eMachineShop.com Forum
Join in our community discussions. Anyone can browse our forums, but you will need to register a user account and login to post replies and original forum threads.
|
| Author |
Message |
| twixter
|
|
Fri Jan 30, 2009 6:35 pm Injection molding |
|
|
I have designed a piece for a board game which stacks with copies of itself like cannonballs do. The ems file is attached. The piece needs to be made accurately enough that it is easy to place and remove tiles. I want to make several small runs using a variety of options before I make my final choice for the main run.
The material should be smooth, non porous, pleasant to handle, but not shiny. It should be rigid, not flexible, or not very flexible anyway. Cost is a major concern, but secondary to accuracy. I need two colors, black and white. Polystyrene for molding looks to be the least expensive. Would that be a good choice?
I used a draft of 92 degrees for the side walls. The shape could be described as two 2-D polygons, the smaller placed on top of the larger. These polygons share an edge at three places. I set these three top edges back with the idea that the resulting sloped edges of the bottom would be flush with the edges of the top. I used the formula (offset distance)=(bottom shape thickness)x(sin(2 degrees)). I specified the coordinates with the full 5 digits of accuracy available. The resulting 3D image seems to have a break between these two surfaces, not flush. Does the 3D image show the draft?
If you see any way I could improve my design, either by reducing cost or something else, please let me know. I considered cutting the two shapes with a laser and attaching them, but I don't know any way to do that accurately enough.
Speaking of which, how accurate is injection molding? And how do I specify where the parting line, ejection pins, and gate should go?
Thanks for your patience!
| Description: |
| Lazo tile shape. The board (not shown) will have holes the same shape as the Y-shaped "pegs" which are actually on the bottom surface of the tile, although here they are labeled as on the top surface because of the draft. |
|
 Download |
| Filename: |
Lazotile.ems |
| Filesize: |
21.4 KB |
| Downloaded: |
27 Time(s) |
|
|
|
|
 |
| twixter
|
|
Fri Jan 30, 2009 7:13 pm |
|
|
Uh- that should be tangent not sine, sorry. But the 3-D image still shows a very slight bump where the two surfaces should be flush.
| Description: |
| using tangent instead of sine |
|
 Download |
| Filename: |
Lazotile.ems |
| Filesize: |
21.16 KB |
| Downloaded: |
19 Time(s) |
|
|
|
|
 |
| Tech2
|
|
Sat Jan 31, 2009 7:50 am |
|
|
Thanks for your informative post.
Please see: Plastic Injection Molding
Most of your questions are addressed at that page and it's sub pages.
About tolerances please see: Tolerances
Polystyrene is a good candidate. You can see typical applications of plastics at: Materials
As for the apparent break in the 3D view, zooming in on the 2D side view shows no problem so you are OK.
You specify parting line, ejection pins, and gate using Comments To Machinist.
|
|
|
|
 |
| twixter
|
|
Sat Jan 31, 2009 11:13 am |
|
|
Thanks! Wherever the parting line is, the draft should be different on each side of it, right? For example, if I put the parting line where the Y "peg" shape meets the larger tile shape, which is a concave recess for most of the perimeter except where the two 2-D shapes share an edge, the draft for the Y shape should go opposite from the draft for the rest of the tile, right? The attached file shows what I mean. Now I have the Y shape on the BACK of the main shape, and I don't need to calculate an offset for the shared edges any more. Using the same frame of reference, the draft for the main shape is 92 degrees and the draft for the Y shape is 88 degrees. Would there be any problem specifying the parting line in a concave recess?
The page you referenced about injection molding suggests that I could define more than one piece in a single mold, connected by small bars. The parting line must lie on these bars, right? Would each piece in the mold need its own gate? If there is only one gate, what is the minimum diameter for the bars?
I can experiment with different numbers of tiles per mold to find a minimal cost for my total order, but if I knew what sort of formula the pricing algorithm uses, This would help to narrow down the range to experiment on. Any information about this formula would be greatly appreciated!
David
| Description: |
|
 Download |
| Filename: |
LazotileNoOffset.ems |
| Filesize: |
21.4 KB |
| Downloaded: |
20 Time(s) |
|
|
|
|
 |
| Tech2
|
|
Sat Jan 31, 2009 11:58 am |
|
|
|
About the parting line you are correct and the concave area is OK. You should not create the copies yourself for identical parts - just experiment with the repeat count in the UI.
|
|
|
|
 |
|
|
|
"When a bracket of my shower door broke, I was looking at a complete enclosure replacement, north of six hundred bucks, since the model I had was not manufactured anymore. In desperation, I designed a replacement bracket using my qCAD software. When I was looking for somebody to machine it, I found your company on the Internet and downloaded your ..."
- Ekkehard Blanz
|