The Troubleshooter, Part 18: Dimples in overmolded TPR

This article continues our series of troubleshooting reports from one of the leading on-the-spot problem solvers in the molding industry. Bob Hatch is manager of technical service and customer support for Prime Alliance, the Des Moines-based resin distributor. Before his present assignment, Bob managed a molding oeration for 25 years.

A customer called with a complaint about dimples at the gates of a glass-filled polypropylene roller overmolded with a thermoplastic rubber material. The problem was with the gates on the overmolded part. I knew the part well. I had helped the company with the design of the two molds several years ago. We had used a glass-filled polypropylene to mold the hub and a 40 shore A TPR to mold a tread on the circumference of the hub in a second mold. Both molds were eight cavities and each had a balanced runner feeding material into the cavities (Figures 2 and 3).

We had used a ring gate on the hub mold to keep the parts as round as possible, and it worked well. For the tread we used a horseshoe runner to go around the outside of the hub and gate into the hub from two points that were opposite each other.

Everything worked well except for the gates for the TPR part of the molding. The size of the subgate we had to use for volume considerations was causing a chunk of the material to be pulled out with the gate when the part ejected. So we changed the single subgate of .125 inch to two subgates, side by side, of .090 inch each.

It turned out to be a blessing in disguise for us after all, because the two side-by-side subgates actually did a better job of filling and packing than the single subgate, plus the smaller gates didn't pull a chunk of the material out with them.

What Had Changed?

We had two molds that have been running well for four or five years, and now the customer says he is getting a sink hole or dimple at the gate areas in the tread portion of the part. I reviewed both molds to see if anything had changed since I last worked with them. I looked to see if the company was using a nozzle with an orifice that was smaller than it should be, which would change the machine's ability to fill and pack the parts correctly. But the molder had drilled out a nozzle to use for each mold, which were still being used, so that ruled out the nozzle as a problem.

Next I looked to see if the sprue bushings had been changed and they had not. I looked to see if the gates and runners were different than before and they were still like we left them. I also wanted to be sure some overzealous toolmaker hadn't tried to reduce the size of the runners or gates to keep from generating extra regrind or to keep from slowing down the cycle. Nothing had been changed. Then I looked at the runner and part venting to see if the vents had been hobbed shut or were full of material residue. Nope, we had made the vents self-cleaning originally and they were still working great.

What was left to look at, the material? The molder was still using the same original materials, and since we had optimized the tooling, we could pretty much be assured the slight differences in lot-to-lot variations of the material would not cause us much of a problem in filling and packing either part.

All that was left now was to look at the setup sheets to double check the processing conditions to see if any changes had been made. The heats, speeds, pressures, mold temps, backpressure, and screw rpm were all pretty close to what they should have been.

Next I went out to the molding machine and watched the mold that puts the TPR treads on the hub insert. I like to watch the screw go forward and visualize the material being injected into the mold and see what is going on with the cushion and whether or not it holds properly.

The Slipping Cushion

I like to use a medium to medium/slow injection speed for thermoplastic rubbers, even urethanes for that matter, and this was still a slow injection speed, which is good. The cushion was about .25 inch, which was what we wanted, but during the hold pressure portion of the injection sequence, I noticed the cushion slipped to about .125 inch. Something was wrong there. Either the check valve was leaking or extra material was being pushed through the gate after the inject, pack, and hold pressure sequence was finished.

I like to use the hold pressure in this manner to fill voids as they form on thick-walled parts, but this wasn't a thick-walled part. It had some thick to thin transitions but not the traditional thick walls and definitely not at the gate.

I turned the hold pressure down from 400 psi to 200 psi, and the dimples at the gates went away. We watched several cycles and the problem did not return. I went back to the setup sheet to see why I had not noticed the problem there. Then it hit me; we are used to seeing the hold pressure being about half of the inject pressure. But in this case the hold pressure was originally set at about a quarter of the inject pressure.

No doubt one of the molding technicians had accidentally raised the hold pressure, thinking it was too low, and caused the dimples to form at the gates. The dimples probably didn't show up until the next full shot and by then he had walked away to get a cup of coffee. The operator probably didn't notice the problem until the quality control person spotted the dimples, so the molding technician did not relate the dimples to his change in the hold pressure.

It is pretty easy to miss a problem like this by just looking at the setup sheet. You have to watch the machine run to get a feel for the plastic injecting into the mold, check the cushion and see what it does, watch the screw recover, and then watch it all over again.

From what I've seen, this problem pretty much shows up only on the soft materials, such as flexible PVC, thermoplastic rubbers, thermoplastic elastomers, and the softer polyurethanes. The customer made the changes to his setup sheet, threw the old ones away, and vowed to use only master copies of the setup sheets so this kind of problem would not bother him again.