Home > Blog > Build Logs > Portal Sentry > Fiberglasstastic


Posted by qwindelzorf on April 15, 2011

Today was the (first) day of fiberglass.  We made some progress, but there is still much to do.  We managed to create both of the pod shells, along with the eye.  We als managed to utterly fail at creating the body shell, not once, but twice.

IMG_20110415_105502.jpg IMG_20110415_105524.jpg

The day started with setup.  We got the last coat of mold release applied to the molds, and cut out the first several rounds-worth of fiberglass cloth.  For this project, we are using 4 oz. fiberglass, with 3M resin.

IMG_20110415_113154.jpg IMG_20110415_115513.jpg

Our first mold of the day was to be the pod shell.  We decided early on that two layers of fiberglass would probably be plenty for this project, as the shells don't have to be all that rigid.  Rigidity will be provided by the internal structure. 

The prodedure we are using is to first pre-apply the resin to the cloth (laid out on a sheet of plastic), and then move the cloth into the mold.  So, we laid out the first sheet of fiberglass, and mixed up the first batch of resin. We poured half the resin (4 oz.) over the cloth and spread it around with scrapers.  Once the cloth was soaked, we piced it up and moved it to the mold. 

Now, the resin we are using nominally has a 15 minute work life.  After that, it begins to get hard.  We estimated that we could put down a layer in 7 minutes, so we should be able to get two layers per batch of resin.  This proved not to be the case.  By minute 8, we had finished installing the first layer, and the resin had begun to get hard.  So, we quickly went to apply the rest of our resin (still in the mixing cup) to the other piece of fiberglass.  I poured the rest of the resin, and nothing happened.  Turns out that this resin is exothermic, and heat causes it to harden faster.  The upshot of this being tha the more of it you have in one place, the faster it hardens.  So, at 8 minutes into our first batch, what was left in the cup had turned into a solid puck of resin that was hot enough that the mixing cup was starting to melt.  

I quickly popped out the puck, and left it to cool on the cement.

That's how we discovered that we could really only count on 10 minutes (at most) of working time per batch of resin.


Two hours later, the resin was fully cured, and this is what we had.  One pod shell, complete, and acceptable. There are afew bubbles, and some spots that will have to be filled with bondo, but overall a success.

During that two-hour curing time, we decided to do one of the body molds.  We had already cut out the fiberglass pieces we would need, so it was just a matter of mixing up the resin and applying the fiberglass.  So we did.


This hideousness was the result.  We attempted to put in two pieces, covering the two largest areas of the mold.  Both pieces came out badly.

IMG_20110415_134703.jpg IMG_20110415_134712.jpg

The top piece (left image) completely failed to stick to the mold at all, and ended up developing a very large bubble while we madly tried to smooth it out.  The lower piece (right image) stuck a little better, but had numerous problems conforming to the stair-step geometry, and also wound up bubbling.  By the time it cured, we had determined that the upper half was a total loss, and the lower would have to be massively trimmed, but might be salvageable. 


After that bout of failure, we went ahead and did up the eye mold.  That came out well... eventually. Because it is such a deep and small piece, it proved incredibly difficult to de-mould.  In fact, while trying to get the shell off, the base for the mold popped off, leaving us with a hat-shaped object filled with blue foam.  

I ended up drilling a big hole in the middle of the foam, then spending the next 45 minutes slowly cutting out all the foam with a knife.  RedPeasant spent that time re-prepping the big molds for the next pod and another attempt at the body.


The second pod shell came out beautifully.  Nicer even than the first. The second attempt at the body shell was not so good. This time, our problem was not so mucn with getting the fiberglass to stick, but rather that the resin began to set up far more quickly than we expected.  I suspect that I may have added too much hardener when mixing up the batch.  It ended up turning into big, nasty brownish chunks that utterly destroyed any attempt to produce a smooth surface.  

So, tomorrow is a new day, and will start with re-prepping the body mold.  This time around, we will be trying to put in lots of small strips in order to more easily follow the complex internal geometry of the piece.  We will also be doing the knee shells tomorrow.  

More to come.