xxxThe Troubleshooter, Part 35: Glass-filled nylon parts

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 operation for 25 years.

One of the most interesting parts I received this month is one where a balanced runner feeds an eight-cavity mold with a glass-filled nylon 6/6 . The customer complained that he couldn’t get a decent cycle time and the parts had a dull look to them. To keep up with production, the parts had to cycle faster, and a glossy look, which the parts currently do not have, was needed.

I look at the runner and find the typical restrictions to flow. You know, the restrictions in the nozzle orifice, the sprue O diameter, and a long land length at the gate.

The parts are designed pretty well, with the exception that the part is gated in one of its thinner sections, requiring the material to flow through the thin .100-inch section into a thicker .240-inch section at the other end. Overall the part is only .660 inch long so the designer probably thought it wouldn’t make any difference. Unfortunately, he was wrong. Glass-filled nylons need larger diameter runners and thicker part walls by as much as 50 percent more than unfilled nylon parts.

Knowing Your Flow Ratio
A design book published several years ago by Monsanto (now Dow Solutia) contained a flow ratio chart. It indicated that when you divide the wall thickness into the flow length you will come up with a flow ratio number that can be used to determine the number of gates you will need on a given part and where the gate or gates might be located.

The flow ratio of unfilled nylon is 220 vs. 120 for filled nylon. You can see that working with filled nylon is like working with many of the stiffer flowing materials we commonly mold today.

This part is short enough and the wall thicknesses are generous enough to allow the part to be filled and packed from one gate location. But we still have thick to thin transitions to be concerned with, and this part is gated into a thin section, which is causing some, but not all, of the trouble. To compensate for the design shortcomings, the molding technicians have raised the barrel melt temperatures, resulting in longer cycle times.

In addition, to keep the parts from warping, the molding technicians tried to cool the mold (90F), but all they got for their troubles were dull looking parts. The question is, what should be done to correct the problems without having to run the barrel heats higher, thus increasing cycle time?

Shaping Up the Runner
The gates appear to be sized correctly until I look at the land length. The gates are .060 inch deep, .060 inch wide with a .050-inch land. The land should never be more than .030 inch for any edge gate. Actually the rule is that the land should be one half the gate depth, but never exceed .030 inch. In this case one half of the .060 inch depth is .030 inch, so we know the land needs to be .030 inch, not .050 inch as it is now. Therefore, the land will be part of the corrective action.

The runner is cobbled up a little bit, but I can see this appears to be because the mold has been set up with a common sprue and main runner; inserts are then changed to make the many different parts that run through this mold. The problem is that the main runner is sized smaller than the subrunners, which results in pressure losses in the runner system. The sprue is also undersized, causing more pressure losses. Undersizing the sprue causes the nozzle orifice to be greatly undersized, which leads to shear points in the flow path. All in all, this is not an easy mold to run the way it is set up.

Let’s review what we have. The nozzle orifice is .125 inch feeding into a sprue OD of .140 inch. The main runner is .170 inch, and feeds into .245-inch subrunners. The second level subrunners are .240 inch and they feed into third level subrunners that are .175 inch, which in turn feed the gates. Each gate is .060 inch deep and .060 inch wide, with a .050-inch land. I do not see any evidence that the runner is vented.

Correct sizing for the runner and gate system on this part should be a .290-inch nozzle orifice feeding into a .312-inch OD sprue. The sprue should feed into a .250-inch main runner and the first level subrunners should be .225 inch. The second level subrunners should be .200 inch and the third level subrunners that feed the gates should also be .200 inch. The gates are OK at .060 inch deep by .060 inch wide but the gate land should be .030 inch.

The runner system should be vented at the sprue puller and at the end of every runner. Runner vents are deeper than part vents, so these runner vents should be .003 inch deep, as wide as the runner diameter, and out .060 inch from the parting line of the runners into a .040-inch-deep channel to atmosphere. Then the vent lands are to be draw polished to a mirror finish to make them self-cleaning. The sprue puller can be vented by taking .0005 inch off a side of the core pin, then going down .060 inch from the end of the pin with the .0005-inch dimension. Then cut a groove from there down the length of the core pin to the head of the pin. Finish the job by draw polishing the end of the core pin to make it self-cleaning also.

Bringing Out the Shine
With these changes we will be able to run the glass-filled nylon at a 520 to 540F barrel melt temperature, depending on glass content of the nylon. Properly venting the runner allows the parts to be filled and packed with about half the injection pressure previously required.

I would guess the injection pressure was around 1200 psi before and I’ll bet it will now be around 600 to 800 psi. The mold can be run warmer than before—I would suggest around 150 to 180F anyway, up from the 90F they were forced into by the previously high barrel heats.

The dullness on the part surface will be gone and now we will have a glossy look. I suspect the parts will be more rigid than before as well. All in all, it was a pretty easy mold to fix.

I called the customer to tell him my recommendations and before I could say anything, he said, “I know what you are going to say. You’re going to tell me to open the gates!” I just chuckled and told him the gates were OK, but it was everything else that was wrong. I went on to tell him that I don’t just open gates, I open up the flow path to eliminate the flow restrictions and shear points. I could have told him to open the gates because they are on the small side, but I didn’t have the heart to tell him he did everything wrong.

For glass-filled materials, the depth of the gate is commonly equal to 90 percent of the thick section of the part. His gate is much smaller than that and if he had been molding bigger parts we would have had to suggest to him that the gate be made bigger.

The molder said he would have his toolmakers take a look at the mold to see if they agreed with the recommendations and then go from there. I called him a couple of days ago and he said the toolmakers had made every single change I suggested, and when they put the mold back into the machine all the toolmakers went over to watch it run. They were amazed when all the molding technicians had to do was lower the barrel melt temperatures, warm up the mold, and turn down the injection pressures, just like we suggested.

The parts were glossy, the glass filler had been buried, and a resin-rich surface that we all strive for had been achieved. The cycle was decreased by about 25 percent. Now they want me to come into their molding shop and do an in-house seminar on the basics for the molders, toolmakers, and engineers.

I told them I would look at my schedule and see what I could do. Let’s see, they’re in northern Michigan and winter is quickly approaching. I suspect it will be next spring before I can fit it in, but in the meantime I suggested they keep sending me boxes of parts and runners, and I’ll keep writing reviews for them. I also want them to start thinking about optimizing their molds, using my suggestions. Then when I do sit down with them, they will be much more receptive than they are now. Now they are just pleasantly surprised it worked. Later they will know why it works, time after time.