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.

Injection mold cooling: A return to fundamentals

Fundamentals concept on keyboard
We hear a lot about “breakthroughs” in our industry. But is all that gadgetry driving up profits? Not really, because we have been dazzled by technology and ignored the fundamentals of the injection molding process.

We read and hear a lot about “breakthroughs” in our industry: Robots eliminate operators; auto-feeding systems never allow the machine to go dry; snazzy signal processors and transducers monitor every microsecond of the molding process. With all this gadgetry, however, are we seeing more profit and a return on investment for the money spent? Not really, because we have been dazzled by technology and ignored the fundamentals.

Recently, I got an e-mail from a guy who had just taken over the position of lead technician. He wondered about the use of chillers and their expense. He also wondered about the quality of his products when the setup sheet used "tower water" as the main source of cooling for molds and machines.

The trouble with tower water

Let's hit the simple but often overlooked problem first—tower water. When people first build a molding plant, they decide on the number and size of the molding machines and calculate the power and cooling requirements. What they tend to ignore is what happens when additional machines are purchased, because that is covered in the “safety margins” of the original designs.

Heat exchange is necessary because:

  • the machine generates its own heat; and
  • the mold will heat up because it must cool the 300°+ plastic that is injected into it.

Heated machine oil is cooled directly from the tower. The molten plastic's heat first dissipates into the mold steel, is transferred to the cooling circuits and then to the mold's heat exchanger (generically called a Thermolater, although there are other suppliers) and finally to the tower's evaporative cooling circuits.

Evaporative cooling depends on the evaporation of water. This depends on the outside temperature, relative humidity and a host of other variables. It is obvious that when the outside air changes, the temperature of the tower water also will change. As the tower water's temperature changes, your mold temperature will change, and the dimensions and quality of your parts will change.

Another reason to avoid directly putting tower water in your mold is scale buildup. With water flowing through a mold you have the perfect setup for electrolysis, where the minerals in the water will plate out onto the waterlines. Just 1/64 in. (0.4 mm) of scale buildup can reduce the heat-transfer efficiency of a waterline by 60%, even with adequate flow.

Fun facts about heat distortion temperatures

First fun fact: The ideal ejection temperature for any molded part is when it reaches 80% of the material's heat distortion temperature (HDT). Second fun fact: If you check the literature, no thermoplastic resin's HDT is so low that the 80% figure turns out to be room temperature or lower. There are some practical exceptions: Thin-walled elastomers tend to turn themselves inside out during ejection. If dimensions are not sacrificed, if you “over-cool” the part prior to ejection it can be rigid enough to conventionally eject.

These fun facts beg a simple question: If this is correct, why do we need chillers? You use a chiller in an attempt to overcome inadequate cooling in a mold.

Most molds use a Thermolator to maintain mold temperature so that the part can reach 80% of HDT as efficiently as possible. Keep in mind plastic is a poor conductor of heat. The heat from the plastic radiates relatively slowly into the mold steel. The heat-transfer characteristics of the mold steel and the water in the cooling lines are many times faster.

The weak link in this plastic-metal-water heat-transfer system is the water's flow rate. When water flows smoothly like a gentle stream, it flows in layers: This is called laminar flow. The layer that is in contact with something—the walls of the waterline or the bottom of the stream—will flow very slowly. The water at the top of the stream or the center of the waterline only has to slip past itself and flows must faster. With laminar flow, the heat transfers very slowly because it has to heat up this stationary layer before the flowing layers can pick it up and exit the mold.

The laminar flow effect stops when the flow increases. It ceases to flow in layers and begins to tumble over itself. This is called turbulent flow. With the water tumbling over itself in a waterline, it picks up heat directly from the mold steel. Turbulent flow is measured with a very complex dimensionless number called a Reynolds number that uses flow volume, the size of the flow channel, the heat of the water and the viscosity of the water. Instead of going through the calculus, a rule of thumb is 1 gallon per minute (GPM) per circuit will always give you turbulent flow in normal molding situations.

Image: Maksim Kabakou/Adobe Stock

You can purchase flow meters cheaply. Hook them in-line and see what you have. The results might surprise you. Here are some examples.

