Air Compressors for PET Bottle Blow Molding at Up to 40 Bar

Why 40 Bar

PET preforms during blow molding have an axial stretch ratio of 2.5 to 3.5 times and a hoop stretch ratio of 3 to 5 times. To complete this deformation within less than one second at 95 to 115°C and press the preform against the mold wall, air pressure of 25 to 40 bar is required. Simple bottle shapes under 500ml with thin walls blow well at around 28 bar. Bottles above 1.5L, irregular shapes, and bottles with fine surface textures often will not fill out properly below 35 bar.

Most selection guides stop here and start listing compressor specifications. This misses an upstream variable that has a large impact on selection: the IV value of the preform.

Process

PET Blow Molding Air Supply

IV value (intrinsic viscosity) in the range of 0.80 to 0.86 dl/g indicates a standard preform with sufficient strain hardening. During stretching, the bottle wall tends to self-equalize in thickness, and the tolerance window for blowing pressure is wide. Once IV drops below 0.76 dl/g, the material's self-equalizing ability noticeably decreases, and the same bottle shape may need an additional 3 to 5 bar to achieve acceptable wall thickness distribution. Batch-to-batch IV fluctuations in raw material can reach ±0.03 dl/g, which is already enough to push a compressor sized for 30 bar operation to its limit.

Air Volume Matching

Rotary blow molding machines consume air in pulses. Each station blows for 0.5 to 1.2 seconds, followed by exhaust and mold clamping intervals. The compressor faces repeated peaks, not steady-state flow. Sizing a compressor to the nameplate average air consumption of the blow molder makes the total volume look sufficient, but pressure drops below the process minimum during peak moments, and defects like whitened bottle bases and thin shoulders follow.

Receivers that are installed yet still allow noticeable pressure swings usually suffer from the same cause: the allowable pressure fluctuation window is too narrow, and the effective buffer volume cannot cover one blowing cycle's peak demand.

Temperature, Cooling, and Volume Degradation

Compressing air from atmospheric pressure to 40 bar produces a theoretical adiabatic discharge temperature exceeding 400°C. After three-stage compression with intercooling, final stage discharge temperature typically falls between 150 and 180°C.

The condition of the intercoolers directly determines effective displacement of the entire machine. This is not a slow degradation process. Heat exchanger fins in dusty environments accumulate fouling quickly. Three months without cleaning can raise the second stage inlet temperature from a design value of 40°C to 55°C, corresponding to a volumetric efficiency loss of 4% to 6%. This is greater than the efficiency loss caused by piston rings worn to end of life. Many maintenance schedules specify detailed inspection intervals for piston rings and valves while categorizing intercooler cleaning as "as needed." The priorities are reversed.

Regarding the gap between nameplate displacement and actual displacement, there is a factor more fundamental and more easily overlooked than cooler degradation: intake air temperature. Compressor manufacturers test nameplate displacement at 20°C or 25°C intake conditions. PET blow molding plants commonly see compressor room temperatures of 45°C in summer, especially those that place the compressor room on a mezzanine floor or adjacent to injection molding machines. Intake temperature rising from 25°C to 45°C reduces displacement by approximately 6% to 8%. A machine rated at 100 Nm³/min actually delivers 92 to 93 Nm³/min.

Oil-Free and Oil-Lubricated

Compressed air oil content must be below ISO 8573-1 Class 1 (≤0.01 mg/m³). There is no dispute about this in the PET blow molding industry. The dispute is about which path to use to get there.

Oil-Free

PTFE Ring Technology

Oil-Lubricated

Cascaded Filtration

Oil-free piston compressors do not introduce lubricating oil into the compression chamber. Purchase cost is roughly 40% higher than the same displacement oil-lubricated machine (this premium has been narrowing over the past decade; ten years ago the gap exceeded 60%). The typical replacement interval for PTFE piston rings on oil-free machines is 3,000 to 5,000 hours, while metal piston rings on oil-lubricated machines can run 8,000 to 12,000 hours.

