Energy saving technology for EBM machines has moved from a marketing bullet point to a hard commercial requirement. For most blow molding plants, electricity is now the second-largest controllable cost after resin, and in regions with high tariffs or carbon reporting obligations it is the cost line that grows fastest. A single extrusion blow molding machine running three shifts on HDPE jerry cans can consume more electricity in a year than an entire office building of comparable floor area. Yet the same machine, correctly specified, can deliver identical output on substantially less power.
Apollo, a Wanplas factory based in Zhangjiagang near Shanghai, has spent more than twenty years building automatic extrusion blow molding machines, with over 4,000 sets running in more than 90 countries. Across that installed base, the pattern is consistent: the difference between an efficient machine and an inefficient one is rarely a single breakthrough component. It is the accumulation of a dozen deliberate engineering decisions, each of which shaves a few percentage points off a different subsystem.
This article breaks down exactly where the power goes in an extrusion blow molding machine, then works through each subsystem in turn: the extruder drive, the hydraulic power unit, barrel and die head heating, compressed air, and mold cooling. It explains the measurable difference between fixed-displacement pumps, load-sensing variable pumps, servo-hydraulic units and fully electric actuation. It compares mica, ceramic, infrared halogen and induction heating on a like-for-like basis. It quantifies what blow air recovery actually returns. And it closes with a prioritised retrofit roadmap that any plant can apply to machines already on the floor, whether they carry an Apollo nameplate or one from Bekum, Kautex Maschinenbau, Uniloy or Graham Engineering.
Throughout, energy performance is expressed the way engineers should express it: as specific energy consumption in kWh per kilogram of product, and kWh per thousand pieces, measured at the machine boundary. Relative cost and payback are expressed in qualitative bands and indexed baselines rather than absolute figures, because tariffs, duties and local labour rates vary far too widely for any single number to be useful.
Why Energy Consumption Is Now a Core EBM Specification
Energy consumption is no longer a secondary specification in extrusion blow molding; it is a primary selection criterion that directly determines whether a container line remains competitive over a ten-year service life. Three forces have converged to make this true, and none of them is likely to reverse.
The first force is tariff volatility. Electricity prices in many exporting regions have become far less predictable than they were a decade ago, and blow molding is an electricity-intensive process by nature. Unlike resin cost, which can often be passed through to the customer via indexed contracts, electricity cost is usually absorbed entirely by the converter. A machine specified purely on purchase price and throughput can quietly consume many times its own capital value in electricity over its operating life.
The second force is regulatory and customer-driven carbon accounting. Brand owners in food and beverage, daily chemical and pharmaceutical packaging increasingly require Scope 3 emissions data from their packaging suppliers. That data cannot be produced without per-machine energy measurement. A converter who cannot report kWh per thousand pieces for a given container is at a competitive disadvantage regardless of how good the parts look.
The third force is the maturation of the enabling technology itself. Servo-hydraulic power units, permanent magnet synchronous motors and induction barrel heating were all expensive novelties fifteen years ago. Today they are mainstream, well-supported and available from multiple component suppliers, which has compressed both the price premium and the technical risk. What was once an experimental option is now a defensible engineering default.
The economics of specific energy consumption
The most useful metric in this discussion is specific energy consumption, expressed in kWh per kilogram of finished product. It normalises away machine size, container geometry and shift patterns, allowing a 200ml bottle line and a 220L drum line to be compared on the same axis. For a well-configured HDPE extrusion blow molding machine, total machine-boundary specific energy consumption typically falls between 0.55 and 0.95 kWh/kg. Machines built before the widespread adoption of variable-frequency extruder drives, insulation jackets and closed-loop air systems frequently exceed 1.2 kWh/kg for the same parts.
That gap of roughly 0.4 kWh/kg does not sound dramatic until it is multiplied by annual throughput. A machine converting 400 tonnes of HDPE per year at a 0.4 kWh/kg disadvantage is burning 160,000 additional kilowatt-hours annually for no additional output. Expressed as an index with an efficient modern machine set at a baseline of 100 index points, a poorly configured legacy machine sits at roughly 140 to 165 index points for the same production programme.
Energy performance in extrusion blow molding is not won by one component. It is won by systematically attacking six subsystems, each of which contributes between 3 and 60 percent of total machine load, and by measuring the result rather than assuming it.
Where the Power Actually Goes: Energy Breakdown of an EBM Machine
In a typical single-station extrusion blow molding machine producing 5L HDPE jerry cans, the extruder drive consumes the largest single share of total power, generally between 45 and 60 percent, followed by the hydraulic power unit, barrel and die head heating, compressed air, and mold cooling. Understanding this distribution is the prerequisite for any rational energy programme, because effort spent optimising a subsystem that represents 4 percent of load can never compete with a modest improvement to a subsystem representing 50 percent.
The breakdown below reflects a representative configuration: a single-station machine with a 90mm single-screw extruder, L/D 25:1, continuous extrusion with a single-layer die head, hydraulic clamping unit, and machine-mounted deflashing. Throughput is approximately 60 to 80 kg/h of HDPE with a melt flow rate around 0.3 g/10 min at 190°C and 2.16 kg, measured per ASTM D1238. Container weight is roughly 180 to 220 g. Ambient temperature is assumed at 25°C.
