is part of the Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Making scientific molding work

Scientific molding enables faster setups, optimized cycle times and higher yields. When you boil it down, this means improved profits.

When I first started consulting, one of my mentors told me the secret to embedding change was to answer the question, "What's in it for me?"

The outcome of employing scientific molding is faster setups, optimized cycle times and higher yields. When you boil it down, this means improved profits. Great for the company, but very quickly the folks who are supposed to implement this find out there's nothing in it for them. They are paid the same, while the boss buys a new Mercedes. What if the profits improved and management decided to split this windfall with those who made it happen? Bingo! There is the incentive and it didn't cost management anything. This will take some work, but when scientific molding is in place, it’s always a money-making success.

Scientific molding has one over-riding tenet: "Show the plastic the same set of conditions and you'll get the same part." Nowhere is there a requirement for new equipment or identical presses! However, poorly maintained or worn out equipment guarantees failure.

Many new engineers fall for the sucker punch of running the mold at the try-out shop and then finding they can't get the same quality parts at the same cycle time when it's run elsewhere. The first question to ask is: "Can my equipment duplicate the conditions I need?" Are your machines capable of being repeatable? Does your auxiliary equipment have enough power to deliver what the mold needs?

A spreadsheet developed by Dr. Rod Groleau (used by permission) that's freely available can help you to answer those questions. Click on Machine Consistency Test. If your equipment needs maintenance, do it.

There are only a few variables in molding that are easily learned and controlled: Setup, heat, pressure, time, position and speed.


  1. Waterline hookups can cause problems. I use a “waterline map”: circuit #1 is manifold #1 is hooked to 13-IN, 13-OUT is hooked to 8-IN, 8-OUT is hooked to manifold return #1. It's like the turn-by-turn directions you get from your GPS. If you hook it up differently or reverse a circuit, you'll get a different cooling pattern in the mold. In addition, wouldn’t it be nice if you turned everything on? If this is too confusing, buy a manifold, hook it to the mold and leave all the waterlines in place. Use a fire hose connection to hook your mold manifold to the machine.
  2. When you use a different size machine, you have to consider shot capacity; if it's less than 20% of the machine's capacity, you'll have trouble controlling the switch-over from fill to pack. If it's more than 80%, you'll have difficulty maintaining an even melt temperature. Like all rules of thumb, there are several exceptions. Those materials that degrade easily (vinyl and urethane) process better when they spend the least amount of time in the barrel.
  3. Wouldn't it be nice if the material were dry? Well-maintained dryers work better and faster than ones that aren't maintained. When was the last time the dryer's filters were changed? When did you last check the dew point?


  1. Have you let the mold come up to temperature and measured the mold's surface temperature? Don't rely on the heater controllers. A good way to check this is to measure the temperature of the ejected part. If it's the same temperature as the ejection temperature of the qualification run, your water hookups, temperature and turbulent flow are good.
  2. The barrel heaters only provide about 20% of the material's heat. The rest of it comes from the mechanical shear of the screw. Definitely monitor the barrel heats. Bad heaters can run away and burn your material, or die and create a cold spot.
  3. Do you know how to take the melt temperature? Keep in mind you are usually plunging a very “cold” pyrometer needle into a 400° F+ puddle of purge. At least preheat the needle to within 30° F of what you expect the melt to be. Make your adjustments with the back pressure, not the barrel heaters.


  1. You need enough pressure in your waterlines to generate turbulent flow. Buy a flow meter and hook it into each circuit. Remember, long circuits will have less flow than short ones. Balance the circuits by adjusting the flow so that all circuits have a minimum Reynolds Number of 4,000 or more. High Reynolds Numbers only make the utility companies rich by wasting power-generating pressure—they don't add to the cooling ability. Need to calculate the Reynolds Number? Go here, scroll down and click on Reynolds Number. It's a simple fill-in-the-blank spreadsheet.
  2. If you're running a 250-ton machine, you usually don’t need 250 tons of clamp pressure. Lower the clamp pressure until you get flash, then sneak back up using the minimum necessary. Maybe you find you only need 190 tons. This gives you the option of running a smaller machine. Regardless of the machine, set the clamp at the 190 tons you initially determined.
  3. Speed and pressure are related. Pressure is resistance to force. If your machine can't pump oil fast enough to achieve your desired fill time, you are "machine (speed) limited." Since the viscosity dramatically changes with fill speed, not all your other process parameters will work, either. Your machine should be able to overpower the melt. If the process calls for a fill in 1.5 seconds and your “new” machine can do it, you'll see little if any lot-to-lot variation from the material.


