Cooling Water Systems for EPP and EPS Moulding Plants: Tower Sizing, Pumps and Water Treatment

2026-10-23Industry News29 min read

Quick answer

Size and run the plant-side cooling water loop for an EPP or EPS moulding line: heat load, cooling tower approach, pump head, water treatment and balancing.

Quick answer: The plant-side cooling loop is the half of the cooling story that does not live inside the tool. Its job is to take the heat that the mould water jacket has already absorbed and reject it to atmosphere, and it is governed by three numbers: the heat load in kW, the supply water temperature you can actually hold on the hottest afternoon of the year, and the condition of the water itself. Size the tower on measured heat load and the local design wet-bulb — not on the installed motor power of the machines — size the pump on the pressure drop of the whole circuit rather than on pipe diameter, and treat water chemistry as a routine maintenance item rather than a commissioning task. Get those three right and the tool-side cooling described in our article on where EPP and EPS cycle time is really spent behaves the way it was designed to.

Customised EPS/EPP mould with its water jacket connections, the load served by the plant cooling water loop — Every cooling channel in the tool ends at a pair of connections on the mould. Whatever the plant loop delivers — temperature, flow and cleanliness — arrives at those connections.

Key takeaways

Cooling is two loops sharing one word. The tool-side loop moves heat out of the steel; the plant-side loop moves it out of the building. This article is the plant side. Improving tool-side cooling is a separate exercise.
Measure the load, do not add up nameplates. Heat rejected is measured with a flow meter and two thermometers: Q = flow × 4.18 × ΔT. Installed motor kW overstates duty badly on a foam line.
Wet-bulb, not dry-bulb, sets your cold water temperature — and the gap between them, the approach, is what you are really buying when you pay for a bigger tower.
Nominal tons are rated at a standard rating point. A tower selected without stating the design wet-bulb, range and cold-water temperature is not a selection at all.
Pump head comes from the circuit, not the pipe size. Budget pressure drop across tool, heat exchanger, piping, riser, strainers and control valve, then add margin — do not guess.
Scale is a cycle-time problem wearing a plumbing disguise. A thin scale film inside a cooling channel insulates exactly where the tool needs to conduct, and cycle time drifts up months later.
Check NPSH on hot return water. Vapour pressure roughly doubles between 20 °C and 30 °C and doubles again by 40 °C, so a pump that was fine on cold commissioning water can cavitate in summer.
A recirculating tower is a water-treatment programme that happens to include a fan and a pump. Evaporation concentrates everything the make-up water carried in.
Cooling towers are a legionellosis risk and are covered by risk-management standards such as ASHRAE Standard 188 and ASHRAE Guideline 12. A written plan is not optional.

Two loops, one word: what the plant-side circuit actually does

Ask two people on a foam plant what "cooling" means and you will get two different answers, and both are right. Inside the tool, cooling water runs through drilled channels or a jacket in the steam chest and pulls heat out of the aluminium so the part can be ejected without collapsing or distorting. That is the tool side, and it is where wall-thickness uniformity, channel layout and alloy conductivity do their work.

Outside the tool, something has to take that heat and throw it away. That is the plant side: a basin or reservoir, a pump, a distribution header to each machine, a return header, and a cooling tower or fluid cooler that evaporates a small fraction of the water to reject the rest of the heat to the air. This article is entirely about that second loop.

Keep the boundary clean. Tool-side cooling — channel diameter and spacing, alloy thermal conductivity, bidirectional steam, vacuum-assisted cooling, and how cooling dominates the moulding cycle — is covered in the cycle-time article. Vacuum pump seal water, including the way seal-water temperature sets the achievable vacuum level, belongs to the vacuum system article. Compressor aftercooling and air-system condensate belong to the compressed air article. What follows assumes those are handled and deals only with the water loop that serves the mould circuit and the plant heat exchangers.

Step 1: measure the heat load before you buy anything

The single most common sizing error on a new installation is adding up installed motor power and buying a tower to match. On a bead foam line that overstates the duty substantially: a large share of the electrical input never becomes heat in the cooling water, and the steam heat that does arrive is intermittent rather than continuous. The reliable way to get the number is to measure it on an existing line, or to have the machine builder state the cooling duty per cycle in writing for a new one.

