The Troubleshooter, Part 25: Thin-wall ABS parts with surface defects

It was a Monday morning, and I received my usual number of packages containing parts and runners from customers for me to look at. I usually take them as they come-first come, first served-or as Chuck says when he walks into my office, "It looks like you use the FISH inventory system here: first in, still here." He doesn't understand I need to keep parts around for a few weeks in case a customer calls back and needs to talk about them some more.

Well, I came to one small box from a molder on the east coast. It was a perfect example of what happens when the wall thicknesses start to get down in the range below .040 inch. So I had my story for this month.

The tool was a two-plate mold with three cavities producing small parts that look like covers for the end of something. The parts had a flow mark at the gate and a surface cosmetic problem at the end of the fill areas. The material was a gray ABS, possibly a plating grade since they are usually gray in color.

I called the molder but couldn't get past the receptionist, so I couldn't get any more detail than what I could see by looking at the parts and runner. Not that I really needed any more information, but I really hate to assume anything when it comes to molding.

Blush and Blemishes
The blush or jetting condition at the gate indicated a gate size problem. I looked at the gate and could see it was really too small for an ABS part. The complication here is that the part had a .070-inch wall running the length of the top, making the thickness at the gate .070 inch. The side walls, however, were thinned down to .035 inch.

The present gate is .040 inch deep and .060 inch wide with a .050-inch land length. The gate should be .060 inch deep and .120 inch wide with a .030-inch land length. Not too far off, but with ABS you can get in trouble if you don't follow the rules. The runners were trapezoidal, which is not good for an edge gate. Full round are the only runners for edge gates. Trapezoidal runners are fine for subgates and three plates but not for edge gates because they set up a high-shear condition and cause a blush at the gate.

Runner and Gate Modifications
So the recommendation is to recut the runner plate to make the runners full round and of the correct diameter for ABS material. The diameter of the subrunner that feeds the gate needs to be .225 inch, and that diameter should go all the way up to the main runner, not split up in two dimensions like it is now. Here the molder used a .155-inch-deep-by-.225-inch-wide section that runs out from the runner to the part; then it is reduced to .120 inch deep by .175 inch wide before it gets to the gate. You only have to reduce dimensions like that when you are splitting the material flow to feed two parts off the same runner.

The correct way to size the runner for these parts is to make the subrunner full round and .225 inch in diameter. Then the molder needs to increase the diameter of the main runner to a full round .275 inch from the current .185-inch-by-.210-inch size. This will help maintain pressure on the material as it leaves the machine nozzle and flows through the runner system. The sprue O diameter needs to be .312 inch with the nozzle orifice at .290 inch to complete the runner system.

Sometimes I think we lose track of the fact that when the material leaves the nozzle it loses its heat source. If we don't size the runners and gates correctly, we cause pressure drops through the runners and gates that, coupled with material that is losing its heat rapidly to the "colder" runner plates, will cause problems with filling and packing of the parts.

Venting
We have now sized the runner and gates. All that is left is to be sure the mold is vented properly. The runner needs to be vented at the sprue puller and the ends of each runner. Runner vents are .003 inch deep with a .060-inch land to a .040-inch-deep channel to atmosphere, and the vents need to be polished to make them self cleaning.

The part needs to have vents at least one per inch of parting line at .200 inch wide, or try using two vents per parting line inch at .150 inch wide. This will give you 20 or 30 percent of the parting line vented, depending on which width you use. Most molds are way undervented. I have been told you can't overvent a mold, and perimeter vents are proof of that.

Part vents for ABS should be .001 inch deep if you're conservative or .0015 inch deep if you're like me. Go out .040 inch from the parting line and drop into a .040-inch-deep channel to atmosphere, and polish the vent lips to a high polish to make them self cleaning.

The only difference in runner vents and part vents is the depth of the vent. Once you learn to optimize your molds, you will see the parting line vents can be deeper than you've ever used before because you are keeping the melt temperature of the material in the barrel at the lower end of the material manufacturer's recommendations. With better venting, you will be able to keep the injection pressure below 1000 psi.

So we have made the gates deeper and wider and shortened the land lengths. That will get rid of the flow marks at the gate. The runner has been sized properly, and that will eliminate any pressure losses in the runner system. The runner has been changed from a trapezoidal shape to a full round with an abrupt transition for the material to go through at the gate instead of going through the high-shear condition we had before we started optimizing.

The cosmetic defect at the end-of-fill area will disappear now that the mold is vented properly and we are no longer trapping air and getting the dieseling effect of compressing air at the end of fill. EDM machines are great, but they have taken away all the "leakage" areas we used to have to shove air out of the way. I suppose this is the one main difference between the molding of years ago and molding today. We no longer have those natural vents to help us out and have to rely on the toolmakers doing their thing.

I finished my review and faxed it off to the molder, putting a copy in my file in case the fax machine screwed up again. I got a call back from the customer 20 minutes later. He told me he had reviewed the fax, totally agreed with my recommendations, and would have the toolroom start on them right away.

A couple of days later, he called and said the parts were running great. I suggested to him we could reduce the diameter of the runners some day if he wanted to build a new mold or replace the runner plate on this one, but he said he could use the regrind without any trouble and wasn't worried about going to smaller runners. He was happy with everything just as it is and didn't want to take any chances on changing anything. Sounds like a former toolmaker to me!

The molder didn't mention anything about cycle time improvements. But with the changes we made, he could certainly work on the processing conditions and possibly speed the cycle up if he wanted. I don't push speeding the cycle up anymore, but it certainly is a possibility once the mold is optimized.

Releasing Stress Concentrations
One last thought I had has to do with where the .070-inch wall joins the .035-inch wall on each side of the part. The molder needs to put a .020-inch radius where those two walls join. The way it is, he will have a stress concentration along the entire length of the part on both sides of the .070-inch thicker section.

The basic rule I use for a thick-to-thin transition is to make the transition over a distance equal to three times the thicker wall or three times .070 inch, which would be a dimension of .210 inches. That is more appropriate for bigger parts, and I am suggesting a .020-inch radius to avoid any future problems. If this part is ever going to be plated, the radius is going to be absolutely necessary.