Sponsored By

The Troubleshooter, Part 49: Trouble in two molds

May 31, 2001

10 Min Read
Plastics Today logo in a gray background | Plastics Today

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. 

I received a call from a molder about a month ago who was asking how to get rid of sinks in a polycarbonate part. I don't know about you but sometimes I have a terrible time visualizing parts over the phone and this was one of those times. I asked him to send me parts and runners and I would see what I could come up with. 

Two days later I was looking at parts and runners from two different molds. Both sets of parts were molded out of the same material, which was a precolored red polycarbonate. Both sets of parts also had similar problems: blush at the gates, flow lines and shear splay on the parts about halfway between the gate and end-of-fill areas, and small amounts of sink in the center of each part. 


I looked at the parts and decided that the blush, flow lines, shear splay, and sinks were all interconnected. The sprue and runner on one set of parts from a family mold (Figure 1) were sized pretty well, but not perfect. In addition, the nozzle orifice feeding the sprue O-diameter was much too small and the land length of the gate was too long. I knew I would have to resize the sprue, but the runner diameter was plenty big at .375 inch. 


The other mold was running similar but smaller-volume parts (Figure 2) and appeared to have the most problems. The sprue was too small in diameter, which made the nozzle orifice quite a bit smaller than it should be; and while the gate was sized OK, it was coming somewhat off the side of the full round runner instead of the middle like it should for most any amorphous material. 

Mold 1 
I decided to call the mold producing the smaller parts mold 1 and the family mold producing the bigger parts mold 2. I started with mold 1 and suggested a larger sprue diameter, more in line with what I usually recommend for polycarbonate. The existing sprue O-diameter was .215 inch and should have been .312 inch. This alteration would provide an opportunity to change the nozzle to a full taper design with an orifice of .290 inch. The main runner sizing was fine at .300 inch—a little bigger than we needed, but it wouldn't hurt us. I operate on the theory that bigger is OK, but small can bite you when it comes to sprues, runners, and gates. 

That took care of the nozzle orifice, sprue, and runner sizing for mold 1. Next I looked at the gate dimensions to see if they would work for these parts. The gate was .092 inch deep and .170 inch wide, and the land was .060 inch. The part walls were .100 inch thick with bubble areas somewhat thicker than the nominal wall. I guessed that the thicker bubble sections measured .140 inch or pretty close to it. I based the gate dimensions on the thicker section of the part since you should always gate into a thick section and flow to the thinner sections to get good fill and pack conditions. 

In a case like this where we can't gate into the thick section, I just size the gate as if I am gating into the thicker section. Based on this practice I would make the gate depth equal to 90 percent of the .140-inch-thick section, or .125 inch for this part. The width of this gate should be about one and one-half to two times as wide as it is deep for a part of this size, which made the width .188 to .250 inch wide. In this case I suggested starting with .200 inch for the width. The gate land is always one-half the depth but never more than .030 inch. So this gate should be .125 inch deep, .200 inch wide, with a .030-inch land. 

As I mentioned before, the gate was not coming directly off the center of the full round runner. If, after making these changes, we still had a cosmetic flow defect just inside the gate, we would have to remove a little bit of steel to move the gate back into the center of the runner. 

Now all that was left was venting the runner and the parts themselves. Runner vents are deeper than part vents so I usually start with runner vents. Runner vents typically should be .003 inch deep, as wide as the runner diameter, with a .060-inch land, dropping into a .040-inch-deep channel to atmosphere. The vent's land should be draw polished to an A1 or mirror finish to make it self cleaning. We should put runner vents in at the sprue puller and at the ends of the runner, either at 90° turns or in the sweeping curves like we had with this runner system. 

Part vents are not as deep as runner vents. Polycarbonate vents are generally .001 to .0015 inch deep, .200 inch wide, with a .040-inch land, to a .040-inch-deep channel to atmosphere. They are also polished to an A1 or mirror finish for self cleaning. One vent should be put in per parting line inch or, in many cases, a perimeter vent can be put in completely around the part. Remember that it is not the width of the vent that flashes, only the depth. 