Where a line is already running, the load is a five-minute job with instruments you should be installing anyway:

Q (kW) = flow (L/s) × 4.18 × ΔT (K)
where ΔT is return temperature minus supply temperature, and 4.18 kJ/(kg·K) is the specific heat capacity of water.

Worked example. A four-machine cell has a common supply header at 30 °C and a measured return of 35 °C, and a flow meter on the pump discharge reads 16.7 L/s. The load is 16.7 × 4.18 × 5 ≈ 349 kW. That is the number the tower has to reject, and the number the pump has to serve.

Where the heat in a bead foam plant's cooling water actually comes from, and how to think about each contribution
Source	How it enters the loop	Shape of the load	Practical note
Mould / steam chest cooling	Steam condenses in the chest, the steel soaks up the heat, water carries it away during the cooling phase	Sharply cyclic, one pulse per moulding cycle	The dominant term. It is a pulse, so the header and reservoir damp it — size the tower on the average, not the instantaneous spike
Vacuum pump seal water	Liquid-ring pumps pass their service water through a separator; it arrives warm	Steady while the pump runs	Often piped to the same return header. See the vacuum article for why its temperature matters
Hydraulic and drive cooling	Hydraulic power packs, hydraulic motors and Variable-frequency drive cabinets on water-cooled models	Steady and modest	Low-grade heat; can usually be served by the same loop if supply temperature is within the manufacturer's limit
Air compressor aftercoolers	Compressor package cooling, where water-cooled	Steady, tracks compressor load	Confirm the compressor's maximum permitted inlet water temperature before connecting it — this is where a hot summer supply temperature bites first
Solar gain and pipe losses	Outdoor headers in sun; long runs	Slowly varying	Insulate outdoor lines. It is a small term in kW but it lands directly on supply temperature

Apply diversity, then margin — in that order, and separately. Four machines do not all peak on the same second, so a diversity factor is legitimate and your machine builder can usually tell you what is normal for the cell. Margin is a different number with a different purpose: it covers fouling, a future extra machine and a hotter-than-design summer. Adding 10–20% margin after diversity is normal; using margin as a substitute for measuring the load is not. A grossly oversized tower tends to cycle its fan more and can hold a wetter, dirtier fill than a correctly sized one.

Step 2: tower sizing is a wet-bulb problem, not a tonnage problem

A cooling tower cools water by evaporating a small part of it. Evaporation is driven by how much more moisture the air can still hold, which is a function of wet-bulb temperature — not the air temperature on the weather report. This is why a tower that performs beautifully in April falls short in August, and why the three terms below matter more than any headline capacity figure.

The three numbers that define what a cooling tower will actually deliver
Term	Definition	Typical planning value	Why it matters
Range	Return temperature minus supply (cold) temperature — how much the tower cools the water by	4–6 K, commonly about 5 K	Range sets the flow rate: for a fixed load, halving the range doubles the flow
Approach	Supply (cold) water temperature minus the ambient wet-bulb	About 3–5 K for a typical film-fill tower; down to 2–3 K with more fill and more fan power	This is the expensive number. Every kelvin of closer approach costs disproportionately more fill area and air flow
Design wet-bulb	The site's summer wet-bulb, usually taken at a conservative percentile rather than the average	Varies by climate; many temperate coastal industrial sites design in the mid-20s °C	Everything else follows from it. Getting it wrong invalidates the whole selection

Cold water temperature ≈ design wet-bulb + approach
Flow (L/s) = Q (kW) ÷ [4.18 × range (K)]

Continuing the example. With a design wet-bulb of 26 °C and a 4 K approach, the tower can deliver about 30 °C supply water. At a 5 K range the return sits near 35 °C, and to reject 349 kW the flow must be 349 ÷ (4.18 × 5) ≈ 16.7 L/s, or roughly 60 m³/h. In refrigeration terms that is 349 ÷ 3.517 ≈ 99 tons, so a nominal 100 TR machine is the starting point — provided it is rated at your design conditions.

Nominal tons are quoted at a standard rating point, and yours is probably not it. Thermal performance certification programmes such as CTI STD-201 define a nominal rating point (commonly quoted as 35 °C hot water, 29.4 °C cold water and 25.6 °C wet bulb) precisely so that different towers can be compared. A selection that says "100 TR" without stating design wet-bulb, range and cold-water temperature is not a selection. When you send an enquiry, state all four: heat load in kW, flow, supply temperature required, and design wet-bulb. A supplier cannot quote against the first two alone.