Table 1: Energy Breakdown of a 5L HDPE Jerry Can EBM Machine
| Subsystem | Share of total machine power | Load pattern | Primary saving levers |
|---|---|---|---|
| Extruder drive (motor, gearbox, screw) | 45–60% | Continuous, near-constant | IE4/IE5 motor class, barrier screw, grooved feed barrel, melt temperature discipline |
| Hydraulic power unit (clamping, carriage, blow pin, deflashing) | 10–20% | Highly intermittent, long idle and holding phases | Servo-hydraulic pump, load-sensing control, accumulator sizing, fully electric actuation |
| Barrel, adaptor and die head heating | 8–15% | High at start-up, low duty cycle in steady state | Insulation jackets, induction or infrared heating, self-tuning PID, zone consolidation |
| Compressed air (blow air plus pneumatics) | 8–15% | Pulsed, high instantaneous demand | Blow air recovery, leak detection, two-stage pre-blow and main blow, pressure right-sizing |
| Chiller and mold cooling circuit | 6–12% | Continuous with seasonal variation | Raised water set point, free cooling, plate heat exchanger, variable-speed pumps |
| Auxiliaries (crusher, conveyor, loader, dryer) | 3–6% | Intermittent, often left running | Demand-linked interlocks, crusher blade sharpening, right-sized conveying |
Three practical conclusions follow directly from this table. First, any energy programme that ignores the extruder is addressing less than half the problem. Second, the hydraulic power unit is the subsystem with the largest gap between average and peak demand, which is precisely why it responds so strongly to servo control. Third, compressed air and heating together represent a share comparable to the hydraulics, and both are usually far cheaper to improve.
It is also worth noting what this breakdown deliberately excludes. Plant-level loads such as the central compressor room, cooling tower fans, lighting and HVAC are outside the machine boundary. A converter comparing machines from different suppliers must confirm whether quoted consumption figures are measured at the machine terminal or include a share of these utilities, because the difference can exceed 20 percent. EUROMAP 60.1 and 60.2 exist specifically to remove this ambiguity by defining the measurement boundary and reference cycle.
Servo-Hydraulic Drive Technology and the Limits of Pump Efficiency
The hydraulic power unit is the subsystem where the gap between installed power and useful work is widest, which makes it the highest-leverage target per unit of engineering effort. In extrusion blow molding, the clamping unit, carriage traverse, blow pin stroke and deflashing cylinders all demand short bursts of high flow separated by long periods of near-zero demand. A conventional fixed-displacement pump cannot distinguish between those states.
Why a fixed-displacement pump wastes so much
A fixed-displacement pump driven by a constant-speed induction motor delivers the same flow every second the motor is running, regardless of what the machine needs. When the clamping unit is closed and simply holding, or when the machine is waiting for the parison to reach length, that flow has nowhere useful to go. It is dumped across a relief valve, and the entire hydraulic energy is converted into heat in the oil. That heat then has to be removed by an oil cooler, which itself consumes cooling water and pump energy. The plant therefore pays twice: once to generate the wasted flow, and again to remove the heat it creates.
In a typical EBM cycle, useful hydraulic work occupies only a fraction of the total cycle time. The remainder is idle, holding or dwell. This is the structural inefficiency that servo technology eliminates.
Load-sensing variable-displacement pumps
A variable-displacement pump with load-sensing control adjusts its swash plate angle so that delivered flow matches demand and system pressure tracks the load plus a small margin. This removes most of the relief-valve dumping and is a genuine improvement over fixed displacement. However, the prime mover still runs continuously at full speed, so windage, bearing friction, pump internal leakage and the standby pressure margin all continue to consume power even when the machine is doing nothing. Load-sensing systems also respond more slowly to sudden flow demand than a well-tuned servo system, because the swash plate has mechanical inertia.
Servo motor plus gear pump
A servo-hydraulic unit pairs a permanent magnet servo motor with a fixed-displacement gear pump and closes the loop on pressure and flow through the drive itself. When the machine needs flow, the motor accelerates; when it does not, the motor slows to a crawl or stops entirely. During holding phases, the servo drive supplies only the small flow needed to compensate for internal leakage, so measured input power during holding drops to a small fraction of the running value rather than remaining at full pump load.
Across a complete EBM cycle, the resulting reduction in hydraulic power unit consumption typically falls between 15 and 40 percent compared with a fixed-displacement baseline. Where the machine falls in that range depends almost entirely on duty cycle. A fast-cycling small-container machine with short dwell periods sits near the lower bound because the pump is genuinely busy most of the time. A large accumulator head machine producing 200L drums, where the clamping unit holds for an extended cooling period every cycle, sits near the upper bound.
Servo-hydraulic systems bring three secondary benefits that are often worth as much as the electricity saving. Oil temperature runs markedly cooler, which extends seal and hose life and reduces the load on the oil cooling circuit. Noise at the machine drops noticeably, which matters for workplace compliance. And pressure and flow repeatability improves, which stabilises parison programming and wall thickness distribution.
Fully electric extrusion blow molding
Fully electric machines replace the hydraulic power unit entirely with servo-driven ball screws and toggle mechanisms for clamping, carriage and blow pin motion. With no oil, no relief valves and no oil cooler, standby losses approach zero and the drive train converts electrical input into mechanical work with the highest efficiency of any architecture discussed here. Apollo’s Fully Electric Series covers containers from 200ML to 20L and is specified most often for pharmaceutical, medical and high-purity daily chemical applications where oil contamination risk is unacceptable.
The applicable boundary matters. Fully electric actuation is most compelling where cycles are short and frequent, annual running hours are high, container sizes are small to medium, and cleanliness requirements are strict. It becomes progressively less practical as container volume rises, because the clamping force and stroke length required for large industrial drums push ball screw and toggle sizing into a region where hydraulic actuation remains more compact, more robust and more economical. For Apollo’s ABLD Series covering 20L to 1500L, servo-hydraulic remains the recommended architecture.
Table 2: Hydraulic and Drive Architecture Comparison for EBM Machines
| Drive architecture | Energy consumption class | Response speed | Maintenance demand | Oil cooling load | Relative capital cost |
|---|---|---|---|---|---|
| Fixed-displacement pump, constant-speed motor | Highest consumption (index 100 baseline) | Medium | High | High | Low |
| Variable-displacement pump with load sensing | Moderate (index 78–88) | Medium | Medium to High | Medium | Medium |
| Servo motor with fixed gear pump (servo-hydraulic) | Low (index 60–85) | Fast | Medium | Low | High |
| Dual servo pump with independent parison and clamp circuits | Very low (index 55–75) | Very fast | Medium | Low | Very High |
| Fully electric (servo ball screw and toggle) | Lowest (index 50–70) | Very fast | Low | None | Premium |
The index values above refer to hydraulic and actuation energy only, benchmarked against a fixed-displacement baseline of 100 index points on a representative 5L jerry can cycle. They should not be read as whole-machine figures. Because the hydraulic power unit represents 10 to 20 percent of total machine power, a shift from fixed displacement to servo-hydraulic typically yields a 3 to 8 percent reduction in total machine consumption, which is meaningful but must be understood in proportion.
