a fixed temperature and the cycle
time is locked in, all parts will match exactly. Subsequent heating
and cooling of a styrene part will not change its physical
properties or dimensions. (As long as you don't melt it!)

Urethane resin is a thermosetting material; two liquid components
(base and catalyst) are mixed together and contained in a mold until
the casting cures. The heat needed to cure the casting is generated
internally, and comes from the chemical reaction between base and
catalyst. The casting solidifies at an elevated temperature when the
reaction "trips" and massive cross-linking occurs. THE TEMPERATURE

Castings that cure at higher temperatures exhibit greater shrinkage
than castings that cure at lower temperatures. Makes sense, because
they have further to go to get back to room temperature, and
coefficient of thermal expansion is a harsh mistress!

But, you say, that should only be true with different materials.
Surely a given material will always react the same way, release the
same amount of heat energy, and the castings will all cure at the
same temperature, n'est pas??

Mais non, mon ami! Yes, the energy released per unit of material is
always the same, but if the casting is very thin, a lot of that heat
will be dissipated into the mold and be unavailable to cure the
casting. Thin castings cure at lower temperatures and as a result
exhibit lower shrinkage. But they are not fully cured – i.e. not
fully cross-linked. (That's why a casting can be rock-hard, while
the flash surrounding it is still sticky.)

Which brings us to the second source of urethane casting shrinkage:
post-curing of incompletely cured castings, which I suspect is the
cause of our problems. The specs for fully-cured urethane are
derived from thick (from our point of view) slab castings, either
1/8" or 1/4" thick and large enough so that the mix reaches its
maximum attainable temperature when the reaction trips. The
percentage of heat dissipated into the mold is relatively small, so
the resulting samples are fully cured and fully cross-linked. A
freight car side around 0.040" thick has few geometrical
similarities with such a slab, even if it contains the same amount
of resin.

The problem comes because the urethane really wants to be fully
cured. Heating castings to flatten them will help, but if no
additional heat is supplied, the castings will continue to cure at
room temperature, even if it takes many (sometimes many, many)
months. This additional cure is really additional cross-linking, and
as this proceeds the molecules in the castings are pulled closer
together. And the castings continue to shrink. It shouldn't take too
much imagination to visualize an assembled resin freight car body:
slab roof, which is a pretty thick casting to begin with and is most
likely fully cured; two ends which are relatively small compared to
the roof; and two sides, still needing to shrink and will do so with
very noticeable results if they are only restrained at the edges.

One more factor needs to be considered, then we'll wrap up this
exposition. And that is Heat Deflection Temperature, or HDT. This is
the spec'd temperature at which cured urethane deforms, and it is
typically quoted in multiple ways. One might be: "135F (24 hrs RT
cure); 150F (24 hrs RT cure followed by 24 hrs @ 180F)". In this
case the shrinkage would be quoted as "0.001" per inch (RT cure
only); 0.003" per inch (post-cured)". Parts have to be supported
during post-cure so they don't deform.

There are a number of implications to all this. First, you know none
of our hobby resin kit providers are post-curing their castings in
any meaningful way. This is good, because it leaves room for us to
heat and flatten castings when necessary. In fact, it might be
useful to heat large thin castings for a couple hours at 150F or so
as a matter of course, to make post-assembly differential shrinkage
less of an issue. You'd do this before sizing the parts, because
post-curing drives castings to a higher state of cure and introduces
additional shrinkage. Secondly, you cannot repeatedly heat and
reheat castings to modify their shapes and expect them to respond
the same way every time. It's that "higher state of cure" thing
again. Third, Tim's suggestion of bonding true structural members to
the backside of thin castings is a good one. Properly done, the
castings may still want to shrink, but they won't be able to – at
least, not so much as to deform. Fourth, the above applies more to
flat cast parts than to one piece bodies. Yes, all of that can
affect one piece bodies to some extent, but more likely causes of
deformed one piece bodies are removing castings from molds too soon
(which can stretch parts irrevocably), or inadequate support for the
castings when they are shipped or stored. Keeping the HDT in mind,
you'd like the castings in a resin kit to be packed so as to survive
being left in the back of a UPS truck sitting on an asphalt parking
lot in El Centro, CA, over the 4th of July weekend. Think that's
ridiculous? You haven't been in my attic recently. I have all sorts
of styrene kits surviving nicely up there, but all my resin kits are
on shelves in my workshop.

Personally, I prefer thicker sides so deformation isn't an issue.
But urethane resin is expensive, and minimizing the amount used is
one way of controlling costs – so long as manufacturer and consumer
are aware of the problems.

Thanks for your attention!

Tom Madden

Join to automatically receive all group messages.