The Troubleshooter, Part 98: Flattening the warp
Published: January 31st, 2008
Published: January 31st, 2008
This article continues our series of troubleshooting reports from one of the leading on-the-spot problem solvers in the molding industry. Consultant Bob Hatch of Bob Hatch & Assoc. has more than 45 years of experience finding solutions to processing challenges. You can reach him at firstname.lastname@example.org.
After rummaging through my projects, I found a part to talk about this month that covers all the key troubleshooting elements needed for a troublesome piece.
The part is a cover for a water pump that does not pump very hot fluids since the olefin material being used wouldn’t handle high temperatures for any length of time. The problem was warpage.
I knew I would find a flow problem after taking measurements. The 0.175-inch nozzle orifice fed a small 0.180-inch cold sprue O-diameter that in turn fed an edge gate into the side of the part where the wall thickness was 0.175 inch.
This presented a pressure loss problem, if nothing else, since the material flow volume running through the sprue, runner, and gate was less than required to correctly fill and pack the part. This is the same problem we have on other cold runner parts when the sprue size is too small to pressurize the runner system properly, and one that should be fairly easy to correct.
But I’m getting ahead of myself.
The part designer had kept a fairly uniform wall thickness throughout the part. I checked the radiuses at all ribs, gussets, and bosses with my trusty medium ballpoint pen and found them to be at least 0.025 inch. This means when I drew a line everywhere that a rib, gusset, or boss attached to the nominal wall, I saw a single, and not a double line.
I wasn’t sure if 0.025 inch was enough of a radius for the notch-sensitive areas. During my initial visual check I could see some ribs, gussets, and bosses that seemed to be pretty thick projections attached to the nominal wall, and this could change the radius required to prevent cracking and breaking during ejection, assembly, or field use.
The first rule of radiusing is that we should use one equal to 25% of the wall we are attaching to when dealing with ribs, gussets, and bosses. Rule #2 tells us to use a radius equal to 50% of the nominal wall when working with an inside corner—where two walls of equal thickness come together. Another rule to remember is that the thickness of a rib, gusset, or boss wall should be no more than 50% of the wall to which these are attached.
Short of cutting the part in half to measure the wall thicknesses, all I had to do was recommend more generous radiuses to comply with the rules. They needed to be thick enough to prevent cracking or breaking but not so large as to cause sinks on the part’s surface or create a fit and function problem for the customer.
I moved on to the sprue and runner, which, as often happens, were undersized and therefore not allowing material to flow as it should to fill and pack the part. The 0.215-inch sprue O-diameter needed to be opened up to 0.312 inch. Where the sprue attached to the main runner—0.280 inch deep by 0.375 inch wide—it needed to be about 0.100 inch larger. The stubby single main runner seemed to be large enough to feed the 0.180-inch-thick wall and the 0.300-inch-deep and 0.340-inch-wide gate with a tapered land seemed like a good size for this part.
Venting and waterlines
Venting was next. I saw flash at the parting line all around the outside diameter of the part, but I didn’t think this was a venting problem—probably more of a machine tonnage problem. My guess was that this part was being run in a molding machine that didn’t have enough tonnage to clamp the mold together well enough to avoid parting line flash. A part this size with an edge gate and running this olefin-type material needed a minimum of 3 tons per square inch of clamp.
The rule for venting on a round part using an olefin material is to use perimeter venting—100% of the parting line—at 0.001 inch deep, with a land of 0.060 inch, dropping into a 0.040-inch channel to a race track, and from the race track out to atmosphere. Draw polish the vent lips to an A1 or mirror finish to make them self-cleaning.
Water problems can also contribute to warpage on a flat part. Remember, the rules for waterlines are pretty simple. First, don’t use quick-disconnect waterline connectors because they often reduce the flow of water to 1?8 inch or even a quarter inch, both of which are reductions from the 7?16-inch-diameter water channels drilled in the mold. This reduction in waterline diameter can easily reduce the flow of water from turbulent to laminar and kills the mold cooling capability by a factor of four or more.
The second rule for waterlines is not to use jumpers on the backside of the mold. Straight in-and-outs are best for good mold cooling and no waterline should exceed 54 inches in length for best heat extraction. No more than a 5 deg F increase from inlet to outlet on a waterline should be accepted.
That was about all I needed to do to correct a flatness problem when using an olefin-type material. Edge gating is recommended to keep parts flat. Sizing the sprue, runner, and gates correctly would stop the molding technicians from thinking they had to increase the material melt temperatures or injection pressures. The gate needed to be sized correctly so it didn’t restrict flow. The venting needed to be generous enough to stop the molding technicians from thinking they needed to slow down the injection speed to avoid blush at the gate. And once the flow restrictions and the pressure losses in the runner were eliminated, most, if not all, of the flatness issues would disappear.
After the flow path restrictions and pressure losses have been taken care of, it is time to reduce the barrel melt temperatures as a way of speeding up the cycle time, if so desired. The cycle time for a part like this should be 250 times the thick wall section of the part (0.180 inch in this case=45 seconds), plus a few seconds more based on the size of machine being used and the type of material being molded. I estimated the target cycle for this part to be 50 seconds once the changes were made to the mold and processing conditions. Just lowering the barrel melt temperatures and correcting the waterline problems usually speeds cycles up by 25% or better.
Sometimes the molder, especially a captive molder, prefers to leave the cycle where it is and just lower the injection pressure—their choice, of course. Often it’s related to timing the molding machines to the speed of the assembly lines.