Extruder Energy Efficiency: Motor Class, Screw Geometry and kWh/kg
The extruder is the dominant energy consumer in every continuous extrusion blow molding machine, so a one-percent improvement here is worth more than a ten-percent improvement in the crusher circuit. Extruder efficiency is determined by three interacting factors: the electrical efficiency of the motor and drive, the mechanical efficiency of the gearbox, and the thermodynamic efficiency with which the screw converts shaft work into melt at the required temperature and homogeneity.
Motor efficiency classes under IEC 60034-30-1
IEC 60034-30-1 defines efficiency classes IE1 through IE5 for line-operated AC motors, and this classification is the single clearest specification lever available to a machine buyer. The difference between an IE2 motor and an IE4 or IE5 motor of the same rating is a few percentage points of efficiency, but those points apply continuously to the largest load on the machine, twenty-four hours a day.
A standard AC induction motor driven by a variable-frequency inverter remains the most common extruder drive and, when specified at IE3, is a perfectly respectable choice. The important detail is that induction motor efficiency degrades at partial load and at reduced speed. Extruders very often run at 50 to 75 percent of rated speed, precisely the region where induction motor efficiency curves begin to fall away, and where the motor power factor deteriorates.
A permanent magnet synchronous motor, or PMSM, behaves differently. Because the rotor field is provided by magnets rather than induced current, rotor losses are dramatically lower and the efficiency curve stays flat across a much wider speed range. PMSM extruder drives commonly reach IE4 and, in larger frame sizes, IE5. They also run cooler, which reduces motor fan losses and extends bearing and insulation life. For a direct-drive configuration, a PMSM can sometimes eliminate the gearbox altogether, removing another two to four percent of transmission loss.
Screw geometry and grooved feed barrels
Screw design determines how much of the shaft work becomes useful melt and how much becomes unnecessary shear heat that must later be removed by barrel cooling fans. A conventional single-flighted metering screw relies on drag flow and shear to melt the polymer. A barrier screw physically separates the solid bed from the melt pool with a secondary flight, which melts the polymer more uniformly at lower shear rates. For HDPE and PP, a well-designed barrier screw with a mixing section typically reduces specific energy consumption by a measurable margin while simultaneously improving melt homogeneity and reducing melt temperature by several degrees.
A grooved feed barrel changes the physics at the other end of the screw. Axial grooves machined into the intensely cooled feed zone dramatically increase solids conveying efficiency, so throughput becomes far less sensitive to head pressure and the screw can deliver the same output at lower speed. Lower screw speed means lower shear heat generation, lower melt temperature, less downstream cooling demand and lower specific energy consumption. Grooved feed is particularly valuable when running high-molecular-weight HDPE with low melt flow rate, which is exactly the material class used for fuel tanks, chemical drums and jerry cans.
L/D ratio: 24:1 versus 30:1
Length-to-diameter ratio is often discussed as if longer is automatically better. In energy terms the relationship is not monotonic. A 24:1 barrel is shorter, has less surface area to heat, and imposes less total viscous drag, so at moderate throughput it can be the more efficient choice. A 30:1 barrel provides more residence time and better melting and mixing capability, which allows higher throughput per revolution and better handling of regrind and masterbatch. However, the additional length also adds surface heat loss, additional torque demand and a longer thermal path.
The practical guidance Apollo applies is straightforward. For single-material HDPE and PP containers with modest regrind content, 24:1 to 25:1 delivers the lowest specific energy consumption. Where the recipe includes significant regrind, colour masterbatch, or where multi-layer co-extrusion is involved, 28:1 to 30:1 is justified because the alternative, pushing a short screw harder, generates more shear heat than the extra barrel length costs.
Table 3: Extruder Configuration and Specific Energy Consumption on HDPE
| Extruder configuration | Typical specific energy consumption (kWh/kg) | Melt temperature tendency | Best suited to |
|---|---|---|---|
| Smooth-bore barrel, conventional metering screw, IE2 induction motor | 0.40–0.48 | High, shear-driven | Legacy machines, low-duty operations |
| Smooth-bore barrel, barrier screw with mixing section, IE3 inverter drive | 0.33–0.40 | Moderate | General-purpose HDPE and PP containers |
| Grooved feed barrel, barrier screw, IE3 inverter drive | 0.29–0.36 | Low to moderate | High-molecular-weight HDPE, low MFR grades |
| Grooved feed barrel, barrier screw, PMSM at IE4/IE5 | 0.25–0.32 | Low | High-duty three-shift operation, energy-priority projects |
| As above with direct drive (gearbox eliminated) | 0.24–0.30 | Low | Selected machine sizes where torque and speed match directly |
These figures represent extruder specific energy consumption on HDPE only, excluding hydraulics, compressed air and cooling. They assume steady-state operation at 70 to 85 percent of rated screw speed with correctly matched melt temperature profiles. Different resins shift the baseline materially: PP generally sits slightly below HDPE, rigid PVC requires low shear and careful heat management, PC and PA demand higher melt temperatures and therefore higher energy input, and PETG sits between the polyolefins and the engineering resins.
One frequently overlooked lever is melt temperature discipline. Every additional 10°C of unnecessary melt temperature must first be generated by motor torque and heater power, then removed again by mold cooling, lengthening cycle time in the process. Excess melt temperature is therefore penalised three times over. Apollo’s commissioning engineers routinely find legacy machines running 15 to 25°C hotter than the resin requires, purely because a temperature profile was copied from an unrelated job years earlier and never revisited.
