What a Load Holding Valve Is Preventing and When It Is Not Enough

Why this matters

A load-holding valve is one of the few components in a hydraulic circuit whose whole job is to stop something falling, and it is very good at the one failure it was designed for. The trouble starts when a shop reads that as "the load is held", because the valve only ever protects against pressure lost on the supply side of itself. Five separate things can drop that same load without the valve failing in any way, and this card is organised around those five rather than around how the valve works. Nothing here changes the first rule of the trade: nobody goes under a load that is held only by hydraulics, ever, on any machine, no matter what valve is fitted.

The rule that comes before the theory

Before anything else on a machine with a raised or suspended load:

• Block or crib the load mechanically, on supports rated for it, before anyone works under or between anything. In construction work this is a written requirement: 29 CFR 1926.600(a)(3)(i) requires equipment suspended or held aloft by slings, hoists or jacks to be substantially blocked or cribbed before employees work under or between it. In a general industry setting the same load is hazardous energy under 29 CFR 1910.147, and the machine builder's own manual names the support points.
• The valve keeps the cylinder pressurized after shutdown. That is the point of it. So the volume between the valve and the cylinder stays at load pressure with the machine off, the power locked out and the pump stopped, and opening a joint into it releases that pressure with the load behind it. Relieve through the machine's own controls or its bleed provision, never by cracking a fitting, and never search for a leak in that volume with a hand.
• Discharge any accumulator under the plant's energy control procedure before work begins. Accumulator stored energy survives a lockout, and the sibling SOP on discharging one walks it.

What the valve is, and where it has to be mounted

A load-holding valve, most commonly a counterbalance valve, is a pilot-operated check with a pressure setting. Flow toward the actuator passes freely through a check. Flow back out is blocked by a poppet held on its seat by a spring, and it only opens when pilot pressure appears on a third port, which is normally taken from the line feeding the opposite side of the cylinder. Command the actuator to move and you pressurize the opposite side, that pressure reaches the pilot, and the valve opens by the amount that lets the load move at the commanded rate.

Mounting is not a detail, it is the whole protection. The valve must be on the cylinder port itself, or on a manifold bolted directly to it, with no hose between the valve and the cylinder. A valve mounted back at the directional valve block protects the pump and the spool and leaves the entire hose run inside the danger it exists to remove. If you can see a hose between a counterbalance valve and the cylinder it is holding, you are looking at a machine whose load-holding function ends where that hose begins.

The one failure it is designed to stop

Everything on the supply side of the valve can fail and the load stays where it is. A burst hose, a cracked tube, a fitting that lets go, a stopped pump, a lost prime, a directional valve whose spool has worn enough to leak its port to tank, a control that goes dead: all of these are a loss of pressure upstream, and the poppet simply stays on its seat because there is no pilot pressure to lift it.

That is a large and real category, and it is why these valves are fitted. It is also the complete list.

The five things it does not stop

One: anything that leaks on the actuator side of it. The valve blocks one port. A cylinder with a rod on one side and a counterbalance on the cap port can still settle if fluid crosses the piston and leaves through the rod line, because that line runs to the directional valve with nothing holding it. The counterbalance is closed and correct the entire time. The same applies to a cylinder port fitting or a cracked tube on its side of the valve. The sibling card on cylinder drift owns the localization; the point here is that the valve is not in that path.

Two: a pilot signal that opens it. The valve is designed to be opened by pressure, which means anything that puts pressure on the pilot opens it. A pilot line shared between functions, a shuttle that picks up the wrong signal, a second function pressurizing a common return, an operator moving the wrong lever: none of these are failures of the valve, and all of them lower the load. The tell is that the creep happens only while something else is being operated.

Three: its own seat, held off by contamination. A poppet is a metal-to-metal seat, and a particle on it leaves a path that does not close. This is a leak past a working valve rather than a broken one, and it is why the fluid cleanliness cards belong to this subject as much as to pump wear.

Four: a hose between the valve and the cylinder, where somebody mounted it remotely. See above.

Five: a person opening a joint into the volume the valve is still holding. The valve does its job right up to the moment a fitting is loosened, and then it is holding load pressure against a joint that is coming apart.

The pilot ratio, and the two things it trades

The pilot ratio is printed on the valve or in its data sheet, and it is the multiplier between pilot pressure and the pressure the valve is holding back. A higher ratio means less pilot pressure is needed to open it.

