pre-warping a plastic part in the tool<?>

[caffine]

I am curious if any of you have ever heard of someone pre-warping a part in the tool. For example if you have a part that you want to lay flat and the fasteners are in from the edges, and its supported around the perimeter, if you mold it pre-warped(bowed in) then it may be more likely to stay flat when the fasteners suck it down, tensioned by the perimeter support.

Have any of you seen this done? Think it would work?

 

[MikeHalloran]

I can imagine circumstances where I would camber a part, e.g. to equalize gasket pressure on a nonrigid part. Better to make the part stiff and flat, but sometimes you just have to do weird stuff.

For instance, I have warped a part to get it _out_ of a mold. This part had severe undercuts that would otherwise require a collapsible core, but I didn't have the budget for that. I put some back taper on the undercuts so the part would spread out as it was ejected. The first few parts in a run typically cracked or fractured, but once the mold was warmed up, the parts deformed, well, plasically. The molder put a rubber band around the hot parts to deform them back to the intended shape, and once they cooled, they were just fine.

There is a trick to getting away with stuff like that. Don't ask us; ask your molder. They absolutely _love_ doing "impossible" stuff.



Mike Halloran
Pembroke Pines, FL, USA

 

[deper]

Yes, you can "pre-warp" a part when you KNOW how it is going to warp. We have done this many times on parts with long flat side walls. We know the part is going to suck inward so we will intentionally design the mold with some outward crown to compensate for it.

 

[orbit03]

A lot of radiator end tanks are molded this way. The companies designing and building these tools generally have a large database built up over the years that they use to determine the correct amount of windage to put in the mold. In theory, you could use some of the output from a warpage analysis to assist with this, but my experience has been that warpage analysis is good for predicting trends with warpage but not so good with predicting the amount of warpage.



------------
Jason Williams
K Development, Inc.

 

[caffine]

Thank you for the responses. Very informative I will contact some of the companies mentioned.

Cheers,

Carl