Barrel and Die Head Heating: Heater Technology, Insulation and Zone Control
Barrel and die head heating typically represents 8 to 15 percent of total machine power, and it is the subsystem where the cheapest energy savings in the entire plant are usually found. In steady-state operation, most of the heat that keeps the melt at temperature is actually generated by viscous shear from the screw, not by the heaters. The heaters’ real steady-state job is to replace heat lost to the surroundings, and much of that loss is avoidable.
Heater technologies compared
Mica band heaters are the traditional choice: a resistance ribbon wound on mica insulation inside a steel sheath. They are inexpensive and universally available, but their contact with the barrel is imperfect, they radiate freely from their outer surface, and their maximum surface loading is limited. Ceramic band heaters place the resistance coil inside ceramic bricks with an insulating layer behind, which directs more heat inward and allows higher operating temperatures with longer service life. Cast aluminium and cast bronze heaters achieve excellent thermal contact by being cast around the element, and they can integrate cooling passages, making them attractive for zones that need both heating and rapid cooling.
Infrared halogen heaters transfer energy primarily by radiation at short wavelengths rather than by conduction. They reach operating temperature very quickly and respond fast to control signals, which reduces overshoot and shortens start-up. Their limitation is that radiant transfer depends on the emissivity and cleanliness of the target surface, so performance can drift as the barrel surface oxidises.
Induction heating jackets take a fundamentally different approach. Instead of heating an external element and conducting the heat inward, an induction coil generates eddy currents directly in the steel barrel wall. Because the barrel itself becomes the heat source, there is no element-to-barrel interface loss, and the outer surface of the induction jacket stays comparatively cool. Reported reductions in heating energy versus uninsulated mica bands are substantial, and the jacket doubles as insulation. Induction heating also eliminates the classic failure mode of a burnt-out band heater sitting undetected in a middle zone while neighbouring zones compensate.
Table 4: Barrel and Die Head Heating Technology Comparison
| Heating method | Heat transfer efficiency class | Heat-up speed | Surface loss without insulation | Service life | Relative cost |
|---|---|---|---|---|---|
| Mica band heater | Low | Medium | High | Short to medium | Low |
| Ceramic band heater | Medium | Medium | Medium | Long | Medium |
| Cast aluminium heater with cooling passages | Medium to High | Medium | Medium | Long | High |
| Infrared halogen heater | High | Very fast | Medium | Medium | High |
| Induction heating jacket | Very high | Fast | Low (jacket is insulated by design) | Very long | Very High |
Insulation jackets: the highest-return measure on the machine
Fitting removable insulation jackets over band heaters on the barrel, adaptor and die head reduces radiant and convective surface losses by roughly 20 to 40 percent. The exact figure depends on melt temperature, ambient air velocity around the machine and the proportion of surface area that can be covered. Higher melt temperatures and draughty plant layouts push the saving towards the upper end.
Three practical points determine whether the measure delivers. First, jackets must be removable and correctly tailored so that maintenance access to thermocouples, heater terminals and screen changers is not compromised; jackets that obstruct servicing get removed and never refitted. Second, the die head is often the most valuable area to insulate because it operates at high temperature with a large exposed surface and, unlike the barrel, receives no shear heat contribution. Third, insulation reduces the heat load reaching the machine’s electrical cabinet and the surrounding air, which in air-conditioned plants delivers a second, indirect saving.
Zone control, thermocouple placement and self-tuning PID
Control quality is as important as heater hardware. A poorly tuned PID loop that oscillates around set point wastes energy in two directions: heaters overshoot, then barrel cooling fans run to bring the temperature back down. Every kilowatt-hour spent in that cycle is completely wasted. Modern self-tuning PID controllers with auto-tune routines eliminate most of this by identifying the thermal characteristics of each zone individually.
Thermocouple placement matters more than most operators realise. A thermocouple in a shallow well responds to heater surface temperature rather than melt temperature, causing the controller to shut off heaters prematurely and then over-correct. Deep-well thermocouples positioned to sense the barrel wall close to the melt give a truer signal and produce smoother, lower-energy control. Apollo specifies zone-by-zone deep-well thermocouples with independent auto-tuning on all ABLB and ABLD Series machines.
Finally, consider start-up sequencing. Bringing all zones to set point simultaneously creates a large simultaneous demand peak, which can trigger demand charges on the plant’s electricity tariff. A staged soak-and-ramp start-up routine, where zones heat in a controlled sequence with a soak period to allow heat to conduct through the barrel wall, both flattens the demand peak and reduces the risk of thermal stress on the barrel and screw.
Compressed Air: The Most Underestimated Energy Cost in Blow Molding
Compressed air is the most expensive utility per unit of delivered energy in any blow molding plant, and it is also the one most likely to be wasted. Generating compressed air is inherently inefficient: only a fraction of the electrical energy supplied to a compressor ends up as useful pneumatic work, with the remainder rejected as heat. Every leak, every unnecessary bar of pressure and every cubic metre of air vented to atmosphere is therefore a multiplied electrical cost.
How much air an EBM machine actually needs
Extrusion blow molding uses compressed air for two distinct purposes with very different requirements. Blow air inflates the parison against the mold cavity and typically operates between 6 and 10 bar depending on container geometry, wall thickness and resin. Control air drives valves, blow pin actuators, ejectors and conveying, and rarely needs more than 6 bar.
A common and costly error is to run the entire machine, and often the entire plant, at the highest pressure any single consumer requires. Compressor power rises with discharge pressure, so every unnecessary bar imposes a permanent penalty on all air produced. Separating the high-pressure blow air circuit from the general-purpose control air circuit, either with a dedicated booster or with a properly zoned distribution network, is one of the most effective compressed air measures available.
