The Truth About Hydraulic Cylinder Drift
I had a brief career in construction. We used an old hydraulic manlift for elevated projects, and it had a problem – we would start the day eye-level with our work, and after an hour would realize we were balancing on our toes to reach the same level. Cylinder drift was to blame. The lift cylinder was slowly retracting while the machine was off. In a manlift, it’s an annoyance requiring you to raise yourself every half hour, but the drift is dangerous if lifts are used to support heavy loads with the possibility that people or equipment may be under them. This is one reason why when doing automotive repairs and working underneath a car, you should always use solid jack stands or blocks instead of the hydraulic jack.
A guy on my crew told me the lift slowly sank because the piston seals needed to be replaced. I wasn’t a seal engineer at the time, so it sounded reasonable. Now I know drift is more complicated, and it’s critical to understand if you’re responsible for cylinder design. Hydraulic cylinders have two main components: the piston, which is acted upon by pressurized fluid to create force and motion, and the rod, which transfers force and motion to the machinery (in my case, the lift platform) (Fig. 1). Located elsewhere are valves that open and close, controlling fluid flow into the cylinder.
Figure 1. Typical hydraulic cylinder
How drift works
Let’s remove the piston completely. We now have just a rod in a bore, which is known as a ram-style cylinder (Fig. 2). Assume we have perfect, leak-free valves and rod seals. If we shut both valves, zero oil can enter or leave the cylinder.
Figure 2. Piston removed — ram style cylinder
Since moving the rod changes the fluid volume inside the cylinder (the rod takes up space), fluid MUST flow into or out of the cylinder for it to move.
Since this can’t happen while our perfect valves are closed, the rod can’t move. This is called hydraulic lock.
As you can see with hydraulic lock, bad piston seals wouldn’t cause drift in our manlift. Volume loss or escape from the cylinder is what caused us to slowly droop back to the floor. In my situation, either the valves were leaking, slowly reducing the volume of oil in the cylinder, or leaky rod seals (easier to spot) were allowing the fluid to escape the system.
There are a few caveats to this scenario. We assume oil is incompressible, which is not entirely true. Because the oil does squish and stretch a little, the rod will move a small amount with large load changes (read up on ‘bulk modulus of hydraulic oil’). This is not drift, since the oil quickly reaches equilibrium and the rod will not continue to move.
Single-acting cylinders (Fig. 3) are an exception, because oil leaking across the piston is leaving the system. This is similar to when the rod seals leak in double-acting cylinders – drift occurs. Double-ended cylinders (Fig. 4) are also an exception because the fluid volume in the cylinder does not change as the rods move. Both of these systems require low leak piston seals to prevent drift.
Figure 3. Single-acting piston
Figure 4. Double-ended cylinder
Why leaky piston seals are a problem
Leakage across the piston doesn’t cause drift but can cause a number of other complications. Rod retraction relies entirely on the piston seal blocking pressure from crossing the piston; I’ve already described how simply pumping fluid into one side of a cylinder without a piston seal will only cause the rod to extend. Using fluid pressure to retract the rod is not possible without a piston seal.
When extending the rod or holding a load (valves open, no hydraulic lock applied), leakage across the piston seal slowly allows pressure to equalize on both sides of the cylinder. Once this happens, effective piston diameter drops to only the diameter of the rod (Fig. 5). Pushing or supporting the same load now requires higher fluid pressure. This can raise pressures higher than the system was designed to see, cracking pressure relief valves.
Figure 5. Reduction in effective piston diameter
In summary, a leaky piston seal won’t cause drift, but it’s bad for efficiency and could damage the system.
Picking the right piston seal
In systems that are quickly cycling, a slow piston leak may go unnoticed as a tiny efficiency loss – the pressure reverses before a significant amount of fluid can leak past the seal to cause problems. On the other hand, cylinders that move slowly or must hold a position for extended periods benefit from no-leak piston seals.
At Parker, we see a wide variety of applications, and we manufacturer piston seals in many styles and materials to cover all of them. Softer durometer materials like those used for our PSP and T-seals are better for tight sealing. Harder durometer materials like our BP and PTFE cap seals are more extrusion resistant and wear longer in fast-stroking applications.
We also offer hybrid designs like our CQ profile. The CQ design utilizes glass-filled PTFE for low friction and long-wearing but also features a rubber insert to reduce leakage.
Cylinder drift is a concern in many hydraulic applications. It’s commonly mistaken as piston seal failure but is usually a combination of factors involving the valves. Understanding cylinder mechanics is vital to identifying the root causes of failure and for designing systems that are resistant to drifting.
Recommendations on application design and material selection are based on available technical data. They are offered as suggestions only. Each user should make his own tests to determine the suitability for his own particular use. Parker offers no express or implied warranties concerning the form, fit, or function of a product in any application.
This article was contributed by Nathan Wells, application engineer, Engineered Polymer Systems Division.