Open tower or closed circuit?

This is the decision with the longest-lived consequences, and for a foam plant it usually turns on one fact: the cooling passages inside a mould are not large.

Open evaporative tower versus closed-circuit fluid cooler for a bead foam plant
Option	How it works	Advantages	What you accept
Open tower + plate heat exchanger	Tower water circulates in an open loop; a plate heat exchanger separates it from a clean secondary loop feeding the tools	Best thermal performance per unit cost; tower fill and basin stay accessible; the clean loop stays clean	Two loops, two pumps, two sets of instruments. The heat exchanger costs you a temperature penalty of a few kelvin
Open tower, single loop	Tower water goes straight to the moulds	Cheapest and simplest; no heat-exchanger temperature penalty	Everything airborne — dust, pollen, insect screens, fill debris, biological growth — is delivered straight into the narrowest passages in the plant
Closed-circuit fluid cooler	Process water stays inside a coil; spray water evaporates over the outside of it	Process water never contacts air; lowest fouling risk at the tool; freeze protection is easier	Lower efficiency than an open tower at the same footprint, because heat crosses a coil wall twice; higher capital cost

What we recommend for a bead foam tool loop. Mould cooling channels are commonly drilled in the range of roughly 6–12 mm, and a tool is expensive, slow to replace and central to your cycle time. Spending money to keep debris out of those channels is cheap insurance. In practice that means either a closed-circuit cooler or an open tower with a plate heat exchanger isolating a clean secondary loop — plus continuous side-stream filtration. A single open loop piped directly to the moulds is the option we would argue against for anything but a small, low-value installation.

Step 3: pumps — head comes from the circuit, not the pipe size

Pumps are selected on two numbers: the flow you already calculated, and the differential head the circuit demands at that flow. Head is not a property of the pipe diameter; it is the sum of every pressure drop in the path, and the tool itself is usually the largest single item in that list.

Typical pressure-drop budget for a plant cooling water circuit serving bead foam tooling. Treat these as planning ranges, not as design values
Element	Typical pressure drop	Comment
Mould / tool circuit	2–4 bar	Usually the dominant term. Depends entirely on channel length, diameter and number of passes — ask the toolmaker for the figure at the design flow
Plate heat exchanger	0.3–0.8 bar	Rises as it fouls; specify a pressure-drop limit to the supplier and monitor it
Piping, fittings and valves	0.5–1.5 bar	Long runs and undersized headers are the usual culprits. Velocity around 1.5–3 m/s is a common design range
Strainers and filters	0.1–0.5 bar clean	Rises as it blocks — install pressure gauges either side so you can see it happen
Static lift to tower distribution	0.3–0.6 bar	The vertical distance from basin water level to the distribution deck, not to the top of the casing
Control valve allowance	0.3–0.5 bar	Leave the valve something to work with, or it will hunt
Total differential	About 4–7 bar (40–70 m head)	Plus margin for fouling; then select the nearest standard pump above the duty point

Cavitation is a summer problem

A pump cavitates when the pressure at the impeller eye falls below the vapour pressure of the liquid, and vapour pressure rises steeply with temperature: roughly 2.3 kPa at 20 °C, 4.2 kPa at 30 °C and 7.4 kPa at 40 °C. A pump commissioned on cold mains water in winter can therefore start cavitating in August, for no reason other than the return water being warmer. The check is the standard one:

NPSH_available = (P_surface − P_vapour) ÷ ρg ± static suction head − suction friction losses
Require NPSH_a ≥ NPSH_required + 0.5 m at the highest water temperature the system will ever see, not at today's.

Two layout decisions do most of the work here: keep the pump below the basin water level so the suction is flooded rather than lifting, and keep the suction line short, large and free of elbows right at the pump. A blocked suction strainer produces exactly the same symptom as a badly sited pump, so check it first — it costs nothing.

Drive and control

Cooling demand on a foam line moves with production, shift pattern and season, so a fixed-speed pump running at full duty year-round is paying for capacity it rarely needs. A variable-speed drive on the pump, controlled from differential pressure at the most remote consumer, is usually the cheapest energy measure on the whole utility system — and unlike a throttling valve it removes energy rather than converting it to noise. Sequencing: match the tower fan to the load too, and remember that the fan is a heat-rejection control while the pump is a flow control. They are not the same knob.

