I am interested if anyone out there has attempted (successfully or unsuccessfully) to re-form injection molded parts by re-heating and then compressing or re-forming, after the parts have normalized/stabilized. I am aware of injection-compression techniques, however these do not eliminate post-molding dimensional changes due to crosslinking, moisture absorbtion, stress-relieving etc. My quest is to attempt to remove/reduce warpage on flat parts (PMMA, PA etc.) and to “re-size” parts to 3D accuracies and surface finishes beyond what can be achieved out of the mold.

I seem to remember that the old Sheller-Globe plant in Mora MN did something like this back in the 80's, particuliarly on parts that they were doing for the window industry. I don't have exact details, but the things I do remember was the reheating was done by cooling then reheating of the part inside the mold. There wasn't any big improvements in the part tolerances, and in some cases, the reheating was actually problematic due to the skin heating up and the inner wall remaining cold.

Thank you Barbara,
I did manage to re-flow the outer diameter surface of automotive idler pulleys molded out of 50% glass filled PA to yield a smoother finish and significantly reduce radial run-out. I did this by rotational driving of the pulleys with a heated steel roller, which produce a tiny "wave" of nylon that re-sized to part. I do know from lessons learned that crystalline polymers do not take kindly to this process, fortunately the parts were amorphous.
Do you recall what the polymer was that they tried, and did they re-heat with high-frequency or induction and then use a separate mold for the re-sizing?

I don't believe that they used a different mold - the cost was prohibitive. I think that the windows frames were vinyl, but frankly that was when I was a very young <wink, wink> molder. Their aim was mostly to do a stress relief.

Thanks.
I have the technical folk at Sabic scratching their heads as to if this process could also be used to totally seal specific holes in electrical connector boxes by surrounding the holes with a volcano-type protruding annular feature, and then melting-twisting it back to close up.
Anyway, thanks once again for the imput.

Robert:

Have you looked at the RocTool technology? It has some very interesting results by molding to a hot mold. There are improvements to surface finish, and I believe dimensional stability. Talk to Andy May at Sabic, he is fimiliar with the RocTool technology.

All the best,

Dennis L. Hayford

Robert,
this topic brings back memories, I worked on a developement project where blanks were molded, and than basically compression molded. the materails at the time were amorphous. the reason for the process was two fold, optically flat and stress free. understand that this was 1975. I was successful for the application.
If you look into the RocTool understand that there is a German university with same process whom was giving it away. think was written up in Jan 2009 MPW and exhibited at 2008 tool show in Frankfurt.

It sound like a really interesting project. Nano structures in the surface may depend a lot on the material. When the surface structure has to be replicated the elongation viscosity is very important but for most materials the available data is shear viscosity. It can be compared to a package of spaghetti (un-boiled). It can easily slide (shear) through an opening, but if you try to push them sideways through the opening, then they will break. Something similar happens on a molecular scale when replicating micro and nano structures. Unfortunately there are not a lot of data for this, but I know that Professor Ole Hassager from the department of chemical engineering at Technical University of Denmark has done research in this field.