A stamping fixture with a 40 mm bore cylinder took 1.2 seconds to extend, and the process spec allowed 0.6. The customer swapped the cylinder for a bigger one. It got slightly faster — 1.0 seconds — and then the air supply couldn’t keep up. The real problem wasn’t the cylinder. It was the valve. The 5/2 valve on the cylinder had a Cv of 0.8, and the cylinder needed roughly twice that. Nobody had sized the valve. It was picked because it was on the shelf.

The Cv equation that decides speed

The flow through a pneumatic valve is set by its Cv (flow coefficient), the supply pressure, and the downstream pressure. For a cylinder, the effective flow needed is determined by the volume the cylinder sweeps per second. A 40 mm bore cylinder with 300 mm stroke sweeps 0.38 liters extending, 0.3 liters retracting (rod volume reduces the annulus). To fill that in 0.3 seconds (which would meet the 0.6 s extend spec with margin), you need about 1.3 liters per second, or about 2.7 SCFM at 6 bar supply.

The Cv needed for that flow is roughly:

Cv = Q × sqrt( (273+T) / (ΔP × (P2 + 1.013)) )

For Q = 1.3 L/s, T = 20°C, supply 6 bar, and a reasonable ΔP of 1 bar across the valve: Cv comes out around 1.4-1.6. The installed valve was 0.8. It was limiting the flow to about half of what the cylinder needed. The cylinder didn’t move slowly because it was weak. It moved slowly because the valve starved it of air.

Why the bigger cylinder made it worse

Swapping to a 50 mm bore cylinder changed the swept volume to 0.59 liters extending. The flow requirement went up to 2 liters per second. With the same Cv 0.8 valve, the extension got slower in theory — but the bigger cylinder also generated more force, so the motion in the no-load fixture was dominated by valve flow, and the speed change was small. The air supply then became the limit, and the pressure dropped on every other cylinder on the same line. The fixture fix made the whole line slower. That’s what happens when you fix a flow problem with a force solution.

The fix

The valve was replaced with a 5/2 with Cv of 1.8. The extension time dropped to 0.45 seconds, well under the 0.6 spec. The supply pressure drop on the other cylinders disappeared. The valve cost $20 more than the original. The only real work was reading the flow requirement off the cylinder volume and the cycle time, and picking a valve with the Cv to match.

The rules for valve sizing on cylinders

Three rules cover most applications. First, calculate the swept volume: bore area × stroke. Second, decide the target stroke time and convert to required flow in liters per second. Third, pick a valve with a Cv that provides that flow at your supply pressure, and add 30% margin for fitting restrictions, silencer backpressure, and aging seals. The exhaust side matters too — the valve must be able to exhaust the air from the other port at the same rate, or the cylinder stalls against its own cushion. A valve that’s rated for the supply flow but weak on exhaust will make the cylinder sluggish in one direction.

Speed control valves (flow restrictors) are a different subject. A speed controller is not a substitute for a correctly sized main valve. It’s a fine-tuner that goes downstream of a valve that can already flow enough. If the main valve is too small, the speed controller just makes the problem worse, because it adds restriction on top of restriction. Size the main valve first. Adjust the speed controller second.

The check that catches it

If a cylinder is slow, measure the pressure at the cylinder port during motion, not at the regulator. If the port pressure collapses well below supply during the stroke, the valve is the restriction. If the port pressure holds at supply, the restriction is elsewhere — mechanical binding, a seized guide, or the exhaust path. That one measurement, a gauge on a tee at the cylinder port, splits the troubleshooting in two. Most slow-cylinder calls are valve sizing problems wearing a mechanical costume.

The 40 mm cylinder was slow because the valve’s Cv 0.8 starved it — it needed 1.5. A bigger cylinder made it worse. Size the valve to the swept volume and target time, add 30% margin, and check port pressure during motion. A cylinder that’s starved of air is a flow problem, and no amount of force fixes a flow problem.