Step 4: water chemistry — what decides whether the tool still cools in year three

An evaporative tower concentrates. As pure water leaves as vapour, everything dissolved in the make-up water stays behind and accumulates, so the circulating water becomes progressively saltier, harder and more alkaline than the water you put in. Left alone, one of three things happens, and all three eventually show up as a longer cycle.

The three ways recirculating cooling water fails, what it does to the plant, and the parameter that warns you
Failure mode	Mechanism	What it does to a foam line	Leading indicator
Scale	Calcium and magnesium salts precipitate onto the hottest surface they can find — usually inside the tool	A thin film insulates exactly where the tool must conduct. Cycle time creeps up over months and the tool responds poorly to extra cooling	Hardness and alkalinity as CaCO₃; Langelier Saturation Index; rising tool ΔT at unchanged flow
Corrosion	Dissolved oxygen, low pH and high chloride attack carbon steel, then copper, then stainless	Headers and heat exchanger plates thin out; rust particles travel and block strainers and channels	Iron and copper in the water; visual corrosion coupons; pH trend
Biological growth	Biofilm, algae and slime colonise warm, sunlit, nutrient-bearing water	Fill and strainers foul, heat transfer drops, and the organism risk below becomes real	Visual slime; biocide residual; plate counts if you run them

Control ranges to ask your treatment supplier about

The exact targets depend on your make-up water and your metallurgy, so these are the parameters to establish with a water treatment specialist rather than numbers to copy into a specification. They are the ones worth having on a monthly report.

Parameters commonly tracked on an open recirculating cooling system. Ranges vary with make-up water, metallurgy and treatment programme — agree them with your supplier rather than adopting them blindly
Parameter	Commonly used control range	Why it is tracked
pH	Commonly controlled around 7.0–9.0, often 7.5–8.8 with a chemical programme	Sits between corrosion at low pH and scaling at high pH; also affects biocide effectiveness
Conductivity	Controlled indirectly, by blowdown against a cycles-of-concentration setpoint	The cheapest single indicator of how concentrated the loop has become
Calcium hardness as CaCO₃	Frequently held below roughly 400–600 mg/L in a treated open system	The direct driver of scale at the hot surface
Total alkalinity as CaCO₃	Often controlled below about 500 mg/L, programme-dependent	Feeds the saturation index and the acid demand
Chloride	Commonly limited to well under 1000 mg/L where stainless is present	Pitting and stress-corrosion risk for stainless plate packs
Langelier Saturation Index	Many programmes target a slightly positive value, roughly +0.5 to +1.5	Slightly scale-forming water lays down a thin protective film; strongly scaling water plugs the fill
Turbidity / suspended solids	Drives the side-stream filtration duty	What you are physically trying to keep out of the mould channels
Biocide residual	Programme-specific, usually oxidising biocide with an alternating non-oxidising partner	Controls biofilm and the organism risk below

Cycles of concentration and blowdown

Blowdown is deliberate, controlled wasting of concentrated water, and it is the lever that holds the whole chemistry steady. Both quantities are defined simply:

Cycles of concentration ≈ conductivity of circulating water ÷ conductivity of make-up water
Blowdown ≈ Evaporation ÷ (cycles of concentration − 1)
Make-up = Evaporation + Blowdown + Drift (+ windage and any leaks)

Worked example. Take the 60 m³/h loop with a 5 K range. A common planning rule puts evaporation at roughly 1% of circulation for every 5–6 K of range, so call it about 0.6 m³/h. At 4 cycles of concentration, blowdown is 0.6 ÷ (4 − 1) ≈ 0.2 m³/h, and total make-up lands near 0.8–0.9 m³/h. That is your water bill, and it is also the number that tells you whether a conductivity-controlled automatic blowdown valve will pay for itself — it usually does, because manual blowdown is nearly always over-conservative.

Side-stream filtration is not optional on a foam plant. A continuous filter taking 5–10% of circulation flow and cleaning itself automatically removes the suspended solids that would otherwise find the narrowest passage in the system. Size the filter media to what you are protecting: if the smallest cooling channel in the tool is 8 mm, a screen that passes 2 mm debris is still doing its job, but a strainer sized for a 50 mm header is not.

