Slow or uneven pillow fill is usually blamed on the blower, and the blower is usually the last thing wrong. Five causes account for most of it, and all five produce the same symptom from the operator’s position: the cushion does not reach pressure. Here is how to tell them apart with a pressure gauge and a clamp meter.

Our air cushion machine blower category covers the 12 V, 24 V and 48 V options that packaging OEMs build into cushion lines, and the WS9070A-24-S400 page carries the three-condition performance table used as the reference unit here. The sequence below comes from returned blowers in our lab, where a unit with no fault code is investigated at its three interfaces first, because that is where the answer usually is.

Cause 1: the blower was sized against the free-air figure

This is the most common cause and the easiest to confirm, because the datasheet and the machine disagree by a third. A blower publishes three load points and they are not interchangeable. Our published reference unit is the clearest example.

The free-air row is the one quoted in a selection meeting, and it is the one that cannot inflate a pillow. The cushion needs air at the pressure required to hold the workpiece, and that is the middle row. Fix the operating point rather than the peak: the maximum flow and pressure figures on a datasheet are open-circuit conditions, and a cushion lives in the middle of the curve, where the blower spends 95 percent of its life.

The current column matters too, and it is the one people get wrong on the power supply. Current rises as back pressure rises, so the reference unit draws 3.7 A at free blowing and 7.2 A at blocked outlet, about 1.9 times, while speed drops from 41,000 rpm to 36,500 rpm. Size the supply for the blocked condition rather than the running condition, because that is the current the protection and wiring must survive.

Cause 2: the inlet filter is undersized

The inlet filter is the cheapest component on the machine, which is exactly why it is the one most often under-specified, and it is the first thing to check before condemning a blower. The published reference is that a 50 Pa drop across a too-small filter turns a 380 m3/h blower into a 340 m3/h blower, so a filter that looks adequate on the drawing can quietly remove more than a tenth of the designed air.

The same filter does a second job. Because an inadequate filter lets dust through, it loads the bearings with abrasive particulate, and the published guidance is that service life is shortened by elevated inlet temperature, dust without filtration and continuous operation above the rated static pressure. A dirty filter therefore costs you twice: less air now and a shorter blower life later. Fit the specified element, confirm it is present at commissioning, and check it on a service interval.

Cause 3: DC bus voltage sag at start-up

This one is frequently misdiagnosed as a mechanical fault, because the symptom is the blower tripping on protection and the operator reports a machine that randomly stops mid-cycle. The mechanism is electrical: an undersized DC bus capacitor causes voltage sag at start-up, and the controller misreads that sag as a stall and trips.

Our published recommendation is to rate the DC bus capacitor at no less than twice the inrush current. That check costs nothing and separates a real electrical fault from a mechanical one in minutes, because if the supply sags at start-up the blower never reaches working speed, and the cushion fills slowly or not at all on every cycle rather than gradually degrading.

Cause 4: PWM frequency mismatch

If the fill is inconsistent from cycle to cycle rather than uniformly slow, suspect the control signal before the blower. The published controller expects a 25 kHz PWM signal, and the failure we see is a PLC sending 50 Hz. The result is an audible hum and unstable speed, so the machine reaches pressure inconsistently and the cushions come out visibly uneven from bag to bag.

The fix is not a controller replacement, it is a datasheet check. Confirm the PWM frequency in the controller datasheet before wiring, and note that the interface has to match the machine: our published units are driven through an external brushless controller with speed set by 0-5 V or PWM, so a machine built around an internally controlled blower will not accept the part. And confirm rotation at low duty before connecting the load, because these blowers run in one direction only, and reversing the impeller does not reverse the air.

Cause 5: inlet plenum bends and length

The fifth cause is invisible on the machine drawing, and it surprises most OEMs. The published field reference is that a 90 degree bend immediately before the blower inlet costs 8 to 12 percent of the rated airflow on most centrifugal designs, more than most datasheets warn about. In our own reference builds we limit the inlet straight length to at least three times the inlet diameter before the first bend, and the published port sizes are 8 mm inlet and 8 mm outlet, worth confirming against your tubing early because it determines the coupling, the sealing method and the pressure drop the connection introduces.

That last point matters because it is where a specification error and a plumbing error compound. The machine needs flow at pressure; the blower delivers it only after the inlet path, the filter and the manifold have taken their share. When those losses stack, the machine cannot reach its published cycle time even with a healthy blower.

One point worth separating from the troubleshooting, because it changes who owns the work: compliance sits on the finished machine, not the blower alone. In Europe the EMC Directive 2014/30/EU applies to the drive and controls, the RoHS Directive 2011/65/EU applies to substances and REACH (EC) No 1907/2006 to the supply chain, while the Machinery Directive 2006/42/EC governs the cushion machine itself. In the United States the guarding expectations for the finished build are set out in 29 CFR 1910.212. The practical consequence for a blower buyer is that the part must contribute to a file the machine builder owns, so establish what the blower must supply before specifying a part that cannot supply it.

The commissioning record that makes the next fault a lookup

A commissioning record built from four measured values is what makes the next replacement a lookup instead of an investigation. The four are airflow and static pressure at the working point rather than at free air, the ambient temperature logged alongside those readings, the inlet filter part number with its service interval, and the controller board model with its speed command method. Add the blower model number, installation date and running hours, and the record survives a technician change.

This is worth stating plainly because the request that comes back to us most often carries only the machine model. A request carrying only the machine model comes back as a list of candidates rather than a part number, and that is the most common reason a maintenance team ends up with the wrong blower. The six items we ask for before raising a replacement quotation are the failed unit’s nameplate photographed rather than transcribed, the drawing of the unit removed, the port arrangement and lead length, the wire count and connector part number with the controller board model and its speed command method, the duty point in airflow and static pressure, and the installation conditions.

Frequently Asked Questions About Air Cushion Machine Inflation Problems

Last verified: 2026-11-13. Blower performance, environmental limits, port sizes, filter and plenum flow-loss figures, PWM frequency, DC bus capacitor rule, published life ratings and commercial terms are as published by Ningbo Wonsmart Motor Fan. The reference blower data is the published WS9070A-24-S400 figure set. Cushion inflation depends on the deck, the workpiece and the operating point, so treat the published curves as a starting point and confirm on the bench. Compliance sits on the finished machine: the EMC Directive 2014/30/EU applies to the drive and controls, the RoHS Directive 2011/65/EU to substances and REACH (EC) No 1907/2006 to the supply chain, and where the equipment is medical IEC 60601-1 is the reference the customer will ask for. No unit price, market size figure or project cost figure appears here, because none is published.