Sand casting Caterpillar track links in aluminium

As a newbie to metal casting I’m hoping for some help from more experienced members with a project that I’m a bit stuck with. I’ve added some photos of a 3D printed caterpillar track link which I need to cast in aluminium, about 120 needed in total. I think that the radiused ends mean that the split line will have to run through the centre of the end radii as shown by the green line in the end profile photo. However, that will leave the concave mating radii (blue line) as an undercut. Overall size of the part is 97mm x 40mm x 20mm, (approximately 3 3/4” x 1 1/2” x 3/4”)
Any ideas on how to get around the problem would be gratefully received.

 

What I see that complicates things is the area between the trunnions.
The rest of it should pull with proper draft on the verticals using the split line you drew.

 

Yes your right Fishbonz, the problem areas are on both sides of the part, I just need a way to sort the problem. From what I’ve read, the answer might be with loose piece inserts, just not sure yet how to tackle it. I’m hoping someone on the forum will be able to help.

 

Looks just like whats on our airplane towing robot we have here. Is that what this is for?

 

The first thing that struck me in terms of sand casting is all of the sharp internal corners and squared edges. Fillets and draft angles would have to be applied to the patterns which is probably not a big deal, but all of the machining after casting absolutely would be a big deal.

Pete

 

The track is an "elegant" design, but must it be so intricate? If it could be a simpler cruder design it would be far easier to cast and post-machine. I am seeing all kinds of shrink defects at all those sharp inside corners of this somewhat "chunky" part.

Also, what keeps the track on track as it seems tracks segments usually have soem sort of guides to keep them on the drive, idler and rollers?

As a high school kid and college student I used to work summers for Link-Belt assembling and testing crawler cranes. Here is an example of their track shoes. Note the simplicity of the design and the ribs used to keep them on the rollers and drive and idler.



Denis

 

It would definitely be worth creating a special fixture for machining that many pieces..

 

Wow, isn't that a lot of filament and printer time (said the guy who does not print)---100 shoes 1.4 x 4 x 3/4 times 100?

Denis

 

What a waste time! But... educational.

If I ever get you to do a 3-D print. Please give me a cost per hour. You should get paid for it! Or at least a kickback.

Edit; I do have one for you but I want to try it first.

 

Exactly that Jason.

 

Bear in mind that this is not the finished part, just proof of concept. When the main problems are solved, I’ll attend to the details, such as draft angles, radii etc.

 

Not sure that I’ll live long enough to do lost Pla on 120 parts. If I remember right, it took about 4 hours to lay down a pair in pla.

 

I must have looked at dozens of track designs before coming up with this one, it’s a mix of the best bits of several existing track links (best in terms of the application). The 5mm thick top on the finished part will actually be rubber and the drive, intermediate and idler cogs will be in pairs that have big fat teeth to locate in the gaps between the link rollers (or trunnions as Jason correctly called them). If I can’t find an answer to the problems with this one, then a redesign will be needed. But I haven’t given up on this yet.
Jim

 

I agree Jason, a 3D printed machining jig is planned to gun drill the parts on my X3 mill. I might try making cores first, but 6mm diameter x 100+ mm long might be a long shot, but worth a try. I’m also looking at a sliding jig to square up the draft angle on the trunnions. Something along the lines of finger jointing in woodwork.
Jim

 

David, I really don’t fancy the lost pla route, but I have a fair bit of experience in setting up to process parts in a production environment. With the correct jigs and tooling, each link could be machined and fettled in about 1.5 minutes, allowing for tea breaks, call it 30 an hour. If the cores don’t work out and the links need gun drilling; that would double the machining time, a days work in all.
Jim

 

The only thing I can see is a three piece pattern with the middle piece being a core.
Mold the bottom piece in the drag normally. The pad holder is in the cope with a follower to get the spacing for the core.
Whew, that's a lot of molding.

 

I hope this makes it all a bit clearer

 

Hi Fishbonz,
I like the idea of a three part pattern, just not sure how to make it work. Any chance of a rough sketch to get me started?