Oil-lubricated machines equipped with cascaded activated carbon filters and coalescing filters can achieve discharge oil content below 0.003 mg/m³, which on paper exceeds Class 1 requirements. The cost is that filter element replacement discipline must be strictly followed. The adsorption capacity of activated carbon elements does not degrade linearly. During the first 80% of service life, filtration efficiency remains stable. During the final 20%, it deteriorates rapidly. Replacing on fixed time intervals rather than on differential pressure or residual oil concentration means either replacing too early and wasting money, or replacing too late and letting oily air through.

Installing an inline residual oil monitor on the discharge piping catches this degradation trend early. A monitor costs roughly 20,000 to 30,000 RMB, about the same as one piston ring replacement. Among all known PET blow molding plants, the installation rate for this equipment is very low.

Number of Compression Stages

This topic is treated as background knowledge in most compressor selection articles, with a conclusion of "three or four stages" and nothing more. It warrants expansion because stage count and interstage pressure distribution are the core architectural decisions that determine machine efficiency and reliability.

Total compression ratio of 40. Three stages give approximately 3.4 per stage. Four stages give approximately 2.5 per stage. Three-stage designs are compact, have fewer parts, and cost less. Per-stage temperature rise is high, and dependence on the cooling system is heavy. Four-stage designs have gentler temperature rise, higher volumetric efficiency, and more uniform mechanical loading across stages. The trade-off is an additional cylinder set, an additional intercooler, and an additional set of valves, increasing overall size and weight. Machines below 50 Nm³/h are mostly three-stage. Above that, four-stage advantages become apparent.

The interstage pressure distribution has a greater impact on performance than the number of stages itself. Equal ratio distribution is not the optimum. In three-stage compression, one well-proven distribution is: first stage outlet 3.5 bar, second stage outlet 13 bar, third stage outlet 40 bar, corresponding to compression ratios of 3.5, 3.7, and 3.1. This arrangement puts the highest discharge temperature at the second stage. The second stage is usually positioned in the middle of the machine, where cooling piping runs are shortest and heat dissipation conditions are best. The third stage (final stage) compression ratio drops to 3.1, keeping discharge temperature within 160°C, which is favorable for final stage valve and piston ring thermal loading.

Asking a compressor manufacturer "how are your interstage pressures distributed, and why did you choose that distribution" reveals whether the company does its own thermodynamic optimization design or merely assembles purchased cylinders and valves.

Specific Power and the FAD Labeling Trap

Specific power is the core metric for evaluating compressor energy efficiency, expressed in kW/(Nm³/min). For 40 bar blow molding compressors, specific power ranges from 5.5 to over 9. Compressor electricity consumption accounts for 25% to 35% of total energy consumption for an entire blow molding line. For every 1 kW/(Nm³/min) difference in specific power, at 100 Nm³/min machine capacity, 8,000 operating hours per year, and industrial electricity price of 0.7 RMB/kWh, the annual electricity cost difference is 560,000 RMB.

Before comparing specific power across brands, one prerequisite must be unified: which reference basis is used for FAD labeling.

The compressor industry uses three FAD labeling methods. The first is Actual Inlet Conditions, calculated at the temperature, pressure, and humidity at the compressor intake. The second is Standard Reference Conditions, calculated at 1 bar, 20°C, 0% relative humidity per ISO 1217 Annex C. The third is Site Conditions, calculated at buyer-specified local environmental conditions.

For the same compressor, FAD at actual inlet conditions is 8% to 12% larger than FAD at standard conditions. Specific power calculated from inlet condition FAD is a smaller number (looks more efficient). Specific power calculated from standard condition FAD is a larger number (looks less efficient). Both numbers describe the same machine.

Some quotation documents use inlet condition FAD on the technical specification page (the specific power number looks good, helpful for winning bids) and standard condition FAD in the contract appendix for acceptance criteria (easier to meet, reducing breach-of-contract risk). This is not fraud. Both numbers are accurate. The problem is that if the buyer does not recognize the two systems are different, comparing Brand A's inlet condition FAD against Brand B's standard condition FAD can completely reverse the ranking.

Valves

If only one component of a 40 bar blow molding compressor could be studied in depth, it should be the valve.

