Cooling is usually the longest single segment of the molding cycle, and it’s almost always the most ignored. Operators will spend an afternoon tuning fill and pack to shave a few tenths off the cycle, then run the part on a tool that’s being fed through a tangle of undersized hoses and half-blocked lines — leaving real seconds on the table and accepting warp they’ve decided is “just how that part runs.”
There’s a useful comparison here. People keep houseplants alive on the bare minimum of water, and some manage to kill cactus because they decide a desert plant needs none at all. Molds get treated the same way: as long as some water is moving, the assumption is that the tool is being cooled. Whether it’s being cooled enough, and evenly, is a question that rarely gets asked until the parts warp or the cycle won’t come down.
This article doesn’t reproduce historical WJT Associates material. It applies the same shop-floor reasoning — confirm the tool is actually getting the cooling you think it is — to the water side of the process, which is where a surprising amount of cycle time and dimensional trouble actually lives.
The standardization trap: undersized fittings
There’s a real case that illustrates the problem perfectly. A molder built his business on small machines — he was, for years, the king of the 50-ton press. Small machines, small parts, small molds, and everything plumbed with ¼-inch water fittings: the molds, the machine manifolds, the temperature controllers. It was consistent, it was tidy, and it worked fine — at that scale.
Then the business grew into bigger tools that needed real cooling capacity, and the ¼-inch standard came along for the ride out of habit. Now you have a large mold that wants substantial flow, fed through a fitting and hose sized for a fraction of it. The pump and controller have plenty of capacity; the bottleneck is the ¼-inch restriction nobody questioned because “that’s what we use.”
The lesson isn’t about one shop. It’s that standardizing the hardware is good practice right up until the standard becomes a constraint nobody re-examines. A water circuit is only as big as its smallest restriction — one undersized fitting, one kinked hose, one clogged quick-disconnect starves everything downstream of it.
Laminar versus turbulent flow
Here’s the part that surprises people: pushing water through a cooling line isn’t the same as cooling the tool. How the water flows matters more than whether it’s moving.
- Laminar flow moves in smooth layers. The water touching the channel wall picks up heat and forms an insulating boundary layer; the water in the center never really contacts the steel. The line is full, water is moving, and heat transfer is poor.
- Turbulent flow churns the water so it continuously mixes against the channel wall. Every bit of coolant gets its turn against the hot steel. Heat transfer climbs dramatically the moment flow crosses into turbulence.
You don’t reach turbulence by guessing. It’s governed by the Reynolds number — a function of flow velocity, channel diameter, and the coolant’s properties — and the practical target is to keep each circuit comfortably in the turbulent range (a Reynolds number around 4,000 or above is the usual rule of thumb). A circuit running laminar can be “full of water” and still cooling the tool poorly, which shows up as long cycles and inconsistent dimensions that no process setting fully fixes.
This is why the right question isn’t “is water flowing?” It’s “is each circuit turbulent, and is every circuit getting its share?”
What actually starves a tool
When a mold isn’t being cooled the way it should be, the cause is usually mechanical and unglamorous:
| Cause | How it shows up | What to do |
|---|---|---|
| Undersized fittings/hoses | Flow can’t reach turbulence no matter what the pump does | Size fittings, hoses, and disconnects to the circuit, not to a shop “standard” |
| Series-plumbed circuits | Last circuit in the chain runs hot; pressure drop starves it | Plumb critical circuits in parallel; keep series runs short |
| Scale, rust, and biological fouling | Gradual rise in tool temperature and cycle over weeks | Clean and descale on a schedule; treat tower water |
| Blocked or restricted line | One area of the tool runs hot; localized warp or sticking | Confirm flow line by line, not just at the manifold |
| Laminar flow | ”Full” lines, poor cooling, long cycle | Raise flow rate or reduce restriction until the circuit is turbulent |
| Coolant too warm at the source | Whole tool runs hot; cycle won’t come down in summer | Check tower/chiller capacity and supply temperature under real load |
Notice how many of these are invisible from the operator station. The display shows the controller setpoint and maybe a return temperature. It does not show you that the third circuit is half-scaled or that the line you think is flowing is actually plugged.
Confirm the water, don’t assume it
The fix for an invisible problem is to make it visible. You confirm cooling the same way you confirm anything else on a scientific process — by measuring it, circuit by circuit, instead of trusting that the plumbing does what the schematic says.
A simple inline flow meter on each circuit is the tool that ends most of this guessing. The good ones let you read flow rate, see whether the flow is turbulent or laminar, and check inlet and outlet temperature and pressure on each line. With that in hand you can answer the questions that actually matter:
- Is each circuit flowing, or is one plugged?