Legionella: a documented risk-management obligation

Cooling towers are a well-established source of legionellosis outbreaks, which is why the risk is addressed by formal management standards rather than by good intentions. In North America the relevant documents are ASHRAE Standard 188 (Legionellosis: Risk Management for Building Water Systems) and ASHRAE Guideline 12 (Managing the Risk of Legionellosis Associated with Building Water Systems); many other jurisdictions have equivalent guidance or a legal duty to manage the risk. The engineering content of such a plan is mostly common sense:

Avoid the growth band. Legionella multiplies in warm water — commonly quoted as roughly 20–50 °C, fastest around 25–42 °C — and does not grow much above about 60 °C. A tower basin sitting in that band with nutrients and biofilm present is exactly the wrong environment.
Control biofilm and nutrients through the treatment programme above, not through occasional shock dosing alone.
Hold water chemistry within the agreed control range and record it. The record is the evidence that the plan is being run.
Fit and maintain drift eliminators. Drift is the pathway from the tower to people.
Site the tower sensibly — away from fresh-air intakes, openable windows and pedestrian routes — and record the assessment.
Assign a named responsible person and a written programme. Standards of this kind are about management systems as much as about hardware.

Step 5: distribution, balancing and what to instrument

Most "the cooling is not working" complaints on a running plant are not capacity problems. They are distribution problems: one machine is starved while another floods, and the tower looks guilty because the return temperature is high.

What to install at each machine connection, and what each item tells you
Item	Where	What it gives you
Isolation valves	Supply and return at every take-off	The ability to service one machine without draining the plant
Balance / commissioning valve with flow measurement	Return line at every take-off	Turns balancing from trial and error into a recorded setting
Two temperature gauges or pockets	Supply and return at every take-off	The ΔT across the machine. Rising ΔT at unchanged setting means falling flow — the single most useful diagnostic on the loop
Y-strainer with differential pressure taps	Supply, upstream of the tool	Lets you see blockage develop instead of discovering it as a hot tool
Union or quick connection	At the tool	Speeds up mould change and tool removal; see the quick tool change article on standardising machine interfaces
Automatic air vent	High points; drain valve at low points	Air locks in a cooling channel look exactly like scale: local hot spot, no flow

Design the header as a ring, not a tree. A looped header fed from more than one direction keeps pressure more even along its length, so the last machine on the line is not penalised for being last. Insulate outdoor runs, protect the basin and exposed piping against freezing where winter temperatures demand it, and slope lines so they drain to a point you can actually reach.

Symptoms, causes and the first thing to check

Troubleshooting the plant cooling loop: what you see, what usually causes it, and the cheapest check to run first
Symptom	Likely cause	Check first (cheapest to most involved)
Supply temperature climbs every afternoon, recovers overnight	Tower is at its wet-bulb limit; or fill is fouled; or airflow is obstructed	Compare supply temperature against a wet-bulb reading taken at the tower. If the gap matches the design approach, the tower is doing its job and the specification was too optimistic
One machine consistently hotter than its neighbours	Flow imbalance, partly closed valve, blocked strainer, or air lock	Read the supply and return temperatures at that machine. A ΔT wider than its neighbours means too little flow, not too little cooling capacity
Cycle time lengthens gradually over months, partly recovers after cleaning	Scale or biofilm inside the tool channels	Trend the tool ΔT at fixed flow, and pull the water chemistry records for the same period. Clean the tool and re-baseline — see mould maintenance and cleaning
Pump is noisy, discharge pressure unstable	Cavitation from blocked suction strainer, high return temperature, or suction lift	Suction strainer differential pressure, then return water temperature, then re-check NPSH at that temperature
Make-up water consumption suddenly high	Blowdown valve passing, float valve failed, or a leak in the basin or piping	Watch the make-up meter with all machines stopped — if it still moves, it is the tower or the pipework, not the process
White deposit at tool connections and on the hottest fittings	Hardness or alkalinity out of range; cycles of concentration too high	Conductivity and hardness against the agreed control range; verify blowdown is actually operating
Heat exchanger differential pressure rising	Fouling on the tower side, or debris	Compare against the commissioning baseline. If it is seasonal, suspect biology; if it is monotonic, suspect scale
Supply temperature fine, but part still ejects soft or distorted	Not a plant-side problem — this is tool-side cooling or the cooling phase of the cycle	Go to the cycle-time article and the moulding defects article. Do not fix the tower for a tool problem

Energy and lifecycle: where the money actually goes

A cooling water system is dominated by two running costs: pump and fan electricity, and water. Both respond to decisions made during specification, which is worth remembering when a cheaper tower looks attractive.

