Thermoformed ABS defects? 1

[DisappointedDad]

Our crew recently ordered some thermoformed ABS two-part, open cast molds for some rough prototyping with a urethane material but ran into some strange issues. It appears that the ABS has small hairline cracks running up and down the part in corners and radii of the geometry. This is usually seen when trying to cast something with the mold as it appears to seep into these hairline cracks. Additionally, it seems that the ABS becomes very brittle after the curing process. Now my initial assumption is that the vendor did not process the ABS correctly as it should not be brittle/cracking. Has anyone else witness such madness?

Some processing notes: The molds, when used, are placed in an oven at ~150F to cure. We have tried this with ABS/PVC blends, PC, and other plastics with good results, so this is a new one for us. Especially ABS as we have tried 3d printed ABS to good effect in the past as well. Is it possible we are annealing the plastic at those temps, thus causing some stress relief or something from the thermoforming process? Or is it purely a improperly processed part?

 

[3DDave]

ABS is a family of plastics. Perhaps it's a bad formulation for this job.

 

[DisappointedDad]

ABS is a family of plastics. Perhaps it's a bad formulation for this job.

Yeah I guess that's the underlying question, whether or not its a process or material issue. The vendor says it can be "mitigated with proper tooling" but its difficult to justify that type of investment if we can't get it to work at the current level.

 

[btrueblood]

It sounds like the abs used was either stretched/formed at too low a temperature, or a too-deep draw or too tight a bend radius, etc. etc. Can you share details of the mold design used to form the sheet? What was sheet starting thickness, and what is the final wall thickness in the area of the cracks?

Has the vendor suggested any rework ideas, or have you tried to fill the cracks with ABS/solvent cement or other materials? What mold release are you using, and does it have solvents as part of its formulation, and what solvents are those? Are you seeing adhesion of the PU to the mold surface in areas without visible cracks?

 

[DisappointedDad]

That's what I suspect, but can't seem to get the vendor to agree to the issue(CYA issues likely). The part has roughly a 5in depth over a 10in width. Running sheets at 0.25in stock thickness. We have measured the thinning down to only 0.2in at the deepest regions. It is performed on male tooling, with the interior surface being the "mold surface".

The vendor has only suggested we get aluminum tooling, which makes sense to me(it is currently tooling foam), but can't justify until we can get through this prototyping phase. They "have no idea" why such things would occur. Either they are withholding intel in a false fear of some sort of retribution, or they genuinely don't have the institutional knowledge (which I suspect). No comments on change to geometry or radii.

I will add a few things. As mentioned in the original post, we had other plastics perform quite well. There doesn't seem to be a geometry issue in the forming of the parts as they all look identical. However, the cracks seem to only exist on the regular ABS parts. Furthermore, the cracks seem to always initiate from the interior surface, or the surface in contact with the tooling. From the external side, there are no visible indicators of said issue, and when a crack fully ruptures, it is very clear it is from the inside out. This leads to support a tooling temp issue, but just unsure if this was common or not. It bothers me that these folks popping these out would "have no idea", when it seems like it may be just a common processing issue. Perhaps they cannot meet that with the given tooling, but they have yet to fully correlate the two.

Finally, the mold release is a general purpose urethane mold release. I suspect it is a silicone oil of some type and haven't seen any issues with ABS compatibility. The parts do not stick where cracks do not appear. They only "stick" because there are small "veins" of material that flows into these cracks during the cast/cure process it seems. We haven't tried to fill any cracks as we only see them once they are used. Might have to try a slurry if I can get my hands on another test subject.

 

[btrueblood]

Check the can of mold release, the stuff we have here (CRC Silicone spray) has toluene in the formulation, I guess to help thin the oil and make it aerosol-ize. Might try hand-applying silicone oil or grease. Yes, foam tools are not ideal, they usually don't get preheated and so you can have the part's inside wall starting to "freeze" while the outside wall is still hot enough to stretch. The usual result is a lot of warping, especially if the part sees a temperature rise (we had a thermoformed abs/pvc tray that got subjected to hot water and warped horribly...the fix was a change to aluminum preheated tool, and a post-forming annealing process). So, inner wall freezes, outer wall gets stretched in tension, and the part cools - the shrinkage of the outer wall then puts inside wall into tension leading to warping, or in your case cracking. The reduction doesn't sound that severe, and as you say you aren't having problems with other materials - were those molds done on other tooling?