Most moldmakers don't like to make polycarbonate part vents very deep because they are used to the processing people running high barrel heats—580 to 590F—and with all the injection pressure and speed they can muster. After sizing the nozzle, sprue, runner, and gates correctly we could use barrel melt temperatures of about 530F with hydraulic pressures of 1000 psi or less, depending on the viscosity of the polycarbonate being used. The injection speed could still be fast, as it should be, because vented runners and parts help avoid burns at end-of-fill areas that trap air. 

When air is trapped, it compresses and heats up enough to cause a dullness or a burn on the part. By properly venting, we eliminate the air traps, and the injection speed can be as fast as needed. Polycarbonate requires a fast injection speed because it gives up its heat to metal very quickly. Polycarbonate has what we call an affinity for metal. 

Mold 2 
In the second mold, since the parts were bigger, the gates needed to be wider—and that's really the only difference we should have seen. As I reviewed the parts and runner system, however, I could see a couple of other changes that would be necessary. 

Polycarbonate requires a fast injection speed because it gives up its heat to metal very quickly.

The sprue had a .275-inch O-diameter feeding into a .375-inch main runner. The main runner was bigger than it needed to be, but we could work with that. A main runner of .250 inch would be adequate for these parts. The biggest problem was that the nozzle being used for this mold had an orifice of only .100 inch. But before we did anything we needed to get the sprue diameter opened up so that the O-diameter was at least .343 inch to feed the .375-inch main runner diameter. 

I knew a .343-inch sprue O-diameter wasn't as big as it should be, but it would work. Once you get to a fairly large size, such as .300 inch, you can get by with slight variations from what is considered normal and still mold parts without raising the barrel melt temperatures and increasing injection pressures. 

This mold was slightly different in that we were gating into a tab on one of the parts that was attached to the wall of the part. In this case I would cut the gate land down and have the runner touch the tab, making the tab become the land of the gate. This way we could eliminate shear and pressure loss at the gate and still have a point where we could clip or even break off the gate. I suspected a heated gate cutter would be used when all was said and done. If this tab was involved with any of the optical qualities of the part, we couldn't do it, but it was a neutral area so it was OK. 

The depth of the gate ended up being the thickness of the tab (.095 inch), not as deep as I would have liked, but with the larger-diameter runner here I could make the gate wider than on mold 1 and compensate for the lack of depth. So we ended up with a gate on this part that was .095 inch deep and .350 inch wide. 

The other part in this family mold was slightly smaller in volume, and didn't have a tab to gate into. For polycarbonate we just needed a traditional edge gate that was .075 inch deep to feed the .090-inch-thick wall. It needed to be wide like the other cavity since we had a gate that wasn't very deep for the volume of material that would be running through it. A width of .250 inch seemed adequate. The land would default to .030 inch, so we ended up with an edge gate that was .075 inch deep, .250 inch wide, with a .030-inch land. This took care of the flow path. All that was left now was the runner and part venting. 

Troubleshooter's Notebook

Part: Red polycarbonate parts. 

Tools: Mold 1—two plate, cold runner; mold 2—family mold. 

Symptoms: Blush at the gates, flow lines and shear splay between the gate and end-of-fill areas, and minor sinking in center of each part. 

Problem: Mold 1—bad gate location; mold 2—gate land too long; both had undersized sprue O-diameters and nozzle orifices. 

Solution: Enlarge sprue O-diameters and nozzle orifices, resize gate dimensions, and add venting to runners and parts. 

Result: Cosmetic and optical problems eliminated; parts running fine. 

Implementing Changes
I called the molder with these suggestions and told him what needed to be done to mold parts that not only looked good but would also have the optical properties needed for the customer. His reaction was good and I sensed he was going to make all the changes and get back to me on how everything turned out. 

A month went by before I heard from him. He wanted me to visit his shop to watch one of the molds run, mostly to see if he had missed anything. I found the mold running beautiful parts. He had made all the changes I suggested and what he really wanted me to do was look his shop over to see if he was as up to date as other molders. 

I was impressed. His machine and auxiliary equipment were almost brand new and he was using a dryer that pulls a vacuum to dry material. I couldn't see any splay on the parts so I guess the new dryer was working like it was supposed to. 

All in all, it was a very nice molding operation. He was going to test the second mold the next day but I had to continue on with my schedule and couldn't be there to watch it run. From what I could see, though, he wasn't going to have any troubles with this mold, either. 

Sign up for PlasticsToday newsletter

You May Also Like