Fan and pump power follow the affinity laws, so a modest reduction in flow produces a much larger reduction in power — which is exactly why variable-speed drives pay back quickly on a plant whose load moves with the production schedule. Water cost is governed by cycles of concentration: running at 3 cycles instead of 5 saves electricity on pumping make-up and treatment chemical, but costs significantly more water and blowdown. The optimum is an economic calculation with your local water and effluent tariffs, not a technical constant.

Then there is the cost nobody budgets: the cycle time lost to a scaled tool. A few percent of cooling-phase time on every cycle, on every machine, every hour, is worth more over a year than the difference between two tower quotations. That is the argument for filtration and water treatment, and it is the same argument as trending part weight with SPC — the point is to see the drift while it is still cheap to correct.

Eight steps to specifying or upgrading a plant cooling loop

Measure the load. Flow meter and two thermometers on the existing return header; or get written per-cycle cooling duty from the machine builder. Do not add up motor nameplates.
Apply diversity, then margin. They are two different numbers with two different purposes. Record both.
Fix the design wet-bulb for your site at a conservative summer percentile. Everything downstream depends on it.
Choose range and approach deliberately. Range sets flow and pump size; approach sets cold-water temperature and tower cost. Write both into the enquiry.
Decide open-and-clean or closed-circuit before you price anything, remembering how narrow a mould cooling channel is.
Build the pump head budget line by line — tool, heat exchanger, pipework, strainers, static lift, control valve — then verify NPSH at the highest water temperature the loop will ever see.
Appoint the water treatment supplier during design, not after start-up, and agree the control ranges, the cycles-of-concentration setpoint and the test schedule in writing.
Commission with instruments and record baselines: flow and ΔT per machine, heat exchanger differential pressure, pump discharge pressure, and water chemistry. Trend them monthly. Those baselines are what make the troubleshooting table above usable.

Keep utility projects separate from cycle-time projects. A cooling upgrade and a cycle-time reduction both improve output per hour, and doing them together makes the gain impossible to attribute. Run cycle reduction as its own exercise following the sequence in where EPP and EPS cycle time is really spent, and treat the plant loop as a separate, verifiable improvement.

Frequently asked questions

Why does my cooling water supply temperature rise in summer when the tower is running correctly?

Because an evaporative tower cannot cool water below the ambient wet-bulb temperature, only towards it, and the gap it can achieve is its approach. As the summer wet-bulb rises, the achievable cold water temperature rises with it even with the fan at full speed and the fill clean. Compare your supply temperature against a wet-bulb reading taken at the tower: if the gap matches the design approach, the tower is performing as sold and the specification was optimistic. If the gap has widened since commissioning, suspect fouled fill, blocked distribution nozzles or restricted airflow.

What is the difference between range and approach, and which one costs money?

Range is how much the tower cools the water — return temperature minus supply temperature — and it sets the flow rate. Approach is how close the supply temperature gets to the wet-bulb, and it sets the tower's size and cost. For a fixed heat load, choosing a wider range lets you pump less water, and choosing a closer approach buys a colder supply temperature at a rapidly increasing price in fill area and fan power. Approach is the expensive number.

How much water does a cooling tower actually consume?

Make-up water equals evaporation plus blowdown plus drift, plus any leaks. Evaporation is roughly 1% of circulation flow for every 5 to 6 kelvin of range as a planning rule, blowdown is evaporation divided by (cycles of concentration minus one), and drift is small on a tower with good eliminators. A 60 m³/h loop at 5 K range and 4 cycles therefore uses roughly 0.8 to 0.9 m³/h of make-up. Your water treatment supplier can refine this, and a conductivity-controlled automatic blowdown valve usually reduces it compared with manual blowing down.

Should a bead foam plant use an open tower or a closed-circuit cooler?