Two-stage blowing: pre-blow and main blow
Two-stage blowing splits the inflation into a low-pressure pre-blow that gently expands the parison and positions it against the cavity, followed by a high-pressure main blow that completes forming and presses the material against the mold surface for detail definition and cooling contact. The energy benefit is direct: the majority of the container volume is filled at low pressure, and only the final forming stage requires full pressure. Two-stage blowing also improves wall thickness distribution and reduces the risk of parison blow-out on thin sections, so the quality argument and the energy argument point the same way.
Blow air recovery systems
At the end of the blow phase, the air inside the finished container is still at or near full blow pressure. In a conventional machine, an exhaust valve simply vents it to atmosphere, discarding all of that compression energy and generating a loud noise pulse in the process. A blow air recovery system instead routes the exhausting air through a check valve into a lower-pressure receiver, which then supplies pre-blow air, machine pneumatics, or feeds back to the compressor’s intermediate stage.
Realistic recovery rates fall between 25 and 50 percent of the blow air volume. Where a specific installation lands depends on several factors. Larger container volumes recover more absolute air per cycle. Higher blow pressures leave more recoverable energy. Short cycles mean more recovery events per hour. Conversely, very small containers, long cycles and low blow pressures reduce the attractiveness of recovery because the fixed cost of the receiver, valves and piping is amortised over a smaller recovered volume. Recovery systems also deliver a welcome secondary benefit: noise at the machine drops significantly when high-pressure air is no longer vented directly to atmosphere.
Leak detection and distribution design
Leakage is the silent tax on compressed air. In plants that have never run a systematic detection programme, leakage commonly accounts for a substantial share of total compressed air production, and because leaks run twenty-four hours a day including weekends and shutdowns, they consume energy when no production is happening at all. Ultrasonic leak detection equipment identifies leaks by their characteristic high-frequency signature, allowing a technician to survey an entire plant during normal operation without shutting anything down. A quarterly ultrasonic survey with tagged repair follow-up is one of the lowest-cost, highest-return energy measures available to any converter.
Distribution design matters almost as much. Undersized pipework creates pressure drop, which operators typically compensate for by raising compressor discharge pressure, penalising the whole system. A correctly sized ring main, adequate local receiver capacity near the machine to buffer the pulsed blow demand, and short, generously sized final connections all allow the compressor to run at the lowest viable set point. Local receivers are particularly important in blow molding because the demand profile is sharply pulsed; without buffering, the compressor sees repeated pressure dips and cycles unnecessarily.
Table 5: Compressed Air Measures for EBM Machines
| Measure | Effect on compressed air energy | Secondary benefit | Relative investment |
|---|---|---|---|
| Quarterly ultrasonic leak detection and repair | Large reduction, recurring | Stabilises system pressure | Low |
| Separate blow air and control air pressure levels | Moderate to large reduction | Longer valve and seal life | Low to Medium |
| Two-stage pre-blow plus main blow sequencing | Moderate reduction | Better wall thickness distribution | Low |
| Blow air recovery system (25–50% recovery) | Large reduction | Significant noise reduction | Medium to High |
| Local receiver sized for pulsed blow demand | Moderate reduction | Fewer compressor load/unload cycles | Low |
| Compressor heat recovery to plant hot water | No change to air energy, offsets heating fuel | Reduces boiler load | Medium |
Mold Cooling and Chiller Efficiency
Mold cooling consumes 6 to 12 percent of machine-related power and simultaneously governs cycle time, which makes it the one subsystem where energy optimisation and productivity optimisation can conflict. Colder water shortens cooling time and raises output, but producing colder water costs disproportionately more chiller energy. The correct answer is not the coldest possible water; it is the highest water temperature that still achieves the required cycle time and dimensional stability.
Water temperature and chiller coefficient of performance
Chiller efficiency is described by its coefficient of performance, the ratio of heat removed to electrical energy consumed. Coefficient of performance rises significantly as the chilled water set point rises, because the refrigeration cycle has less temperature lift to overcome. As a rule of thumb widely used in industrial refrigeration, each degree Celsius of raised evaporator temperature improves chiller efficiency by a few percent.
Most extrusion blow molding applications on HDPE and PP run mold cooling water between 8 and 15°C. Many plants default to the bottom of that range out of habit. In practice, container geometry, wall thickness and resin crystallinity determine what is actually required. Thick-walled industrial containers genuinely benefit from colder water. Thin-walled bottles often cool adequately at 12 to 15°C, and the difference in chiller power between 8°C and 14°C over a full year is substantial. The risk of running too cold is not only energy: below the local dew point, condensation forms on mold surfaces and cooling lines, which causes surface defects on the container and corrosion on the mold.
Free cooling and plate heat exchangers
Free cooling, sometimes called economiser operation, uses ambient conditions to reject process heat without running the compressor. When outdoor wet-bulb temperature is low enough, a cooling tower or dry cooler can produce water cold enough to serve the process directly through a plate heat exchanger, with the mechanical chiller idling or running at partial load. In temperate climates this can cover a significant portion of annual cooling hours; in cold climates it can cover most of the winter.
A plate heat exchanger between the cooling tower circuit and the closed process circuit is essential to this arrangement. It keeps the open tower water, with its dissolved oxygen, biological load and suspended solids, separated from the clean closed loop that feeds the molds. Without that separation, mold cooling channels foul rapidly, heat transfer degrades, cycle times lengthen and the plant loses far more than free cooling ever saved.
Conformal cooling channels and mold thermal design
Cooling channel design inside the mold determines how effectively the water actually removes heat. Conventional straight-drilled channels are constrained by drilling geometry and often leave hot spots at handles, neck finishes, pinch-off areas and base corners. Conformal cooling channels, produced by additive manufacturing or by fabricating the mold from bonded layers, follow the contour of the cavity at a constant offset. The result is more uniform heat extraction, fewer hot spots, shorter cooling time and reduced warpage.
Conformal cooling carries a High relative cost in mold tooling and is not justified for every job. It makes economic sense on high-volume, long-running molds where a cycle time reduction of even a fraction of a second compounds into meaningful annual output, and where warpage or shrinkage variation has historically caused rejects. For short-run or frequently changed molds, conventional channels with careful attention to turbulent flow rates deliver most of the benefit at a fraction of the cost.
