The Troubleshooter: Part 95: Watch the subgates and the venting

Even without the material type or part prints available, the Troubleshooter shows how to cure the short shots.

What a week! Every day three or four packages showed up in my office for review, but I found one to write about that presented a good lesson in subgates and venting.

The part looked like an air or water filter. It was approximately 4 inches long and appeared to be molded out of a crystalline material—possibly nylon 6, polyethylene, or polypropylene. The deciding factor would be the operating temperature of the air or fluid in which this filter had to function; all three materials would be chemically resistant to most fluids and have a similar shrinkage factor (roughly 0.017 in/in).

(Note: It’s important to check with the material manufacturer in cases like this to be sure the shrinkage numbers are known prior to building the mold. I have seen molders get burned because they guessed incorrectly.)

I could see from the cold runner that the mold had four cavities, but with a twist: The gates going into each of the cavities had a split design—one that I refer to as a “snake tongue.” Each of these subgate diameters was 0.040 inch.

The reason a gate design like this is used is to keep the subgate diameter from getting too large. During ejection, a single, larger gate could break a chunk of material out of the part.

I measured the part walls and found the thickness of the bottom to be 0.090 inch, the sidewalls 0.070 inch, and the filter ribs 0.030 inch. The open spaces between the webbing ribs were 0.020 inch. The actual filter openings were too small for me to measure accurately but I estimated them to be about 0.015 inch.

Other dimensions I measured were the nozzle orifice (0.120 inch), the sprue O-diameter (only 0.140 inch), the main runner (0.150 inch), and the subrunners (0.120 inch). I could see trouble with these dimensions when trying to fill a thin-wall part. Packing pressure was not a concern, but venting and a fast injection speed were critical.

Tool modifications

The only defect I saw on the parts was short shots along the side, all in the filter portion of the webbing; I had no idea if the length of the part or the inside and outside dimensions were to print.When gating into a 0.090-inch-thick wall, use a subgate diameter that’s 50-60% of the part wall being gated into. I would continue to use the snake tongue subgate design, but I would increase the subgate diameters to 0.050 inch from 0.040 inch to help fill out the thin sections.

The 0.120-inch subrunner feeding each of the split gates was big enough to pressurize the melt adequately as it split at the gate. The 0.150-inch main runner diameter was also big enough to pressurize the two subrunners. However, the 0.140-inch sprue O-diameter wasn’t big enough to pressurize the main runner. The sprue bushing needed to be replaced with one that had a 0.175-inch sprue O-diameter. With the sprue O-diameter opened up, we could increase the nozzle orifice diameter from 0.120 inch to 0.160 inch. What a difference that would make!

All that was left was venting. It’s difficult to vent long, skinny parts like these; sometimes we need to see how the mold is constructed to find extra places to vent. In this case, both the runner and cavity parting line needed to be vented.

Runner vents should be at least 0.003 inch deep, no matter what material is being used. Part vents, however, are more critical. With PE or PP, normal parting line vent depth is 0.001 inch. If the material is nylon, the parting line vent depth should be no more than 0.0005 inch deep. (Runner vents are always deeper than parting line vents.) Since I didn’t see any evidence of a reverse-taper nozzle being used, I guessed that the material was polypropylene; the part’s stiffness helped in making this guess.

When I spoke with the molder about my suggestions, he asked the toolroom supervisor to listen in on the conversation. They both seemed to be on board with the changes. Just a day or two later, I heard from them: The short shots were gone and it did not take more than a couple of hours to make the changes.