A hydraulic cylinder that hammered at the end of stroke. The 50 mm bore cylinder traveled 500 mm at 0.5 m/s. At full extension, it slammed into the end cap. After a month, the end cap cracked. The customer thought the flow control valve was too open. It was. But the real fix was proper cushioning — not just slowing the whole stroke. This is about cylinder end cushioning.
The impact energy
At end of stroke, the moving mass has kinetic energy that must be absorbed:
E = ½ · m · v²
For a 50 kg load moving at 0.5 m/s: E = 0.5 × 50 × 0.25 = 6.25 J. That’s small. But the impact force is high because the deceleration distance is short. Without cushioning, the piston hits the end cap in about 1 mm. The deceleration: a = v²/(2s) = 0.25/(2×0.001) = 125 m/s². The impact force: F = m·a = 50 × 125 = 6,250 N. That’s 625 kg of impact force on the end cap. After thousands of cycles, the cap cracks.
How cushioning works
Cylinders with cushioning have a needle (or sleeve) that blocks the main exhaust port near end of stroke. The oil must escape through a small adjustable orifice. The restricted flow creates back-pressure that decelerates the piston. The cushioning energy absorbed:
E_cushion = P_cushion × A × s_cushion
Where P_cushion is the back-pressure (bar), A is the piston area (mm²), and s_cushion is the cushion length (mm). For our 50 mm bore: A = π × 25² = 1963 mm². With cushion pressure of 40 bar (4 MPa = 4 N/mm²) and cushion length of 20 mm: E_cushion = 4 × 1963 × 20 = 157,040 N·mm = 157 J. That’s way more than the 6.25 J impact energy. The cushion works.
What I changed
1. Adjusted the cushion needle. The cushion screw was fully open (no restriction). I closed it 2 turns. The oil had to escape through the small orifice. The piston decelerated over 20 mm instead of 1 mm. The impact force dropped from 6,250 N to about 312 N (20x lower). No more hammering.
2. Verified the cushion length. Some cheap cylinders have short cushions (10 mm). For high-speed applications, I specify cylinders with 30-50 mm cushion length. The longer cushion decelerates the mass more gently. The end cap sees lower peak force.
3. Added external deceleration. For very high-speed or heavy loads, I add an external flow control valve that slows the cylinder for the last 50 mm of stroke. The cushion handles the final 20 mm; the external valve handles the approach. Combined, the cylinder decelerates over 70 mm. Smooth stop.
The cushion adjustment rule
Start with the cushion needle fully closed. Run the cylinder. If it doesn’t reach end of stroke (stops short), open the needle 1/4 turn. If it hammers at end, close it 1/4 turn. Iterate. The correct setting is: the piston reaches end of stroke in about 0.5 seconds after deceleration starts, with no audible impact. Too much cushion: slow cycle. Too little: hammer.
Double-acting cushioning
Both ends of the cylinder need cushioning. A cylinder that extends fast but retracts slowly still hammers on retraction. I adjust both ends. For vertical loads (gravity extending the cylinder), the cap end cushion is more critical — gravity accelerates the load. I spec longer cushions for vertical applications.
The adjustment I make: start closed, open 1/4 turn increments until the piston reaches end of stroke without impact. The hammering cylinder had the cushion fully open. Close the needle 2 turns and it stops. For heavy/high-speed loads, add external flow control for the approach deceleration.