Pump and flow control
Cooling water pumps are often the forgotten load. A fixed-speed pump sized for the worst-case mold runs at full power regardless of how many stations are active or which mold is fitted. Fitting a variable-speed drive with differential pressure control allows pump power to follow actual demand, and because pump power varies roughly with the cube of flow, even a modest flow reduction produces a large power reduction. Equally important is ensuring flow in each circuit remains turbulent rather than laminar; turbulent flow transfers heat several times more effectively, so a slightly higher flow in the right circuit can shorten cooling time far more than a much higher flow spread indiscriminately across all circuits.
Measuring What You Save: Energy Monitoring, kWh/kg and ISO 50001
No energy saving measure should be implemented without a means of verifying it, because unmeasured savings have a habit of disappearing. Per-machine energy metering is the foundation of every credible energy programme, and it is inexpensive relative to almost every other measure discussed in this article.
Per-machine metering and the right indicators
A three-phase energy meter installed at the machine’s main disconnect, logging active power, reactive power, power factor and cumulative energy at short intervals, converts energy from an abstract plant-level line item into an operational parameter that a shift supervisor can act on. Once that data exists, two indicators do most of the useful work.
Specific energy consumption in kWh per kilogram normalises for part weight and is the correct indicator for comparing machine efficiency, screw configurations and resin changes. Energy per thousand pieces, expressed as kWh per 1,000 pcs, is the correct indicator for commercial quoting and for customer carbon reporting, because customers buy containers, not kilograms. A plant should track both. A change that improves kWh/kg while worsening kWh per 1,000 pcs usually means part weight has drifted upward, which is a resin cost problem hiding inside an energy metric.
Linking energy to OEE
Energy data becomes far more powerful when overlaid on overall equipment effectiveness. The reason is simple: a machine that is heated, pressurised and running its extruder but producing no saleable containers is consuming a large fraction of its full-production power while producing nothing. Idle and non-productive states typically consume a substantial share of running power, because heaters, hydraulics and the chiller remain active.
This means that availability losses carry an energy penalty that never appears in a conventional OEE calculation. Plotting kWh consumed against good parts produced, hour by hour, exposes exactly which shifts, which changeovers and which fault modes are the most expensive. In many plants this analysis reveals that the single largest energy saving opportunity is not a technology upgrade at all, but the elimination of extended idle running during changeovers and quality holds.
Standards frameworks
ISO 50001 provides the management system framework for energy: policy, baselines, energy performance indicators, action plans and management review. ISO 50006 gives practical guidance on establishing baselines and indicators, which is where most plants struggle. Certification is not necessary to benefit; adopting the baseline-and-indicator discipline alone produces most of the value.
On the machine side, EUROMAP 60.1 and 60.2 define standardised measurement procedures and reference conditions for plastics machinery energy performance, which is what makes supplier comparisons meaningful. Without a common measurement boundary and a defined reference cycle, quoted consumption figures from different suppliers are not comparable. IEC 60034-30-1 governs motor efficiency classification, and buyers should specify the required IE class explicitly in the purchase specification rather than accepting whatever the supplier’s standard configuration includes. CE conformity under the Machinery Directive, with EN 422 addressing blow molding machine safety specifically, remains a separate requirement and should not be conflated with energy performance.
Building a defensible baseline
A baseline is only useful if it accounts for the variables that legitimately move energy consumption. Those variables include product mix, resin grade and melt flow rate, regrind ratio, ambient temperature, and running hours. A baseline built on a single week of summer production will make every winter month look like an improvement. Apollo recommends establishing baselines over at least one full production cycle covering the plant’s normal product mix, expressed as an index with the baseline period set at 100 index points, then tracking monthly deviation with product mix normalisation applied.
The Apollo Energy Saving Retrofit Roadmap
The correct order of implementation matters as much as the measures themselves, because early low-cost wins fund later high-cost upgrades and because metering must precede investment. The roadmap below is the sequence Apollo engineers recommend to customers operating existing extrusion blow molding machines, whether those machines came from Apollo or from another supplier.
Table 6: EBM Energy Saving Measures Ranked by Priority
| Measure | Saving on affected subsystem | Saving on total machine power | Implementation difficulty | Relative investment | Payback class |
|---|---|---|---|---|---|
| Compressed air leak detection and repair programme | 10–30% | 1–4% | Easy | Low | Short |
| Barrel and die head insulation jackets | 20–40% | 2–5% | Easy | Low | Short |
| Per-machine energy metering and dashboard | Enabling measure | 3–8% via behaviour change | Easy | Low | Short |
| Melt temperature profile review and PID auto-tuning | 5–15% | 2–6% | Easy | Low | Short |
| Chilled water set point raised within process limits | 8–20% | 1–3% | Easy | Low | Short |
| Auxiliary equipment interlocks (crusher, conveyor, loader) | 20–40% | 1–2% | Easy | Low | Short |
| Two-stage pre-blow and main blow sequencing | 10–25% | 1–3% | Moderate | Low to Medium | Short |
| Variable-speed drives on cooling water pumps | 25–45% | 1–3% | Moderate | Medium | Medium |
| Blow air recovery system | 25–50% | 2–6% | Moderate | Medium to High | Medium |
| Servo-hydraulic power unit retrofit | 15–40% | 3–8% | Difficult | High | Medium |
| Free cooling with plate heat exchanger | 30–70% seasonal | 2–6% annualised | Difficult | High | Medium |
| Induction heating jackets on barrel and die head | 30–60% | 3–7% | Difficult | Very High | Medium to Long |
| PMSM extruder motor upgrade to IE4/IE5 | 8–18% | 4–10% | Difficult | Very High | Medium to Long |
| Barrier screw and grooved feed barrel conversion | 10–25% | 5–14% | Difficult | Very High | Medium |
| Replacement with fully electric machine | Whole-machine architecture change | 15–30% | Project | Premium | Long |
How to read the roadmap
Notice that the percentages in the second and third columns are not additive. Once insulation jackets have reduced heating losses, a subsequent induction heating retrofit delivers less incremental benefit than the table’s standalone figure suggests. The same applies to compressed air: after leak repair and pressure right-sizing, a blow air recovery system operates on a smaller base. Sequencing therefore has real financial consequences, and the cheap measures should always come first, both because they fund the expensive ones and because they establish the reduced baseline against which the expensive ones must be justified.