That single number sets a trade the machine builder made, and it is worth understanding before anyone swaps a valve for whatever the counter had:

• A higher ratio opens more easily, so the opposite side of the cylinder needs less pressure to lower the load, which costs less energy and less heat. It is also less stable, because a small change in load or pilot pressure produces a large change in opening, which is what makes a boom chatter or bounce on the way down.
• A lower ratio is more stable and needs more pilot pressure, which shows up as back pressure, wasted power and heat.

Whichever ratio is fitted, the valve is a designed restriction on the way down, so it is a designed heat source. That is why the thermal survey card gives throttling elements an expected rise rather than an expected zero.

The setting itself belongs to the machine builder, and it is set above the maximum pressure the load can induce by a margin the builder specifies. Set too low, the valve cannot hold the load. Set too high, the function needs more pilot pressure to lower, which costs pressure on the opposite side and can leave the machine unable to lower at all under load. Neither number is a rule of thumb, and a valve set by trial in the field is a valve nobody can defend later.

Two functions on one machine, and the valve was innocent both times

A boom machine came in with two complaints, and both got blamed on counterbalance valves.

Complaint one: the boom creeps while held. Measured, marked and timed: about 0.5 inches per hour with the machine idle. The counterbalance valve is correctly mounted, on a manifold bolted to the cylinder cap port with no hose in between, so item four is not in play.

The valve holds the cap port. The rod line runs straight back to the directional valve with nothing on it. Capping the rod line at the cylinder, with the boom on rated mechanical supports, the machine locked out, the circuit relieved through its own controls and any accumulator discharged, stopped the creep completely over a 10 hour hold. The exit path was the rod line, the crossing path was inside the cylinder, and the counterbalance valve was closed and doing its job for every one of those 0.5 inches per hour. That is item one on the list above, demonstrated.

One reconciliation worth stating rather than leaving as a trap: the drift-localization card's rod-side cap test assumes the opposite port is freely connected to its valve. A counterbalance valve on that opposite port breaks that assumption, because it blocks the port whether the piston seal is good or not. On a function with a counterbalance fitted, run the cap sequence knowing which ports are actually free, or the test will clear a piston seal that is passing.

Complaint two: the outriggers creep down. Measured idle over a shift: no movement at all. Measured during operation: movement, every time, and only while the swing function was being used. That is item two, a pilot signal, and it was a shared pilot line that picked up swing pressure. The valve was opening because it was being told to open. A new valve would have done exactly the same thing, which is what the shop had already proved once by fitting one.

What the two complaints have in common. In both cases the component that would have been replaced was working correctly, and in both cases the actual finding was about what the valve is connected to rather than what is inside it. That is the general shape of this fault family: the valve is a boundary, and a complaint about a load moving is a question about which side of that boundary the path is on.

The failure mode of getting it wrong. Replacing a healthy counterbalance valve on a load-holding function is not a neutral mistake. It means a load-holding component was removed and refitted on a machine that still drops its load, and the shop now believes the function has been serviced. The next person to work under that boom inherits that belief.

How to verify a load-holding function on a machine in front of you

• Look for the hose. Valve directly on the cylinder port or on a manifold bolted to it, or the function is not load-holding no matter what the parts list says.
• Ask which port it holds, and what is on the other one. A single counterbalance on a differential cylinder leaves the opposite line as an exit path, and knowing that in advance saves a wrong diagnosis later.
• Trace the pilot line to its source. If it is shared with another function, creep during that function is expected behaviour and not a fault to chase.
• Read the setting and the pilot ratio off the valve or its data sheet, and compare against the builder's specification. A valve of the right physical size with the wrong ratio fits perfectly and behaves differently.
• Test holding with the load on mechanical supports and nobody underneath, marked and timed, and compare against the builder's allowable drift rate for that function rather than against zero.

References

• 29 CFR 1926.600(a)(3)(i), for the construction requirement to block or crib equipment held aloft before employees work under or between it, with 29 CFR 1910.147 covering the same load as hazardous energy in a general industry setting
• Machine builder and valve manufacturer documentation for the counterbalance setting, the pilot ratio, the mounting requirement and the allowable drift rate for each held function
• ISO 4413, general rules and safety requirements for hydraulic fluid power systems, in the edition your machine builder declared conformity to or your contract calls out, which is how it binds you rather than on its own
• See related: What Cylinder Drift Tells You About Which Seal Gave Up; Discharging a Hydraulic Accumulator Before Any Work Begins; How to Find Where a Hydraulic System Is Making Its Heat