The Troubleshooter, Part 121: Give the gates some room

A few dimensional changes make for a happy molder.

Lucky me: I received two deliveries from FedEx in one day. Perhaps this means the economy is picking up a little bit. I hope so, because it has been slow for much too long.

I opened the first package and found a small runner with parts attached to each of the four edge gates. Overall I could see that it was a two-plate cold runner tool and the material was a crystalline material of some kind—probably an unfilled nylon 6/6, from the slippery feel of the part surface and the limited flexibility I felt when I flexed the runner and the cold sprue. Each of the four parts was identical, so this was a multicavity mold.

Each part was short of filling all the way out, so I suspected the problem was that the molder could not get the parts completely filled using normal heats, speeds, pressures, mold temperatures, and cycle times. Heck, he was no doubt pressure limited and had the heats as high as he dared raise them. I would say he had a pretty bad problem and really needed my help.

The sender of this package was off on vacation somewhere according to the receptionist in his front office, so I started working on my observations and would modify my answers accordingly when I finally got together with him.

Checking the dimensions
We were at the point that I was ready to put some dimensions to these parts, runners, and sprue. That meant I needed to get out my digital micrometer, which meant I needed to round up my son and find out where he was keeping my digital micrometer.

Sometimes it’s in my desk drawer, but not that day. Sometimes we lay this delicate measuring device on the blueprint table in the Engineering Department but again, not so on that fine winter day. Then I got lucky. I saw Rob off in the distance, so I did what all parents do to get their children’s attention: I called him on his cell phone and asked him to join me in the breakroom.

It was over a cup of coffee for me and a Mountain Dew for him that he told me the micrometer was on my desk but resting safely in the in-basket along with recent correspondence. Then he said something about my desk being too cluttered to leave anything as important as a measuring tool anywhere other than in the basket. He couldn’t get me upset, though; everything was going very well and I wanted to keep it that way.

Now I was ready to do some measuring. First up: part wall thickness. Surprisingly, all of the nominal walls were quite uniform at 0.100-0.110 inch. The vertical walls were 0.100 inch thick and the horizontal surfaces were 0.110 inch. I’m sure they were all supposed to be 0.100 inch thick, but I would guess that the clamp tonnage was too low to hold the mold clamped tightly. They were probably using a hydraulic machine, which is often guilty of this particular thickness problem.

Next to meet the measuring jaws was the subrunner feeding the edge gates. It was full round—perfect for edge gates—with a 0.180-inch diameter, which meets our rule of 1.5 times the wall thickness by a nice margin.

Next was the main runner diameter, measuring 0.200 inch—not bad for the preferred increment between these runner levels.

Next, I measured the sprue O-diameter. It turned out to be 0.065 inch—way too small, no matter how you look at it. It should have been 0.050 inch larger in diameter than the main runner, so this was a big problem for these guys.

Then I measured the sprue length: 3.5 inches. I use a 0.0175-inch taper calculation for each inch of sprue bushing, which means the root diameter of the sprue (where it attaches to the runner) should have been 0.062 inch larger than the existing sprue O-diameter. The root diameter turned out to be 0.245 inch, so it was easy to see that the taper on the sprue bushing was excessive. According to my rules, the proper dimensions for the sprue O-diameter would be 0.250 inch, and 0.312 inch for the root diameter.

With the new dimensions for the sprue, the general purpose nozzle needed to have an orifice of 0.290 inch, just a little smaller than the sprue O-diameter. Instead, the nozzle orifice was 0.060 inch—greatly undersized.

Then it was time to look at the edge gates. I measured each of them at 0.030 inch deep, 0.125 inch wide, and with a land of 0.100 inch. Here again, two of the dimensions were way off. The gate depth needs to be 50% of the wall thickness, or 0.050 inch in this case; its small size was causing jetting at the gate. The width should be twice the depth on parts of this volume, so 0.125 inch would be OK. But the gate land for an edge gate should be half the depth, never to exceed 0.030 inch. This meant the gate land was four times greater than the rules dictate. Additionally, the flow marks just inside the gate indicated the land was too long. Both of these sizing errors made it extremely difficult to fill and pack a part. Now I could see why these parts ended up on my desk.

Wrapping it up
By this time, I had about 5 hours invested in reviewing this part and runner. I figured it was time to write it up and send an e-mail off to our vacationing molder.

A week later I got my reply, and it was a good one. From the excitement in his response, I guessed his batteries got recharged to their fullest. He said the toolroom had made the sprue and gate changes I recommended, and then he wanted to know what I thought the cycle time should be on parts like these. I multiplied the wall thickness (0.100 inch) by 160 for nylon, added 1 second for the 100-ton or so machine that he no doubt was using for this job, and came up with a 17-second target cycle.

He called me several minutes later and suggested I must be crazy, because they had to run a 30-second cycle on that mold. When he slowed down to take a breath, I reminded him that once he made the changes to the mold, he will have opened the material flow path and eliminated two of the restrictions to flow. I told him he would be able to lower his barrel melt temperatures and speed up the cycle by quite a bit, probably down to at least 20 seconds, and maybe he would be able to achieve the 17-second cycle time that I had calculated.

I suggested he make the changes to the mold, lower the material heats, and find out for himself what the new cycle time would be. Well, it didn’t take more than a few days and I got a check in the mail for my well-received suggestions and a big thank you from the men and women in the molding department and the toolroom.

It’s always a pretty thing to see a well-thought-out plan come together like this one did.

Troubleshooter’s 
notebook
Part/material: Unfilled nylon 6/6.
Tool: Two-plate cold runner, four-cavity.
Symptoms/problem: Jetting and flow marks inside the gate; incomplete filling.
Solution: Increase sprue O-diameter, root diameter, and nozzle orifice; enlarge gate depth and reduce gate land length.

 Dimensional changes (inches)
                                 Old    New
Sprue O-diameter      0.065    0.250
Root diameter           0.245    0.312
Nozzle orifice            0.060    0.290
Gate depth               0.030    0.050
Gate land                 0.100    0.025

Following the rules of thumb for gate and runner sizing would help fill this part.