Age: Like everything else, Thermolators and chillers wear out with time. A Thermolator is not something we tend to do maintenance on. First, check your Thermolator's output pressure as stated in its operating manual. In many cases, the pumps are old, tired and incapable of creating enough pressure to pump the needed 1 GPM per circuit. Repair, replace or retire worn-out equipment.

Line resistance: Let’s do a mind experiment—you want to water your front and back lawn at the same time. You only have one faucet in the front of your house. With a T connector, you hook up a 15-foot hose for the front yard and a 75-foot hose to go around your house and into the backyard. You put two identical sprinklers on each hose. Turning on the water, you think equal amounts will go to each yard but you notice the front yard sprinkler is shooting 25+ feet into the air while the backyard sprinkler is only shooting up five feet. You wonder why—the same pressure source should be the same flow. You forgot about the energy it takes to push the water through the longer hose. Water will always take the path of least resistance. Twenty-foot lines from the Thermolator to the mold require the unit to work harder and only make your utility company rich.


Figure 1: In the best of all worlds, each cavity has individual cooling with a circuit that goes to the main machine manifold. All pressures and flows are equal.

Figure 1.

Figures 2 and 3 show four cavities looped together. The resistance of each baffle will compound on the next, severely impeding flow. Figure 2 shows an external loop; figure 3 shows an internal loop.

Figure 2.


Figure 3.

Figures 4 and 5 show a ladder loop. You can fall into a productivity trap if it isn't designed well (one of the major excuses to use a chiller). Both figures give you the illusion of one circuit cooling only two cavities. Figure 4 shows the "in" and "out" at the bottom of the ladder. This is like our mind experiment. The majority of the flow will cool the cavities closest to inlet. It will become less and less the farther away you get from the inlet and outlet. Figure 5 shows the inlet at the bottom and the outlet at the top. While the bottom of the ladder sees a high inlet pressure, it also sees a high resistance to the outlet, thus balancing the flow.

Figures 4 and 5.

Many molds are built with short and long circuits. Look at the mold designs and designate the circuits that can be looped and what should not be making all the circuits close to the same amount of flow.

How do we document waterline hookups? I've seen photographs (hard to see with more than eight circuits), sketches (hard to read) and descriptions (sometimes hard to do). The best solution is to take from the written description you get for driving directions from your GPS.

In Figure 6 you can see circuit #1 has five loops while circuit #4 is straight through without any loops. It's amazing how techs think they can commit these “waterline maps” to memory. They can't.

Figure 6.

Another “flow killer” is so obvious it is sad. If you have more than 15 machines, there is a better than 80% chance you can find at least one pinched-off circuit. You have looped two circuits together with a length of hose that is too short, bending it sharply and pinching it closed.

Follow up—avoiding common mistakes

Have a setup guide available when hanging a mold and have someone:

  • Check the hookups;
  • make sure the water is turned on; and
  • periodically splice a flow meter into the circuits to be sure they aren't clogged and that your pumps are working properly.

Waterline diameter

Realizing how silly this sounds, I have seen large molds cooled with ¼-in.-diameter waterlines. The physics of the Reynolds number and commonsense will tell you a small-diameter waterline and a complex path, or one with several restrictive fountains or bubblers, will require an extremely high pressure to get 1 GPM flow. Good mold design will also tell you a cooling line can only efficiently cool within three diameters of its outer wall.

If you can only remember one thing on the topic of cooling a mold, put the Mississippi River through the mold. Temperature is easy to regulate but flow determines the mold's temperature.


Injection molding should be both fun and boring. The fun side comes with psyching out the process to balance everything. All this effort should come when you first run/qualify the mold. If you did your homework properly, the boring part is sitting back and watching the profits roll in.


At the end of a few e-mails, my guy tested and scraped a few worn-out Thermolators. He got the engineer to re-write clearly the process sheets and waterline diagrams. He and his crew went on constant hunts and found pinched off waterlines.

The results:

  • Less scrap, higher productivity;
  • more consistent outputs;
  • less use of utility-eating chillers;
  • reduced time invested in troubleshooting.

In other words, more profits for less work.

About the author

Bill Tobin has more than 30 years of hands-on experience in injection molding. Through his company, WJT Associates, he writes articles, presents papers and teaches seminars helping people improve their profits and productivity. He can be contacted at [email protected].

TAGS: Auxiliaries
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.