Notice also that the measures with the largest total-machine impact, screw and barrel conversion and PMSM motor upgrade, are those that address the extruder. This follows directly from Table 1. A plant that has completed all the easy measures and still needs a step change must eventually go there.
How Apollo Engineers Low Power Consumption into Every Series
Apollo, a Wanplas factory, designs energy performance into the machine at the specification stage rather than treating it as an optional accessory list, because retrofitting efficiency is always more expensive than building it in. With ten series and more than eighty models covering containers from 200ML to 1500L, the specific configuration differs by application, but the underlying engineering philosophy is consistent across the range.
ABLB Series, 200ML to 20L
The ABLB Series comprises eight machine types covering the container sizes that dominate food and beverage, daily chemical and pharmaceutical packaging. Because these machines run fast cycles with high annual hours, Apollo prioritises the measures that scale with cycle count: servo-hydraulic power units for the clamping and carriage circuits, two-stage pre-blow and main blow sequencing as standard, insulated die heads, and deep-well thermocouples with independent zone auto-tuning. Extruder drives are specified with inverter control at IE3 as standard, with PMSM at IE4 available where the customer’s duty cycle justifies it.
ABLD Series, 20L to 1500L
The ABLD Series covers large-capacity containers, industrial drums, chemical intermediate bulk containers and large technical parts. These machines have fundamentally different energy characteristics: cycles are long, clamping forces are high, and the holding phase during cooling dominates the cycle. That profile is precisely where servo-hydraulic technology delivers its largest relative benefit, because a fixed-displacement pump would be dumping flow across a relief valve for the majority of every cycle. Accumulator sizing is treated as an energy decision as well as a process decision, since a correctly sized accumulator allows a smaller pump to serve a large peak demand.
Fully Electric Series, 200ML to 20L
The Fully Electric Series eliminates the hydraulic power unit entirely. For customers producing pharmaceutical containers, medical packaging or high-purity daily chemical bottles, the absence of hydraulic oil removes a contamination pathway that no amount of maintenance discipline can fully close. The energy argument is equally strong for plants with high electricity tariffs and three-shift operation, where the near-zero standby consumption compounds across thousands of annual hours. Apollo positions the Fully Electric Series as the recommended choice where container volume, cycle frequency and cleanliness requirements all point the same direction.
Material handling and the wider Wanplas ecosystem
Energy performance does not stop at the machine. Apollo machines process PE, PP, PVC, PA, PC, ABS, PS, EVA, TPU and PETG, and each material carries its own energy profile. High-molecular-weight HDPE with low melt flow rate demands more torque and benefits most from grooved feed barrels. PVC requires low shear and tight thermal control to avoid degradation, which means heating control quality matters more than raw heater power. PC and PA require thorough drying, and the dryer itself becomes a significant energy consumer, so dew-point-controlled dehumidifying dryers with demand-based regeneration are worth specifying.
Where a converter runs significant regrind, the crusher and conveying system become part of the energy picture, and flake consistency directly affects extruder specific energy consumption. For plants scaling into full closed-loop recycling, Wanplas’s Polyretec factory supplies washing and pelletizing lines, and Wanplas’s Kerke factory supplies co-rotating parallel twin-screw extruders for compounding and pelletizing recovered material into consistent, low-variability feedstock. Consistent feedstock is itself an energy measure, because a stable bulk density and melt flow rate allow the blow molding extruder to run at a steady, optimised operating point rather than continuously correcting.
Commissioning, verification and support
Apollo tests machines at its 8,000 square metre factory in Zhangjiagang before shipment, with engineers available for on-site installation and commissioning. Energy verification during commissioning is straightforward and should be requested as part of the acceptance protocol: log machine-boundary energy over a defined production run at the contracted throughput, calculate kWh/kg and kWh per 1,000 pcs, and record the figures as the machine’s baseline at 100 index points. That baseline then becomes the reference for the machine’s entire service life, making future degradation immediately visible.
Across the Wanplas brand, the shared commitment includes machine customisation for molds and voltage, on-site installation by engineers, usage status tracking, factory inspection before shipment, an annual complimentary spare parts allowance, transportation guarantee, production capacity guarantee and quality standards guarantee. Competing suppliers including Bekum, Kautex Maschinenbau, Uniloy, Graham Engineering and Automa offer their own energy packages, and buyers should compare them on a common measurement basis rather than on headline claims. The reference conditions defined in EUROMAP 60.1 and 60.2 exist for exactly this purpose.
Frequently Asked Questions
How much electricity does a typical extrusion blow molding machine use per kilogram of product?
A well-configured single-station HDPE machine producing 5L jerry cans typically lands between 0.55 and 0.95 kWh per kilogram of finished product measured at the machine boundary, covering extruder drive, hydraulics, barrel heating and machine-mounted pneumatics. The extruder alone usually accounts for 0.25 to 0.45 kWh/kg of that total. Older machines with fixed-displacement pumps, mica band heaters and no insulation jackets frequently exceed 1.2 kWh/kg for identical parts. Always confirm whether a supplier’s quoted figure includes plant utilities such as the compressor room and chiller, because that alone can shift the number by more than 20 percent.
Does a servo-hydraulic EBM machine really save 15 to 40 percent of energy?