  1. Can you duplicate all the time settings you had in the original qualification run in every machine you run this mold? This includes open, eject and close times.
  2. If residence time—the time the material sits in the screw and is cooked by the heaters— is less than a minute, you probably don't have an even melt temperature. If it's more than five minutes, you'll probably have burned material.
  3. Time is related to heat, which is further related to pressure. If you slow down the cycle, the parts will stay in the mold—the world's most expensive shrink fixture—longer and come out cooler and larger. You will also observe a drop in mold temperature and an increase in melt temperature because the residence time increased. Shorten the cycle time and the reverse happens.
  4. The “curing” or cooling time takes the majority of the cycle. Most techs set the cooling time so that the part comes out cool enough to touch. You should only cool long enough so that the part will not warp. In scientific molding terms, this means the part is 80% of the material's heat distortion temperature when ejected. If this is too hot to touch, wear gloves or use a robot.
  5. With the ideal cooling time established, you must refill the screw for the next shot and decompress the melt. This is achieved with the RPM setting on the screw. When done properly, the material only sits and cooks for about a second more than it takes to open and close the mold.
  6. Packing time can only pack the part until the gate freezes off. You are not in the business of making good-looking runners. Do a gate freeze off study and set your times accordingly. You can't do any more packing once the gate is frozen off.


  1. What's the usual position of “mold open?” Unfortunately, it's usually the thickness of the setup tech plus the room needed for the last tool that was used to remove a stuck part. It should open only far enough to get the parts out of the mold. When you open and close a mold, you're usually moving a few tons of steel. This takes time. The less distance it needs to move, the less time it takes. The mold open distance should have been set at the qualification run.
  2. Where is the transition between fill and pack? Try this experiment: Turn off the packing time completely. Make a part. This is called a “fill only shot.” You should see a completely filled part full of sinks and voids. If it's a short shot, increase the load. If there are no sinks or voids, decrease the shot slightly. The rule: Fill with the fill; pack with the pack.
  3. If your shot is 45 grams, it will always be 45 grams. However, different machines have different diameter barrels. This means your 45 grams will be at a different screw position depending on the size of the injection unit. Translating from one machine to the next is simple algebra.


  1. Speed is movement over time. If you found the mold filled best at 1.5 seconds, always make sure it fills in 1.5 seconds. Injection speed affects the melt viscosity. Usually, going faster gets you nothing; going slower, however, will show a dramatic increase in the pressure required to fill because the melt's viscosity is thicker.
  2. Don't jerk the mold open or closed. You need to let the suction break occur so the part will stick on the ejector side. Jerking it open will stick parts in the cavity. Don't slam the ejector system. If the second suction break doesn't occur on the core side, the ejector pins will punch through the parts. Slamming the mold closed or kicking the ejector system back is an easy way to break the ejector rods and/or beat down your parting line shut offs.

The “button-pushing cowboys” Garrett MacKenzie mentioned in his article (“Top 10 reasons companies fail at scientific molding”) are born from ignorance with years of experience at guessing. These self-appointed gurus are the kiss of death in a production environment. I met one of those guys when I was in China. The process had drifted away and he sat down at the control panel and started changing things. Through my interpreter, I asked, "Why did you change XYZ?" His answer was, "I fix." When further pressed, it became obvious he didn't have a clue. He literally just kept changing things until he got a good part. He fixed short shots on a simple part in five hours. Later on I showed the technician I was training how to fix it in 20 minutes (15 of which involved standing around to ensure the process had stabilized).

If you don't understand the molding process, you shouldn't be changing things.

To avoid button pushing, you have to answer two questions. First, "do I have enough knowledge to make changes?" This is a function of training, not experience. Second, "what changed?" If the process was stable before, it should be stable now, if nothing else changed or malfunctioned. Pull out the setup guide and compare it to what you have now. If this is a troubleshooting issue, remember that changing the process will usually give you multiple outcomes. Make your changes carefully.

Scientific molding is easy to learn and implement. You only have to take the time to understand what the process is doing.

Your choice.

Bill Tobin is a consultant who teaches seminars and helps clients improve productivity. He can be contacted at or by e-mailing him at [email protected].

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.