Scientific molding, Part 1: Portability

One of the great promises of scientific molding is portability. With it a mold can be moved almost seamlessly from a 400-ton Van Dorn with an 8:1 intensification ratio and a 60-mm screw to a 450-ton Milacron with a 10:1 intensification ratio and a 68-mm screw. All this is thanks to the fact that scientific molding process parameters are derived from plastic variables, not machine or press variables. 

Still, it's hard for molders who use traditional, nonscientific methods to comprehend that a tool can move between such different presses, even hydraulic to electric, without lengthy and careful requalification. It sounds too easy. In reality it must be more complicated. 

To find out, IMM visited Hobbs Corp. in Springfield, IL. Hobbs molds and assembles pressure and vacuum switches, hour meters, halogen lamps, shifters, and ignition and turn signal switches. Since early 2000 Hobbs has been exploring and implementing scientific molding, led by manager Gary Chastain. 

Along the way Chastain has learned what it takes to bring scientific molding in-house. More importantly, he and Hobbs have learned how the adoption of this molding technique brings order to the molding process, shaving seconds from cycle times and improving operation and throughput. (For a primer on scientific molding, see the October, November, and December 1997 issues of IMM; or search the Article Archive.) 

The occasion of IMM's visit is to witness a test of scientific molding's portability: Optimize a mold and establish process parameters on one press using scientific molding principles, and then move it to another press and produce good first-shot parts based on the established parameters (see test specifics). 

On hand to coordinate this effort is John Bozzelli, principal of IM Solutions, consultant, and scientific molding originator. Bozzelli is generally interested in "keeping our molders internationally competitive." At Hobbs he wants to test his Universal Setup Sheet, an Excel worksheet that helps molders convert processing parameters for a given mold between different machines. Assisting is Mike Van Duine, a General Polymers technical service rep who has been instrumental in helping guide Hobbs' use of scientific molding. 

Press One
Before the mold is optimized, a series of tests is conducted to assess the mold and capabilities of the press. First, Bozzelli and Chastain perform a rheology study with a series of short shots. All four parts from each short shot are laid out in rows, and each row is labeled with the pressure at transfer and fill time for the shot. Data from these shots are used to derive a rheology curve for the material in this mold, which in turn is used to establish optimum velocity for first-stage injection. 

The test also reveals that cavity four in the mold consistently produces the least full part and is most susceptible to fill problems. It's determined, then, that a cavity transducer should be installed in cavity four, the thinking being that, as the lowest common denominator, a stable process in this cavity should indicate stability in the other three. 

Next, a series of shots to determine pressure loss is conducted to check for pressure limitations that might exist in the system (Figure 2). The series runs through the nozzle, full runner, runner and gate, and a full shot. The pressure required for each shot is plotted; unusually high pressure losses are highlighted in a bar graph. This series of shots for the ignition switch show no pressure loss problems or concerns. 

Ready to optimize the mold, Chastain and Bozzelli hook up RJG's new ePak data acquisition system to read cavity pressure performance. The crew optimizes injection pressure, injection velocity, transfer position, fill time, and pack and hold pressure, and then conducts a gate seal study to determine pack time. Once the process is stabilized and acceptable parts are produced, basic data are recorded for the mold. Again, these data are specific to this mold and material, irrespective of the molding machine: 

The cavity pressure curve for an ideal shot is saved for comparison after the mold runs in the second press. 

Press Two 
Ready to prove portability, the mold is pulled from the first press and dropped into the 250-ton Magna. With the above critical data entered into Bozzelli's Universal Setup Sheet, new control parameters are established given variations in screw size and other plastic variables. Bozzelli's setup sheet performs these calculations automatically, but they're worth exploring for understanding. 

Remember, the goal of scientific molding, when transferring on position, is to consistently move the same volume of material at the same velocity from the same position (when transferring on cavity pressure, pressure targets govern switchover). Like setting your car to run on cruise control, a machine properly configured to meet this target then automatically compensates for changes in temperature, viscosity, and other variables. 

The most important values to account for when moving from press to press are volume, plastic pressure, and fill time. The constant value is the shot volume, 103.42 cu cm in this case. Moving from the 220-ton machine to the 250-ton machine saw a change in screw size from 45 to 52.6 mm. The goal is to determine the shot size (and by extension transfer position) required for the 52.6-mm screw to produce a shot of 103.42 cu cm. 

The formula to do this calculation is V=r²(L), where V is volume and L is length (shot size). We solve for L: 103.42 cu cm/3.1416(2.63 cm)²=4.76 cm or 47.6 mm. (Volume units are usually cubic inches or cubic centimeters; screw size was converted to centimeters from millimeters.) So, a barrel with a diameter of 52.6 mm and a length of 4.76 cm contains a volume of 103.42 cu cm. 

In this case, given the relatively large screw diameter, shot size is only 23 percent of barrel capacity, below the recommended minimum. As a result, the screw moves only a short distance for each shot, leaving little room for error. Bozzelli says that ideally the screw/barrel combination would be smaller, the shot size would be larger, and the margin for error would be minimized. 

Still, transfer position is set to 18.5 mm, adjusted to compensate for slight suck-back on the screw. Next, Chastain turns pack and hold off and modifies velocity to meet the target fill time. The screw rpm percentage is adjusted to meet the target recovery time. The cooling time is dialed in and the ePak pressure monitor is hooked up again. Total setup time to this point is 15 minutes on the second press. 

The first shot is run and parts are collected. Together all parts weigh 68.54g, the same as those from the first press. Individual part weight is also the same. Cavity pressure overlays reveal just 25 psi variation from good shots on the first press (Figure 1). Hydraulic pressure shows more variation—attributed to a different nozzle. But, again, because critical plastic variables are met, changes in hydraulic pressure are irrelevant. 

What Does it Mean? 
Most molders readily grasp what scientific molding offers: process control based on easily measured plastic variables, independent of machine behavior. Because of this, a mold can be moved from press to press without laborious setup, tweaking, and adjustment to compensate for differences in machine capability. 

In a molding world where tools move so readily from machine to machine, the ability to quickly establish a process that produces good-quality parts is paramount. 

Next month we'll chronicle Hobbs' discovery and implementation of scientific molding. 

Gary Chastain will speak on Hobbs' scientific molding experience at IMM's technology conference, "Molding Technology 2001—Technology Investments that Pay Off," Oct. 1-2 in Chicago, preceding Plastics USA. For more information, go to www.immconference.com.