Lessons from Another Casting Experiment

This looks a good bit worse than the last one, doesn't it?

DSC_0093.JPG

The excess metal, called "flash," is the result of cracks in the mould (or from a mould seam, as the term is usually applied). As I had feared, even a small mould cannot be heated too quickly. Therefore, I'll be stuck babysitting my non-programmable kiln and ramping the temperature manually, increasing it by 40 degrees every ten minutes, leaving it alone only during the two-hour hold periods. That's what I did the first two times, and it worked.

DSC_0094.JPG

Luckily, at least the new process for making the mould seems to be an improvement. There were far fewer bubbles, owing to the longer de-gassing cycle.

DSC_0095.JPG

The line running up the side of the cup is from a very shallow surface crack in the mould. Honestly, this was the result I was expecting the first time, when I used ordinary plaster to make the mould (and ramped up the temperature the way I was supposed to). Oh well, live and learn!

DSC_0096.JPG

Thanks to the new orientation of the cup, the bottom isn't filled with metal. There is one moderately-sized bubble in the bottom (no idea why that's there), and the only other complaint that I have is that I'm still getting voids in the walls. This time, it's considerably more noticeable.

DSC_0097.JPG

The mushroom, meanwhile, came out in two pieces, and a lot of it is missing. I have no idea why, but I'm going to attribute it to the improper burnout cycle. I think I'm going to buy another flask (I can't make one at the moment, but I can once I'm back in my machine shop) so that I can have more than one mould in the works at a time. I can still have multiple part trees under construction, of course. The next part tree will probably be another batch of cups, with a different configuration and a slag trap. If the slag trap works as intended, I'll put one on every future part tree.

Despite my extensive video documentation of these experiments, I don't think I'm going to publish any footage of failed casting runs. After all, my intent here is to figure out on my own what I can get away with before trying to teach anyone else how to do this. Besides, if someone were to ask me "what happens if I don't do [insert procedure here]," I'd like to have an answer.

H2
H3
H4
3 columns
2 columns
1 column
Join the conversation now
Logo
Center