Pneumatics

Compressed air is the most expensive utility on the plant and the one treated most casually. It is also the one most likely to couple two machines that have nothing else to do with each other, because everything shares the same header.

Force and speed are different problems

This is the distinction the whole chapter turns on, and it is the one an operator's report will not make for you. "The cylinder isn't working properly" covers two faults with nothing in common:

• Weak — it does not push hard enough. That is force, and force is pressure acting on an area.
• Slow — it gets there eventually. That is speed, and speed is about how fast air can get out of the way on the other side.

Establish which one you are looking at before anything else, because the searches do not overlap. Turning the regulator up is a reasonable response to one of them and useless for the other.

The system, end to end

Compressor → receiver → dryer → distribution → an FRL at the machine → valve → actuator → exhaust to atmosphere. Two things about that chain are worth noticing straight away.

First, it ends in the atmosphere. Unlike a hydraulic circuit there is no return line, which means the exhaust path is part of the circuit and a blocked silencer is a genuine fault rather than an inconvenience.

Second, everything upstream of the FRL is shared. A demand somewhere else on the site is a pressure change here — which is exactly the Palletiser 01 mission, where a case erector a hundred and thirty feet away drops the header enough to collapse a vacuum generator.

Where force comes from, and where speed comes from

Force = pressure × piston area. A 2 in bore cylinder at 90 psi develops roughly 280 lbf extending. Retracting, the rod occupies part of the piston area, so the same cylinder at the same pressure pulls appreciably less — which is why a machine that extends happily and struggles to retract is often not faulty at all, just specified tightly.

Speed is flow, and the flow that matters is usually the exhaust. Air has to leave the far side of the piston before the piston can move into that space, so a restricted exhaust slows a cylinder just as effectively as a restricted supply — and blocked silencers are far more common than blocked supply lines.

This is also why speed controls are fitted as meter-out: throttling the exhaust means the piston always works against a cushion of trapped air and moves smoothly. Metering the inlet instead gives a cylinder that lurches, because nothing resists it once it starts moving.

Air preparation, and why water is the enemy

Compressing air concentrates the moisture that was in it, and cooling it downstream condenses that moisture out — inside your pipework. Water carries rust and scale into valves, washes lubricant out of cylinders, and freezes in outdoor lines.

The FRL at the machine is the last defense:

• Filter — particulates and bulk water. The bowl tells you the truth about the whole system's condition, and it is free to look at.
• Regulator — sets the pressure this machine actually works at, which is usually below header pressure.
• Lubricator — where fitted. Many modern components are designed lubricant-free, and adding oil to them washes out their factory grease.

A filter bowl full of water is not a filter doing its job well. It is a dryer that is not doing its job at all.

Reading a valve symbol

The common types you will meet:

• 3/2 — three ports, two positions. Single-acting cylinders, blow-off, pilot signals.
• 5/2 — the workhorse for a double-acting cylinder. Spring return means it has a default position it falls back to when de-energized.
• 5/3 — three positions, with a defined center: closed center holds position, open center lets the cylinder float. Which center it has decides what the machine does on a power failure, which makes it a safety-relevant choice rather than a preference.

The manual override on a solenoid valve is the most useful diagnostic feature on the whole assembly. Pressing it operates the valve mechanically, bypassing the coil and the PLC entirely. If the cylinder moves on the override but not on command, everything downstream of the valve is fine and the fault is electrical — Chapter 9's chain, in one press.

Leaks

A pneumatic system leaks by default and the leaks are treated as background noise because nothing stops working. They are worth taking seriously for two separate reasons.

The economic one: a single 1/8 in hole at 90 psi wastes a meaningful fraction of a compressor's output continuously, and a plant with many of them runs a compressor that never unloads.

The diagnostic one matters more here. Leaks and large intermittent demands lower header pressure, and the symptom appears on whatever component has the least margin — which is very often not the leaking one. That is a whole mission on this site, and it is the reason a "weak cylinder" complaint sometimes has nothing to do with that cylinder.

1. Decide whether it is slow or weak, before anything else

   Watch it move under load. A cylinder that reaches full force but takes too long is a flow problem; one that moves at normal speed and stalls against the load is a pressure problem. Getting this wrong sends you to the opposite end of the circuit from the fault.

2. Read the pressure at the machine, while the machine is working

   Safety

   Not at the header, and not at rest. The regulator's own gauge under demand is the number that matters, and a reading that sags when the machine cycles is telling you about supply rather than about the actuator.

3. Look at the FRL — the filter bowl is free information

   Water level, element condition, regulator setting against what it should be. The bowl reports on the whole system's health, and if several machines' bowls are full you have found something bigger than this callout.

4. Use the manual override on the valve

   Safety

   The single most informative action in the chapter. Moves on override means everything from the valve outwards is healthy and the fault is electrical — straight into Chapter 9. Does not move means the spool, the supply or the actuator. Confirm nobody is in reach first: this bypasses every interlock.

5. Check the exhaust path for anything slow

   Silencer, flow control setting, tube condition, valve exhaust ports. Silencers block gradually and are the commonest single cause of a cylinder that has got slower over months. Removing one temporarily to test is a legitimate diagnostic and a very loud one.

6. Listen, with the machine still

   Leaks are audible in a quiet cell and invisible on any instrument. A hiss that continues when the machine is at rest and pressurized is either an external leak or a valve passing internally, and walking the circuit with the machine idle finds a surprising amount.

7. Test for internal leakage past the piston

   Hold the cylinder at one end of stroke and watch the valve's exhaust port. A continuous flow there means air is crossing the piston seal. This distinguishes a worn cylinder from a leaking fitting, which look identical from the machine's behavior.

8. Ask what put the water or the dirt there

   A contaminated valve is a symptom. If the filter bowl is full, the dryer or the drains are the actual fault, and the valve you have just cleaned will stick again. Chapter 5's principle, applied to a different utility.

• Turning the regulator up to fix a slow cylinder

  Why

  The regulator is right there, the adjustment is free, and more pressure feels like more of everything. It is not: speed is governed by how fast air can leave the far side of the piston, and raising supply pressure barely touches that while quietly increasing the force on every mechanical part downstream.

• Fitting flow controls to meter the inlet

  Why

  Both orientations restrict flow and both slow the cylinder, so meter-in looks like it works. What it produces is a cylinder that hesitates and then lurches, because there is nothing resisting the piston once it breaks away. Meter-out leaves a cushion of trapped air to work against and gives smooth motion.

• Ignoring water in the filter bowl

  Why

  It is normal, every bowl has some, and draining it is somebody else's routine job — so it stops being information. But a bowl that fills quickly is the clearest signal available that the dryer or the drains have failed, and that failure is upstream of every pneumatic fault on the site.

• Not pressing the manual override

  Why

  It takes two seconds and splits the entire circuit into electrical and pneumatic halves, and it gets skipped because the meter feels like the proper tool. A great many pneumatic callouts are electrical faults that would have been identified in the first minute.

• Replacing the cylinder when the valve is passing

  Why

  The cylinder is the part that visibly is not doing its job, so it is the part that gets ordered. But a valve leaking internally starves or back-feeds the cylinder and produces exactly the same weak, drifting behavior. The exhaust port test separates them before anything is bought.

• Treating leaks as cosmetic

  Why

  Nothing stops working, so nothing gets raised. Meanwhile the compressor never unloads, the header sits lower than it should, and every component with a thin margin becomes intermittent — usually somewhere else entirely, which is what makes the eventual fault so hard to attribute.

Five questions. Each is answerable by reasoning about where air has to get to or out of, rather than by recalling a component list.