For a tool loop, prefer either a closed-circuit cooler or an open tower with a plate heat exchanger isolating a clean secondary circuit, together with continuous side-stream filtration. The reason is mechanical rather than thermal: mould cooling channels are commonly drilled in the range of about 6 to 12 mm, and an open tower delivers everything airborne in the plant straight into them. A single open loop piped directly to the moulds is workable on small, low-value installations, but it is the option most likely to cost you cycle time later.

How do I know if the pump is correctly sized?

Calculate the differential head from the circuit rather than from the pipe size: tool circuit, heat exchanger, pipework and fittings, strainers, static lift to the tower distribution deck, and control valve allowance. Typical totals for a foam plant cooling circuit land around 4 to 7 bar, plus margin for fouling. Then select the nearest standard pump at or above that duty point, and separately verify that available NPSH exceeds required NPSH by at least 0.5 m at the highest water temperature the system will ever see.

The pump was fine at start-up but cavitates in summer. Why?

Because vapour pressure climbs steeply with temperature: about 2.3 kPa at 20 °C, 4.2 kPa at 30 °C and 7.4 kPa at 40 °C. A pump checked against cold commissioning water can fall short of its NPSH requirement once the return header runs hot. Before replacing anything, check the suction strainer differential pressure, since a blocked strainer produces the same symptom for a fraction of the cost, then re-run the NPSH calculation at the summer water temperature.

What causes scale inside mould cooling channels, and how do I stop it?

Scale is calcium and magnesium salts precipitating onto the hottest surface available, which in this system is usually the inside of the tool. It is driven by hardness and alkalinity, pH and cycles of concentration, and the Langelier Saturation Index summarises the tendency. Control it by holding cycles of concentration within the agreed band through controlled blowdown, keeping the treatment programme's hardness, alkalinity and pH targets, and filtering the loop continuously. Watch the tool's supply-to-return temperature difference at a fixed flow setting: a rising trend is scale forming, and it will show up as longer cycle time before it shows up as a failure.

How often should cooling water be tested?

That is a question for your water treatment supplier, because the schedule depends on the programme, the make-up water and the system size. As a general pattern, conductivity and pH are cheap enough to check on site frequently, often weekly or continuously, while hardness, alkalinity, chloride, iron and microbiological results come from a periodic laboratory or service report. The important part is less the interval than the record: a trended log is what lets you distinguish a real drift from a single bad reading.

What are the legionellosis obligations for a cooling tower?

Cooling towers are a well-established source of legionellosis, and the risk is addressed by formal management standards rather than by good housekeeping alone. ASHRAE Standard 188 and ASHRAE Guideline 12 are the commonly referenced North American documents, and many other jurisdictions impose an equivalent duty. The engineering content is consistent: keep water out of the growth band of roughly 20 to 50 °C where practicable, control biofilm through a real treatment programme, hold chemistry within agreed limits, maintain drift eliminators, site the tower away from air intakes and pedestrian routes, and run a written plan with a named responsible person and records.

Are variable-speed drives worth fitting to cooling water pumps?

Usually yes, and often ahead of other energy measures. Pump and fan power follow the affinity laws, so a modest reduction in flow produces a much larger reduction in power, and load on a foam line moves with production schedule and season rather than sitting constant. Control the pump from differential pressure at the most remote consumer. Note that the tower fan and the pump are doing different jobs — fan controls heat rejection, pump controls flow — so they should not be slaved to the same signal.

How do I balance flow between machines on the same header?

Instrument each take-off with a flow-measuring balance valve and a pair of temperature gauges, then set each machine to its design flow and record the valve position. After that, the supply-to-return temperature difference across each machine is your ongoing flow indicator: a machine showing a wider difference than its neighbours at the same setting is getting less water than it should. Design the header as a ring rather than a tree so that the last machine on the line is not penalised for being last.

Compressed Air for EPP and EPS Moulding Lines: ISO 8573-1 Purity, Demand and Distribution

2026-10-22

In-Process Quality Control for EPP and EPS Moulding: SPC, Part Weight and Traceability

2026-10-21

How to Cut EPS Foam: Tools, Techniques and Tips

2026-10-20

Need a custom EPP / EPS mold for this application?

Send us your drawing, target density and annual volume — our engineers reply with a tooling proposal, lead time and quotation within 24 hours.

Get a quotation