Valves endure 500 to 1,500 opening and closing impacts per minute (depending on speed and stage number), at working temperatures between 80 and 170°C (depending on stage and cooling conditions), with pressure differentials across sealing surfaces ranging from a few bar to over ten bar. They are the most frequently replaced component in the entire machine and the primary cause of unplanned downtime.

Components

Valve Engineering at 40 Bar

On 40 bar blow molding compressors, the mainstream valve type is the ring valve. Ring valve plates consist of several concentric metal rings, each rising and falling independently. Flow area is large and response speed is fast. By comparison, plate valves use a single plate or several sector-shaped plates with relatively small flow area and limited lift. At high speeds, plate valves are prone to delayed seating that causes backflow losses. Above 600 rpm, ring valves show clear advantages in both life and efficiency.

Within ring valves there is further classification. Concentric ring valves have rings arranged concentrically, with motion guided by a center post or guide pins. The layout is symmetrical and sealing integrity is good. Eccentric ring valves have rings arranged around different centers, which structurally permits greater flow area at the cost of reduced uniformity across sealing contact surfaces. At 40 bar, any non-uniformity in the sealing face means leakage. Leakage means reduced volumetric efficiency and localized overheating. Concentric ring valves deliver more reliable overall performance at this pressure level.

A fact that is rarely discussed: design and manufacture of valves for 40 bar reciprocating piston compressors worldwide is concentrated in a handful of specialized European companies. Most compressor OEMs, including some well-known brands, do not design their own valves. They purchase finished or semi-finished valves from these specialist valve manufacturers and assemble them into their own cylinder heads. The same valve model from the same valve manufacturer may appear simultaneously in products from three or four different compressor brands. Conversely, different models within a single compressor brand may use valves from different valve manufacturers. Knowing which valve manufacturer's product is installed in a compressor, which series, and what material grades are used for plates and springs, is more useful than knowing the compressor brand's history and market share. Because valve failure is the number one cause of unplanned downtime for 40 bar compressors, and valve supply and spare parts channels are independent of the compressor brand.

System-Side Matters

Pressure dew point. Compressed air for blow molding requires pressure dew point control between -20°C and -40°C. A pressure dew point of -40°C at 40 bar corresponds to approximately -58°C at atmospheric pressure. If a supplier reports atmospheric dew point but it is evaluated as pressure dew point, the sizing error will be severe. Regeneration losses for adsorption dryers at 40 bar consume 10% to 18% of total air supply and must be subtracted from effective delivery volume.

AIS (Air Recovery System). Recovers exhaust air from the high-blow phase to supply the pre-blow phase. Nominal air savings of 30% to 40%. Recovery efficiency is constrained by exhaust valve response speed. On high-speed blow molders, actual savings rates often fall short of nominal values. Reducing compressor displacement based on 80% of the nominal recovery rate is a more reliable approach.

Piping pressure drop. Every 1 bar of pressure drop means the compressor must deliver approximately 2.5% more power. From compressor discharge through dryer, filters, receiver, main header, branch piping, to blow molder inlet, cumulative pressure drop of 2 to 4 bar is common. Pipe diameter, number of elbows, valve types, and filter cleanliness are all sources of pressure drop.

Machine Layout Preferences

W-type and L-type cylinder arrangements on large displacement machines offer better vibration balance than V-type, with lower foundation requirements. Forged steel crankshafts have fatigue life an order of magnitude higher than cast iron crankshafts. This difference only becomes visible after three to five years of continuous 24/7 operation and can only be confirmed during the selection phase by requesting material certificates.

Cylinder cooling methods are water-cooled and air-cooled. At the 40 bar level, only water cooling is considered. Air cooling at this pressure level cannot provide sufficient heat dissipation, discharge temperatures cannot be controlled, and piston ring life and volumetric efficiency suffer as a consequence. Water quality management (hardness, pH, corrosion inhibitor) in the water cooling system affects heat exchanger life more than water temperature itself. In hard water regions, heat exchanger internal surfaces scale quickly. Without chemical cleaning for two to three years, heat exchange efficiency can drop by over 30%.