The Troubleshooter, Part 20: Warping and voids in nylon

This latest problem is more common than most of us think. The parts are round with a solid molded shaft running through the middle of a disk. The shaft extends 2 inches perpendicularly above and below the disk. The material is a 33 percent glass-filled nylon 6/6 that is precolored with carbon black at a high enough loading to make the material fairly UV resistant.

How the Parts Were Molded
The problems mentioned in the customer's note were warped parts and voids in the shaft, mostly in the thicker cross section of the part where the shaft and the disk intersect. I reviewed the process conditions from the setup sheets and saw the molder was using high barrel heats (580F) and high injection pressures (1400 psi), both good indications of flow problems in or near the runner system. From looking at the runner system, I could see the machine nozzle orifice was definitely undersized and the gate, on the end of the runner, also appeared to be undersized based on the wall thickness of the part.

The injection speed was set a little bit slower than I like to see for glass-filled nylon, indicating the technicians may be giving any trapped air time to get out instead of taking care of venting problems in the mold. Maybe the runner isn't vented, or perhaps it is a lack of vents on the part itself, or maybe it's both.

Mold temperatures were okay at 180F, which will bury the glass and give the best looking part surface one can expect from a glass-filled nylon 6/6. Besides, if a better looking surface were important, the customer would have selected nylon 6 with glass. He must have wanted the higher heat deflection temperatures of the nylon 6/6.

The backpressure was set at 50 psi, which will keep the glass fibers from getting chopped up during screw recovery. The screw rpm was set at a low speed, just enough to get the screw back before the mold opened, which again is perfect for glass-filled nylons. So everything looked good except for the high barrel heats and the high injection pressure.

Usually, I like to watch the pressure gauges on the molding machine to see where the injection pressure starts to build up, so I can figure out where the restriction to the material flow is. This works great on older machines, but on new machines I just look at the computer screen. Things sure have changed! If the pressure starts to build quickly, then the restriction is in the machine nozzle or in the runner system. If pressure starts to build later in the fill sequence, it probably means the material made it through the nozzle, sprue, runner, and gate and has filled and started to pack the part.

But this time I am not standing beside the molding machine, so I was only able take a guess at the reason for the high injection pressure. This mold has a heated sprue bushing feeding a cold runner that in turn feeds directly into the part right on the end of the shaft. I looked at the cold runner again and could see that the orifice in the heated sprue bushing was far too small to feed and properly pressurize the trapezoidal runner. The runner was .250 inch deep and .250 by .300 inch wide and about 4 inches long.

Checking Both Ends of the Runner
I saw the molder was trying to feed the runner with a .060-inch orifice coming out of the heated sprue bushing, and it should be at least a .250-inch orifice-if not a .312-inch orifice-to feed a runner of this size. The way this is designed will cause the glass fibers to bunch up in the heated sprue bushing orifice and cause tremendous restrictions to material flow, which will cause a big pressure loss. This would be a reason for the high injection pressures I see on the setup sheet.

On the other end of the runner, the gate feeds into the end of the part. The gate diameter feeding the part is only .030 inch, which is feeding into the .250-inch stem of the part. The minimum gate diameter I use for glass-filled nylon is .090 inch for gates like this on parts with a .100-inch wall. In this case, with the .250-inch stem dimension, I would make the gate quite a bit bigger, like .125 to .150 inch.

If the gate is too small on a part like this, with a direct gate straight into the part and no core or anything to break up the flow, you will get jetting just inside the gate. In these cases, I size the gate somewhere between 75 and 90 percent of the wall thickness I'm gating into. That would make this gate diameter somewhere around .180 inch, even though I would probably suggest starting at about .150 inch and staying steel safe. (Does this make the toolmakers that read this column happy?)

The only area left to cover is venting. I like to use perimeter venting on the parting line of the parts and individual vents for the runner system. The depth of the parting line vents can be .001 inch deep for glass-filled nylons and the runner vents can be .003 inch deep. The vents should come out .060 inch from the parting line on either type of vent, then drop into a .040-inch-deep channel to atmosphere and polish the vent lip for self-cleaning.

I told the molder the changes I would make and that I'd be waiting to hear back from him. It didn't take long. I got a phone call the next day from the molder saying he had made the changes, over his toolmaker's objections. He said the toolmaker didn't think the changes would help that much.

The molder went on to say the nylon parts were now running warp free and the voids were gone. With the larger flow path, he had been able to lower the barrel heats to 540F across the board. With the mold and runner vented, he had lowered the injection pressure to 800 psi. He added a little extra for the hold pressure so he could pack out the voids as they formed. It added 4 seconds to the cycle, which he took off of the cooling time, so the cycle stayed the same. He was no longer getting rejects for warpage or voids-a big change from prior runs.

The molder said he had thought about speeding up the cycle since his barrel heats were lower and the part was setting up more quickly, but had decided against it in favor of matching speeds with an assembly operation he was running in conjunction with the molding of the nylon parts. With the rejects pretty much eliminated, he felt he could live with the slower cycle, especially since the increased productivity more than paid for the mold rework.

I see parts all the time with gates that are too small, and I often see parts with heated sprue bushing orifices that are too small, but I don't see both problems in the same mold very often. I think this specific problem is because these particular toolmakers were used to using small gates for unfilled nylon parts, not realizing that glass-filled nylon parts need gates that are twice as big as those for unfilled nylon. It's actually kind of confusing, since runners need to be only about 10 percent larger in diameter when switching from unfilled to filled nylon, but gates need to be twice as big.

Actually, I did talk with the toolroom manager. He didn't understand why the restrictions were a problem for the molding technicians. He thought that they could just inject the nylon faster through the restrictions to make up for any pressure losses, and that would take care of the orifice and gate sizing problem. I spent some time helping him understand that most of the molding problems I am called in on are caused by tooling problems and he was in a unique position to help the operation more than he ever thought possible.

He agreed to spend more time on flow path sizing and less time on trying to stay steel safe. He also agreed to work on venting the molds a whole lot more. He will have plenty of help from the design engineers and the molding technicians now that he is willing to make needed changes to the molds.