- Is each circuit turbulent, or just full?
- What’s the temperature rise across each circuit? A large delta means that circuit is doing a lot of work — and may be starved relative to its load.
- Are inlet pressures consistent, or is one line being robbed by a restriction upstream?
A practical baseline routine, no exotic equipment required:
- With the tool running, confirm flow on every circuit individually — never just at the manifold.
- Check the temperature rise (inlet to outlet) on each circuit; flag any circuit that’s much hotter than its neighbors.
- Confirm each circuit is in the turbulent range; fix the laminar ones by clearing restriction or raising flow.
- Put cooling on a preventive schedule: descale lines, service disconnects, and verify flow at intervals — not after the parts start warping.
Choosing the unit that feeds the tool
Flow and turbulence are half the picture; the other half is what’s supplying the coolant and at what temperature. The right temperature-control equipment depends on the target mold temperature, and using the wrong class either can’t hold the setpoint or wastes energy fighting itself. A practical selection guide by target tool temperature:
| Target mold temperature | Appropriate unit |
|---|---|
| ~8–20 °C | Chiller |
| ~20–35 °C | Ambient / tower water |
| ~35–85 °C | Water-type temperature control unit (TCU) |
| ~85–160 °C | Oil-type TCU |
(Ranges are practical guidance, not hard cutoffs — overlap exists and the resin and part drive the actual target.)
Once you’re running a TCU, a few operating rules prevent the expensive mistakes:
- Never mix water and oil units on the same tool, and purge water from the mold and lines before connecting an oil unit. Oil units use heat-transfer oil and the appropriate high-temp hose — water left in the circuit flashes to steam at oil-unit temperatures.
- Connect supply and return in the correct direction, and check the medium level: too low alarms, too high overflows or won’t start.
- Respect the shutdown order. Kill the heat and let the tool cool below roughly 60 °C before powering off and draining — and never close the circuit valves while the pump or heater is still running.
- Start a chiller in order: water valves first, then pump, then compressor; stop in reverse. A compressor that keeps tripping on overload is telling you the cooling demand is too high or the setpoint is too low.
These aren’t just equipment-care notes. A unit that’s mismatched to the target temperature, or run in the wrong sequence, gives you an unstable tool temperature — which lands right back as warp, dimensional drift, and a cycle that won’t come down.
Why this pays for itself
Cooling improvements hit two things at once that the plant actually cares about. They shorten the longest part of the cycle, which is pure throughput. And they make tool temperature even, which is what holds dimensions and kills warp at the source instead of fighting it downstream with fixtures, annealing, and slower handling. A tool that’s fed and balanced is also a tool that holds a tighter, more repeatable process window — which is the whole point of qualifying it in the first place.
For deeper technical references on conformal cooling, channel design, and thermal balancing of tools, MoldMaking Technology (moldmakingtechnology.com) is a strong vendor-neutral source.
FAQs
How do I know if my mold is being cooled enough?
Don’t infer it from the controller setpoint — measure it. Put a flow meter on each circuit individually and confirm three things: that every line is actually flowing, that each line is in the turbulent range rather than laminar, and that the temperature rise across each circuit is reasonable and balanced. A tool can show a correct setpoint at the controller while one circuit is plugged and another is running laminar, and you’ll never see it from the press.
What’s the difference between laminar and turbulent flow, and why does it matter?
Laminar flow moves in smooth layers, so the water against the channel wall forms an insulating boundary layer and the coolant in the center barely contacts the steel — poor heat transfer even though the line is full. Turbulent flow churns the water so it keeps mixing against the wall, and heat transfer climbs sharply. A circuit can be completely full of moving water and still cool badly if that flow is laminar. The practical target is to keep each circuit turbulent (Reynolds number roughly 4,000 or above).
Why does one area of my part warp or stick while the rest is fine?
That pattern almost always points to uneven cooling — a single circuit that’s blocked, scaled, undersized, or starved at the end of a series run, leaving one region of the tool hotter than the rest. The part shrinks unevenly and warps, or stays soft locally and sticks. Confirm flow and temperature rise on each circuit separately; the offending line usually stands out immediately once you stop measuring only at the manifold.
Can undersized water fittings really cost me cycle time?
Yes, and it’s one of the most common hidden causes. A water circuit is only as large as its smallest restriction. A big tool fed through fittings, hoses, or quick-disconnects sized for a smaller standard can’t reach the flow it needs to stay turbulent, no matter how much capacity the pump or chiller has. The result is longer cooling time and uneven temperature — both of which read as “that’s just how the part runs” until someone sizes the plumbing to the tool.