 

[DisappointedDad]

btrueblood, thank you! That is the kind of insight I was hoping to get from the vendor as I am typically far removed from the polymer game. I will check the mold release as well for those types of compatibility issues. The other molds were done on the same tooling but with different materials. I can only suspect that the vendor is more comfortable with some of those materials, or they are better to form? I am not too sure. We did some in PC, and ABS/PVC, and the latter came out the best with the foam tooling. Difficult to speculate when I am not sure if they were cutting corners or if it is just a poor material/tooling choice. Anyways thanks for the insight, will update if I can figure anything out, whether it be mold release, material, or tooling!

 

[btrueblood]

Yeah, abs/pvc is some of the best stuff for thermoforming, gives advantages of both materials. If surface finish isn't very important, you might try straight PVC, as it's stable against a lot of chemicals. ABS alone is too sensitive to a lot of solvents, even trace amounts can give you crazing/cracking at areas like radii where a lot of residual stress remains in the formed part. I'm surprised the PC is holding up, but as long as you are careful with what you let contact the surface, it might be ok (it is sensitive to a lot of solvents and even cleaner type chemicals). PC is also formed from a higher temperature, which might be helping here (less likely to fully freeze when it touches the male plug). Have him try PET film, it can be pretty stable, though you need to use pretty thin sheets to get good results. Another thing, are you doing RIM urethane (i.e. closed mold, foaming urethane)? The pressure rise in the closed molds could be opening up some hidden cracks. Oh, and baking the molds empty in a 150F oven might help anneal them and reduce the internal stresses a bit, no matter the material.

 

[DisappointedDad]

I guess in all fairness, the PC did not do well in the molding process, had a lot of crazing occur, but in contrast to the actual cracks seen in the ABS, I would have to call it an improvement. We are open mold vacuum casting a PU mix. It is not a RIM urethane process, but the mix is somewhat dense. I guess we inadvertently anneal them as they cure in an oven around that temp range. First ABS/PVC part we tried worked well, showed no cracks or defects, but it did bulge under the weight without some external structure, but what was fascinating is we used it a few times more and it never seemed to expand past where it did on the first cycle, which made us begin to ponder about annealing and the like. We gave the parts some more structure and now its working as intended with no obvious signs of defects and I think we will stick with that for now as it does seem to be working. I guess its just kind of frustrated the plastics folks who produced these parts couldn't share this kind of information (assuming they should know about it).

 

[rickfischer51]

I would be curious to see what the thermoformed abs parts look like, and where the cracks are on the part. tool design could make a difference. For instance, a 5 sided box can be made by sucking a sheet into a cavity, or sucking it down over a plug. Parts can be very different because the sheet begins to cool as soon as it touches the tool, so where the stretching occurs, how much material stays hot enough to draw, etc, can vary.

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[DisappointedDad]

I can't share images of the parts unfortunately but there is a picture of the cracks up above (I think). It is pulled down over a plug, over some epoxy tooling and the cracks started near the surface that touched the plug. We discussed with the vendor about this and they do agree the tooling sucks but were hesitant to really explain (not sure if they even know) how that tooling affects the parts.

Everything btrueblood and you mention makes sense to me and we are working with the vendor to see if they come to a similar consensus. Either way, if we go forward, we are going to get the proper tooling, perhaps check out other vendors,etc...

 

[rickfischer51]

Check out empirewest.com. They have a page that shows different thermoforming techniques. Take a look at the snap-back and billow methods. If you let the air do the stretching, the finished wall is usually thicker and more uniform in thickness and temperature. Once the plastic touches something solid, it starts cooling and friction can inhibit flow, so the stretching now occurs in a smaller volume of material, and it thins.

Rick Fischer
Principal Engineer
Argonne National Laboratory