What was the material? ABS? When the mold maker asked you to put a reveal around the perimeter, did you only make a groove in the CAD? Or did you put in a half-lap joint as part of the reveal. Typically a clamshell design will use a half-lap. That gets you the reveal, and also the alignment. The metal of the screws themselves are not used for alignment. You can in addition add heavy draft of about 5 degrees to the ends of the bosses where they meet from each half of the clamshell. The underside of the counterbore for the head of the screw will have a ring that will overlap the top of the boss the screw threads into. That is where you put your 5deg. I also like leaving about 20 thou or .5mm of gap between this mating area, that way the bosses will pull the whole structure tight by .5mm, and insure the outer perimeter mate at the reveal is completely pulled tight. The mating line will almost never be perfect, even with a reveal. Although your mismatch seemed to be very large. Typically is more like .005" to .010" of mismatch. (Unless you are someone like Apple where its a whole different world and they will make prototype tools to test and then make the final tools based on how the parts really shrunk, and not just what moldflow showed). One trick you can use is put a round on both sides of the reveal of about 20 thou. To do this though, the PL has to be on the tangent of the round. And if you have this in your CAD, the toolmaker will most likely ignore it. You would have to check their ppt and the tool CAD before you give the go-ahead. The Chinese toolmaker never read the 2D in my experience, they only look at the 3D. I'm not an expert, but I do have 15 years experience as an ME designing part for mass production injection tooling. I too am surprised the toolmaker made a family tool of those mismatched part sizes as the other commenter was. I guess your numbers are low enough it was not worth making more tools. That could potentially be your source of the mismatched parting lines on the clamshell pieces, but I'm not certain of that.
Edit, I saw your first video after watching this one. It looks like you did put in the half lap. My apologies I couldn't tell from this video. And it looks like the bosses are too close to the wall to do that alignment feature I was mentioning. I'd probably have to send you a cross section of it, my text description is hard to follow. Nice work.
Thanks! This is exactly the type of input and feedback I was looking to get when I started this channel. I'm certainly not an expert, so I welcome any input that will allow me to learn. Thanks for the information. I had to look up what you mean by half-lap.
Hi John, first of all great work, I really enjoy watching your videos. I was wondering if you would consider selling the controller cases. I am also working on some sort of wireless controller for Marklin system but have not even started considering any physical design yet.
Looks pretty nice. Are they going to be able to run the parts simultaneously or are they going to have to block off and run them individually? If they are run individually the molder can tweak the size a bit via molding parameters (at the expense of cycle time).
I don't know what their plans are. However, I did ask for them to include shut offs so we could make extra battery doors. That way they can sell customers replacement parts in case they get broken.
Great Video! I’m curious, did you push the molder to make a family mold that included the smaller two parts? Typically I always get pushback on family molding anything but two very similar size parts. Using Icomold?
Thanks. No, we didn't have to push for a family mold. They suggested it as a way to reduce the cost. And no, it's not Icomold. We're working with a small company (a friend of a friend) that doesn't have a web site. But they know the model train industry well.
That's a good question. I don't know exactly, but I believe the company in Cleveland, OH will be using a 50 ton machine. I don't know what they have in China, where the test shots were done.
Thanks. I think the mold was about $13K, but I don't remember for sure--I'm not the one paying for the mold. That's a lot less than such a mold would cost to have made here in the US. We're working through an injection molding company in Ohio. The mold will be shipped to them for production. It turns out that injection molding itself is comparable here when you include the cost of shipping. We wanted to have the mold made here as well, but the estimates we got were closer to $50K.
