Copyright WWW.DBBP.COM | updated 12-March-2016

As the walls at the outer edges were the thinnest I figured I would pour the aluminium in the thicker center and let the air escape through 4 holes at the edges. Ahem, keep in mind this was the first try... I made some little dams of sand to keep the molten aluminium away from the wood, and after 2 hours in the oven the aluminium was ready to pour.

 
 

When melting aluminium all the stuff like paint, dirt etcetera tends to float to the surface, so this you scoop off and do not use. I have no pics of this or of the actual pouring as I was doing this alone and needed both hands.

 
 

After pouring aluminium in the center hole I was expecting to see aluminium coming up in the risers but this did not happen. That meant that the aluminium was already solidifying in the thin parts and thus blocking the way. I pored some aluminium in the 4 risers knowing there would probably be air trapped in there but what the hell....

 
 

All in all the cover looked pretty decent apart from the airbubble where the aluminium did not get. Back in the crucible!(melting pot) On the second try I made several large diameter risers to make sure the metal would get everywhere, and it did!

 
 

On the second attempt however I saw that the material in the areas where the thick 25 mm riser connected to the thin 4 mm wall the material around the edge had sunken in a bit. This is caused by the fact that aluminium shrinks quite a bit when coolng and as I should have known by now large differences in wall thickness should be avoided. ( I have been designing cast parts for more than 30 years..) Allthough we could have welded some material on the deep spots that would be cheating, so this one goes back in the crucible also. Tomorrow I will try for the third time, this time I think I know how I should do it!


Previous