The Troubleshooter, Part 17: Stress cracks in ABS

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 box the other day that was so big I thought there might be a pony inside, but when I opened it and dug through the packing materials, all I could find was a fan blade assembly. No pony! I was really disappointed. I've got several kids at home who would have loved to get a pony for the back yard. But they'll probably have just as much fun with the fan blades.

The fan blade housing was made up of five fan blades connected to a center housing that had a brass insert molded into the middle, all fed through three gates into the top surface of the housing. The gates were equally spaced at 120¡ intervals around the top surface. The fan blade assembly is the kind used in a central air conditioning unit like the one you may have outside your house or shop.

Protection from UV
The material is ABS, colored black, and the problem is cracking of the blades where they attach to the housing (Figure 1). At first, it appeared to be stress cracking of some kind, possibly due to using ABS in an outdoor application or something to do with the centrifugal forces that occur when the central air unit functions.

The molder probably thought the black color would protect the ABS part for outdoor use, which would be true if he had used carbon black as a colorant at a 2 to 3 percent loading. It didn't look to me like he had enough carbon black in the material to protect it for outdoor use. A better material for this application would be an ASA material with 2 to 3 percent carbon black in it for protection from the rays of the sun.

ASA is often referred to as a "weatherable ABS," which, of course, it isn't. It's like ABS, but it has an impact modifier that won't degrade as easily when exposed to UV in outdoor

applications. Even though the ASA is "weatherable," I still recommend that it be given extra protection by using the heavy load of carbon black; in other colors I suggest using a UV inhibitor in the color package.

The ASA material shrinks like ABS, needs to be dried like ABS, and processes like ABS, so the molder typically doesn't notice any differences when switching from ABS to ASA material, as long as the melt flow is similar. It looked like this problem could be solved simply by switching material, but then I remembered that the five fan blades were being fed from only three gates. My guess was that some of the cracking of the blades was due to an out-of-balance condition of the fan blades due to uneven fill and pack pressures and unbalanced flow patterns.

Runners and Gates
My recommendation to the molder was to use five gates to fill the five fan blades, one gate to fill each blade, all gates still located on the top surface of the center housing and still using the three-plate mold design. Naturally, we will have to modify the runner plate to incorporate the new five-drop design.

The molder had a single- and a two-cavity mold in which to run these parts. He was having the same problems with parts out of both molds. That alone points to a material problem, but no matter what material he uses, he will have to be sure the molds feed material to the cavities in a balanced condition. Otherwise the blades will flutter and vibrate, not to mention that they will tear away from the hub.

For the single-cavity mold, I suggested using a sprue into a disk gate using an old diecast design that works great for pulleys and parts like this. The idea is to shape the sprue and disk gate like an upside down funnel that feeds a .250-inch circular-shaped trapezoidal runner that in turn feeds the five drops to the .090-inch diameter, .005-inch land gates (Figures 2-4). The sprue O diameter will be .375 inch and the funnel or spreader portion (the disk gate) is .300 inch where it is attached to the sprue and .275 inch where it attaches to the circular runner that feeds the drops. Last but not least, we couldn't forget to vent the circular runner.

For the two-cavity mold I suggested a sprue with a .375-inch O diameter into a .325-inch main runner into a .300-inch subrunner that feeds a .275-inch subrunner that feeds a horseshoe runner that feeds material to each of the five drops to the .090-inch diameter, .005-inch land gates (Figure 5). You'll notice I used the extra subrunner level to balance the pressure drops at each of the gates. I cautioned the molder about starving the gates of material by undersizing the sprue, runners, or gates, which would cause the barrel heats to be raised and in turn could cause warpage and slow cycles.

The wall thickness of the blades and the hub ranged from .100 inch at the end of the blades to .125 inch where the blades attach to the center housing. The thickness of the housing itself is .150 inch in all areas. Based on these dimensions, I suggested .090-inch gates with .005-inch land lengths.

Vents Help, Too
I figured as long as I was this close to doing an entire optimization review, I might as well go ahead with the venting details. I suggested the molder use perimeter venting around the parting lines of the parts. The depth of the vent would be .0015 inch deep, a land of .060 inch into a .040-inch-deep channel to a racetrack and then vent the racetrack to atmosphere. I also told him to be sure to polish the vent lip to an A1 or A2 finish in the direction of air flow to make it self- cleaning.

The runner also had to be vented. We vented the runner at the ends of each horseshoe runner with a .003-inch depth (so you can feel a little flash), then went out .060 inch from the runner parting line and dropped into a .040-inch-deep channel to atmosphere. Polishing the vent lips makes them self-cleaning. The width of the vents is the same as the width of the runner being vented. And we couldn't forget to use a self-cleaning vent on the sprue puller also.

It's the combination of the correct design, material selection, and processing expertise that makes a mold run efficiently-one a molder can make a profit with. Molders should check the part design to be sure wall thicknesses have been optimized, the sharp edges and sharp corners have been radiused, and the runners are sized correctly for the type of material and the melt flow being used.

The gate dimensions need to be based on wall thickness and volume of material flowing into each part. It's also important that the material itself performs at or above the level required in areas such as impact, tensile, heat deflection, and chemical resistance, to name a few. By watching these basics, molders can avoid a lot of common molding shop problems that are the cause of slow cycles, rejects, and low profit margins.

So how did the fan parts run after all the changes we suggested were made? Just as we had expected, the parts looked good, they ran on the central air unit like they were supposed to, and the stress cracking problem disappeared.