Pattern making made me som headace. I will try to cast a housing for a km/h speedometer for a motorcycle. I figured out thanks to all good videos posted by myfordboy and others here how to do the pattern. But i can not figure out how to do the cores. How do you think when you solve pattern making. Do i need multiple cores, how do i do it or is it impossible. Does anyone have som links to a description or video how to do this kind of casting? Thanks in advance. Ole Thronborg Sweden
It should be possible to make that as two halves for the outside and one core for the inside so three pieces to assemble to make the complete sand mould. The core would be made by a lengthways split mould with baked linseed oil or a suitable sand binder resin. The lugs would have to be modified by widening to allow core removal from the mould to eliminate the undercut or even made into a single circular ring/flange with the mounting holes drilled in it.
Thank you very much Mark. And specially thanks for the explaning drawings. But i dont now what you mean "The lugs would have to be modified by widening". Sorry i am not so good at the english language. Thanks, Ole
My explanation is not that good but if the core is made by a two piece split mould and there are four lugs? like your drawings, there would be sand caught by the pattern unable to release.
Looking mainly at the main cavity and with respect to the (I think three) lugs, couldn't the core simply be a curved-sided cone with three recesses for the lugs? Split the cope and drag as indicated in the drawing. Use core prints as indicated in Mark’s drawing. But make (mold or machine) lug recesses in the core of any desired shape. I guess I do not see why “sand entrapment” in the core would be a consideration. Maybe I have not had enough coffee yet...;-) Denis Added: with respect to the smallish cable channel, it appears it could be molded using a thin (very delicate) side arm piece on the main core. It would probably benefit from an aluminum tube reinforcing rod. And the main core could have a socket to plug the one end of the small core into the main core. But for one-off, I think it might be easier to just drill out that 1/2” or so diameter channel starting one hole from inside the housing and the short segment from below so that the holes intersect. By eye, it appears there is clearance to drill both portions of that channel.
For a core like this, there's a way around that, use loose pieces for the lugs, attach the lugs to the core box with screws accessible from the outside of the core box, after packing sand into the core box remove the screws so that the lugs stay in the core when removed from the box, once the core is freed of the box, remove the lugs from the core. Myfordboy uses a somewhat similar technique in his video: "Metal casting at home part 3".
To make those lugs isn’t the core more or less like the nose of a rocket with 3 or 4 recesses not outcroppings? Denis
Perhaps ensure the core(s) have adequate friability? (They break down readily after casting). Perhaps run this in Petrobond, and use a green sand core? (Ram it *really* hard, using small dowels or nails as rammers?) Use a four-part mold for the core?
I would use what I am familiar with—-sodium silicate. It breaks out very easily if properly mixed. This is especially true for aluminum. Green sand core could work if you stood the flask vertically. One part core. And yes core mold would need to have multi parts—-two main parts and 3 or 4 loose pcs. Ignoring the side arm portion but same idea:
I agree, there are a few advantages with silicate cores, unlike molasses sand cores, they can be cured without heat while still in the core box and unlike a green sand core, silicate cores are quite robust and won't get damaged with normal handling and yet when used with aluminum, they are easy to remove and any stubborn residue can be flushed out with water. Here's an old video of one of my cores being made. This video was shot in real time with no cuts or edits so the length of the video is exactly how long it took to prepare the core.
Is that regular plaster (in the core box) - or a special kind, e.g. hydrocal or similar? I recall making (plug) molds for fiberglass out of plaster with some *sculptamold* as a backer, to make the molds less brittle.
That's core box is just plain old hardware store plaster of paris. The core is symmetrical about the split line so I made a half pattern of the core out of plywood and bondo and used that to cast the two core box halves. I only intended to make 3 or 4 cores from that core box so I wasn't too concerned about durabilty
Thanks Mark. Now i understand, you are totally right i did not thought about that. It is not easy to think inverting. I put on a 16 hour print on my 3D printer already with the wrong core. I hope i will learn.
That's true, but if a two piece core pattern is made the core pattern catches on the sand where the lugs are kind of like pulling a dovetail joint apart. As Dennis mentioned a four piece core pattern would fix this or removable lugs from a one piece pattern would also work.
If your core mould divides into four sections it would still work as Dennis mentioned. It may be possible to save the print by carefully cutting into four pieces.
You are correct. I shoulda had another cup of coffee before hitting the keyboard. Loose pieces are probably the way I’d go to make the lug the OP shows. Just a single screw would hold the piece in place while the core was cast. Denis
If possible you could add draft to the bottom of the lugs and draw it with no cores.... just have to drill the holes.