Injection Molding: Process Documentation Is Crucial to Making ‘Identical’ Parts

Injection molding is one of the most complex processes in manufacturing. Our industry has a common problem in setting up a molding job to make “identical” parts on every run, especially on different machines.

Defining what “identical” means depends on the application; but whatever the requirements, the goal is to make identical parts run to run, year to year, even on different machines. How we, as an 80-year-old industry, attempt to accomplish this varies and is something of a Pandora’s Box.

There is no general agreement on best practices. It is desperately needed, as it would enable us to compete more competitively within the world markets. Your future in the plastics industry may depend on it. Where do we begin?

There are five components to making a successful plastics application: part design; resin selection/handling; mold design/construction; processing; and testing. Each has hundreds of details and putting them together requires a trained and talented group of people. One person cannot do it all. Most likely your shop has talented people in two to four of these components.

Want to know more about Injection Molding? Visit the Injection Molding Know Zone

Use them for their specialty and go outside to trusted talent for those areas you do not cover or that may be new. I also suggest that you force a meeting with all parties before steel is cut. Few take time for this, but it will save you time and money when you find a problem before the metal chips fly rather than after. Plus, there are always last-minute issues that are best dealt with when all parties are present. This will aid in getting the process started on a sound footing. Then comes documentation to continue successful production.

The focus here is on processing and how to document a process that is running “good” parts so it can be reproduced from run to run, even on different machines. Documentation of the process is usually captured in what most call the “Machine/Mold Setup Sheet.” Seems simple enough, but if you visited the estimated 16,000 molding shops in the U.S. you’d wind up with 16,001 different setup sheets. Even worse, there is a different setup sheet for each different press the mold goes into. And to really bring you down, it is often the case that the process changes on the same machine as shifts change.

Our grizzled processor, “Crusty Sr.,” will argue that his process is better than what is currently written on the setup sheet. Processors “adjust” the process to their liking. That is the state of our industry. Here is where Scientific Molding deviates from the crowd: You use only one setup sheet per mold—not a setup sheet per press. Document the plastic (not machine) parameters and use them on each acceptable press into which the mold goes. Tempted to say it can’t be done?

In a recent case history, the process was initially developed for a 32-cavity hot-runner mold based on plastic parameters only. It ran for six years and then the mold was sent to a different facility in another location. That meant a different press and processor.

The setup sheet was requested and emailed to the processor. He set up the process to those plastic parameters documented. His response was “Six years and dead nuts on.” So, what is the trick?

There is no trick … you use a scientific method with machines that are calibrated correctly. The setup sheet for the mold provides actual volumes, pressures, temperatures etc. The same plastic parameters are achieved on not just the press but in mold cooling as well. It just goes to show that if you properly document the machine, mold and resin process parameters, good parts are produced, shot to shot, run to run, on one machine or another.

If you visited the estimated 16,000 molding shops in the U.S. you’d wind up with 16,001 different setup sheets.

The point is that if we want to make “identical” parts we need to ensure these processing parameters are controlled and measured to be the same within a reasonable tolerance. That mandates that when a processor sets up a process, he or she inputs the process setpoints that provide the same processing conditions (parameters) documented.

The numbers input on the machine may be different—for example, hydraulic pressures vary from machine to machine—but the specified actual plastic pressures are the same. What process variables do we duplicate? That list is too long for this column, but here are six that must be duplicated. I suggest that once a shift, have someone visit each machine running, document each of these process conditions to ensure part reproducibility/accountability between shifts.

1. Fill time, within a range of ±0.04 sec, for most jobs. This does not mean that the processor inputs the same injection rate in mm/sec or in./sec on the controller. Even if you are using cubic centimeters (volume) the input number may vary as the machine, hydraulic or electric, may not give the same result from one run to another. A machine is composed of mechanical and electrical components, and these do wear (take the non-return valve as just one example). The input for injection rate can vary to provide the documented first-stage fill rate of a given weight, size and fill time.
2. Plastic pressure at transfer; it may or may not be peak pressure during injection. It should be the same between runs and machines. This requires use of the same type and size of nozzle tip. The nozzle becomes part of the mold from a pressure-loss point of view.
3. Cushion—this does not have to be the same from press to press but must be consistent shot to shot. I target less than about 1 mm variance. I define cushion here as the screw position at the end of second (hold) stage. This is not minimum screw position. There are two definitions of cushion, depending on which brand of machine you are using.
4. Recovery or plasticating time with identical plastic backpressure. Not a set rpm, but the same plasticating time, achieved by adjusting the rpm to replicate that time. Replicating circumferential screw flight velocity is more logical, but better results are obtained by replicating plasticating time to replicate melt temperatures.
5. Cycle time, replicating pack and cooling time.
6. Whole-part IR temperature picture, measured as soon as possible before or after ejection.

Are there other parameters that are important? Absolutely, like melt temperature, but interrupting a production cycle to measure that temperature is a hassle and staying on cycle is important for consistency. There are also cooling parameters and still others, but those will have to wait for another column. The six parameters above are a good start, but remember: They are process results not machine setpoints.

Bottom line: Replicate actual volumes, times, temperatures, and pressures with machines that are capable and calibrated. This may sound complicated, but it is simpler than fighting Murphy on every other run or working with the idea that molding is art rather than science.

---

ABOUT THE AUTHOR: John Bozzelli is the founder of Injection Molding Solutions (Scientific Molding) in Midland, Mich., a provider of training and consulting services to injection molders, including LIMS, and other specialties. Contact john@scientificmolding.com; scientificmolding.com.