That range applies to the hydraulic power unit specifically, not to the whole machine. Since the hydraulic power unit represents roughly 10 to 20 percent of total machine power in extrusion blow molding, a servo-hydraulic retrofit typically produces a 3 to 8 percent reduction in total machine consumption. Machines with long holding phases, large accumulator heads or slow cycles sit at the upper end because the conventional pump would otherwise be dumping flow for most of the cycle. Fast-cycling small-container machines sit at the lower end.
When does a fully electric EBM machine make more sense than servo-hydraulic?
Fully electric extrusion blow molding is strongest for containers from 200ML to about 20L, high cycle counts, cleanroom or pharmaceutical environments, and plants with high electricity tariffs and long annual running hours. Above roughly 30L, and especially for accumulator head machines producing large industrial drums, hydraulic or servo-hydraulic actuation remains more practical because of the very high peak clamping force and long stroke lengths involved. The relative capital cost of fully electric is Premium, so the business case depends heavily on annual running hours.
How much can barrel insulation jackets reduce heating energy?
Insulation jackets fitted over band heaters typically cut radiant and convective surface losses from the barrel and die head by 20 to 40 percent, depending on melt temperature, ambient airflow and coverage. Since barrel and die head heating is roughly 8 to 15 percent of total machine power, that means a 2 to 5 percent reduction in total consumption at Low relative investment with a short payback period. The die head is often the highest-value area to insulate because it runs hot, has large exposed surface area and receives no shear heat contribution from the screw.
What is a blow air recovery system and how much air does it actually recover?
A blow air recovery system captures the still-pressurised air released from the container at the end of the blow phase and routes it through a check valve into a lower-pressure buffer receiver that feeds pre-blow, machine pneumatics or the compressor’s intermediate stage. Recovery rates of 25 to 50 percent of blow air volume are realistic on machines producing larger containers at short cycles with high blow pressure. The system also reduces machine noise substantially, since high-pressure air is no longer vented directly to atmosphere.
Which energy saving retrofit should be done first on an older EBM machine?
Start with compressed air leak detection and pressure right-sizing, then barrel and die head insulation jackets, then per-machine energy metering. All three carry Low relative investment, easy implementation and a short payback period. Metering should be installed early even though it saves nothing directly, because it converts every subsequent measure from a claim into a verified result. Only after these foundations are in place does it make sense to evaluate a servo-hydraulic power unit, a PMSM extruder motor or induction heating.
Does running regrind change the energy consumption of an EBM machine?
Yes. Regrind has lower bulk density and irregular particle geometry compared with virgin pellets, which reduces solids conveying efficiency in the feed zone and can raise specific energy consumption by 3 to 10 percent at the same throughput. The crusher itself adds a parasitic load. Grooved feed barrels, consistent flake sizing and a controlled regrind ratio recover most of that penalty. Where regrind quality varies widely, compounding and pelletizing the recovered material into a consistent feedstock, using twin-screw equipment from Wanplas’s Kerke factory, stabilises the extruder operating point and often pays for itself in reduced variability alone.
What standards apply to measuring blow molding machine energy consumption?
EUROMAP 60.1 and 60.2 define the measurement procedures and reference conditions used to compare plastics machinery energy performance on a like-for-like basis. IEC 60034-30-1 classifies motor efficiency into the IE1 to IE5 bands, and buyers should state the required class explicitly in the purchase specification. ISO 50001 provides the energy management system framework, with ISO 50006 covering baselines and energy performance indicators. Machine safety is governed separately, with EN 422 addressing blow molding machines under the CE framework, and resin melt flow rate is measured per ASTM D1238 or ISO 1133.
Can raising the chilled water temperature really save meaningful energy?
Yes, and it is one of the least expensive measures available. Chiller coefficient of performance improves measurably with every degree the evaporator set point rises, because the refrigeration cycle has less temperature lift to overcome. Many plants run mold cooling water at 8°C purely out of habit when 12 to 15°C would meet the cycle time requirement for thin-walled containers. The limit is condensation: water below the local dew point causes moisture on mold surfaces and cooling lines, producing surface defects and corrosion. Verify cycle time and part quality at the higher set point before making it permanent.
Conclusion
Energy saving technology for EBM machines is not a single feature that can be ticked on an order form. It is a layered engineering discipline that starts with knowing where the power goes, continues with attacking each subsystem in proportion to its share, and depends entirely on measurement to confirm that the savings are real and durable.
The evidence across Apollo’s installed base points to a clear hierarchy. The extruder drive, at 45 to 60 percent of total machine power, is where the largest absolute savings live, accessible through IE4 and IE5 motor classes, barrier screw geometry, grooved feed barrels and disciplined melt temperature management. The hydraulic power unit, at 10 to 20 percent, responds strongly to servo control precisely because its duty cycle is so intermittent, delivering 15 to 40 percent reductions on that subsystem. Barrel heating and compressed air together represent a comparable share and are far cheaper to address, with insulation jackets cutting surface losses by 20 to 40 percent and blow air recovery returning 25 to 50 percent of blow air volume. Mold cooling closes the loop, where a raised water set point and free cooling deliver savings that cost almost nothing to implement.
The practical recommendation is unchanged regardless of machine brand or age. Install per-machine energy metering first. Fix compressed air leaks and insulate the barrel and die head next, because these are Low investment, easy to implement and carry a short payback period. Establish kWh/kg and kWh per 1,000 pcs as tracked indicators against a baseline of 100 index points, link them to OEE data so that idle running becomes visible, and only then commit to the High and Very High investment upgrades that address the extruder and the hydraulics.
Apollo, a Wanplas factory with more than twenty years of extrusion blow molding experience and over 4,000 machines running in more than 90 countries, specifies these measures across the ABLB Series, the ABLD Series and the Fully Electric Series. Buyers evaluating new capacity should request machine-boundary energy figures measured under EUROMAP reference conditions, specify the IE motor class explicitly, and include an energy verification run in the acceptance protocol. A machine chosen on those terms will still be earning its keep long after a machine chosen on purchase price alone has become the most expensive asset on the floor.
