Rare cosmetic defect

[Horeki]

Hi everyone!

Im ruling right now a very large thin wall IML project, the process is stable right now with scrap rates of 0.75%, and everything it's going well, but after a few days of non stop 24/7 production(just stoping every 50k shots for periodical inspection check) I've noticed that a rare cosmetic defect started to apear in some pieces, the piece its not considered scrap since the defect its located on the base of the cup and it's not noticeable from the inside (client approved).

The curious thing is that this defect only appears after more than 100hrs of running non stop, and as sudden as it appear it disappears too and that's why QA just noticed this "little" defect (after more than 2 months of work).

To me it appears like gas or air trapped inside, but it doesn't bulk or adds bumps to the piece.

Anyway, I wonder if anyone has seen something similar to this?

IMG 1 (https://drive.google.com/file/d/1bRAJ7Md_Npu9k8dcHXNdZWifhPI7OdJR/view?usp=sharing)
IMG 2 (https://drive.google.com/file/d/14WAGcsYroac85wxiRKP40MpgQmtJVxi6/view?usp=sharing)
IMG 3 (https://drive.google.com/file/d/1zMSaffnv9LL3Spkkz7_jrzg45zF7ho3M/view?usp=sharing)
IMG 4 (https://drive.google.com/file/d/1ZXyQgcDFxErZ7Kk5Fn3jdPNFXXBvlmVj/view?usp=sharing)

It's only visible from inside if seen against light.

The "failure rate" reported by QA its juts of 0.0048% (7 out of 145k).

[RobertPHale]

Horeki,

It looks somewhat like a voids are forming in the base of that cup and the layers of plastic are separating from each other. You said it only occurs after running non-stop for 100 hours. Is this a hard number or could it be happening more often than that but its hard to catch? Is it every cavity or just some of the cavities (assuming a multi cavity tool). I would be looking at the tool temperature difference between the core and cavity when you start seeing the defect to see if there is a possibility that one side of the tool is hotter than the other. I would also measure the temperatures of the core and cavity lets say after the press is up and running for an hour or two and then again when the defect occurs to see if the tool temperature on one side or both sides are changing.

Sticking with water, I would make sure consistently have turbulent flow through the channels and that flow isn't being restricted anywhere.

Hopefully this helps.

-Robert

[rickbatey]

To me it looks like either poor color dispersion or unmelted pellets. I would watch the screw recovery pressure at start and then check it every half hour. I think you’ll find it low at the start and then increasing as it runs proving throughput is out pacing temperature settings on the barrel. Then just increase heats slightly across the board until the screw drive pressure drops back down to where it started.
Rick.

[joeprocess]

Unmelted resin or contaminated with some higher melt resin maybe?