Discharging a Hydraulic Accumulator Before Any Work Begins

Purpose

To bring a hydraulic circuit containing an accumulator to a genuinely zero-energy state before any component is opened, and to verify that state by a means that does not depend on a pressure gauge. Opening the electrical disconnect stops new energy entering the circuit; it removes none of the energy already stored. 29 CFR 1910.147(d)(5)(i) makes this an explicit separate step: after the energy isolating devices are locked out, all potentially hazardous stored or residual energy must be relieved, disconnected, restrained or otherwise rendered safe.

The consequence of skipping it is not abstract. A one gallon accumulator at 3,000 psig can still drive a 3 inch bore cylinder over a foot of travel at more than ten thousand pounds of force with the disconnect locked and nothing turning. The companion reference card on what an accumulator stores carries that arithmetic.

Scope

Applies to any hydraulic circuit fitted with a bladder, piston or diaphragm accumulator, on any machine, whenever a hose, fitting, valve, cylinder, filter housing or the accumulator itself is to be opened, removed or adjusted.

Does not cover disassembly of the accumulator shell, replacement of a bladder or piston, or any work that breaks into the gas side. Those require the nitrogen precharge to be deliberately released first through a charging assembly, and they follow the accumulator manufacturer's own documented procedure, not this one.

Does not apply on its own where the accumulator serves a safety function such as an emergency brake or a controlled stop. On those machines, discharging the accumulator removes the protection, and the machine builder's documented sequence governs the order in which things are made safe.

Work that involves opening an energized electrical enclosure or metering conductors is outside 1910.147 by its own exclusion at (a)(1)(ii)(C) and is governed by 29 CFR 1910.333(b)(2) in general industry or 29 CFR 1926.417 on a construction site.

Roles and responsibilities

Before you start

Three things must be known and none of them are guessable from the outside:

• Where every accumulator is. Machines carry more than one, and the second is often remote from the power unit, at a manifold near the actuator it serves. Walk the schematic in the cabinet, then walk the machine; the two disagree more often than anyone would like.
• Whether any of them serves a safety function. Ask before you bleed.
• Whether a manual bleed valve is fitted to each one. If one is not, this procedure stops at step 5 and the machine goes on a list for a bleed valve to be fitted during a planned shutdown. Cracking a fitting to relieve an accumulator is never the answer; fluid escaping a partly loosened joint injects through skin at a fraction of these pressures.

Procedure

1. Bring the machine to rest and deal with mechanical stored energy first. Retract or lower actuators to their rest positions using the machine's own controls. Where a load must stay raised, set mechanical blocking or a stand rated above the load before anyone approaches. A raised load is stored energy in its own right and this procedure does not address it.

2. Shut down at the operator control, then open the motor disconnect, apply your own lock and tag, and try the start command to confirm the drive will not run. This is mechanical isolation under 29 CFR 1910.147.

3. Close the accumulator isolation valve where one is fitted, and record that you did. Do not treat a closed isolation valve as a discharge; it traps the energy rather than removing it, and the oil behind it is still at full pressure.

4. Put on eye protection and position yourself out of line of the bleed outlet and of every fitting on the pressurized side. Have a container and absorbent ready. The oil coming out may be at operating temperature and will burn on contact, and any that reaches the floor is a slip hazard until it is cleaned up. Handle the recovered fluid per its safety data sheet.

5. Open the accumulator's manual bleed valve slowly, and leave it open. A quarter turn is plenty. Opening it fast on a large accumulator produces a jet at the outlet and a pressure surge in the return line. Watch the gauge fall and time it roughly; a large accumulator that goes to zero instantly has either been discharged already or is being read by a gauge that is not connected to it.

6. Release residual pressure in each function's own lines. Trapped oil in a long hose run or between a cylinder and a closed valve holds pressure independently of the accumulator. Cycle each directional valve through both shifted positions and back to center with the bleed still open. Nobody is in the machine's motion path while this is done, because the whole point of the step is that you have not yet proved nothing will move.

7. Leave every bleed valve open for the duration of the work, and tag it open. A closed bleed lets the circuit repressurize from a source nobody accounted for, including thermal expansion of trapped oil in a hot bay.

8. Do not touch the gas side. After the oil side is at zero, the nitrogen remains at its precharge pressure behind a closed poppet or piston, and that is correct, not a fault. Never loosen the gas valve guard, the valve core, the shell fasteners or any retaining ring. Those are pressure vessel components and releasing them under precharge launches them.

9. Tag the accumulator as discharged, with the time and your name, so the next person on the machine is not deciding for themselves whether the gauge in front of them is trustworthy.

Verification is not a gauge reading

A gauge reading zero is necessary and it is not sufficient. Three ways it reads zero on a live circuit, all of them common:

• The gauge is isolated. Most gauges are fitted behind a small needle valve or a snubber that gets closed to stop needle flutter and never gets reopened.
• The gauge has failed. A bourdon tube that has split reads zero forever and looks entirely normal from the front.
• The gauge is on the wrong side of a check valve. Read at the pump outlet, upstream of the check that keeps the accumulator charged, it correctly reports zero at the pump while the accumulator behind the check is at full pressure.

Verification is the bleed valve left open plus a commanded movement that produces nothing. With the bleed open, operate every directional valve in both directions and confirm no actuator motion, no sound of oil moving, and no recovery on the gauge. That test does not care whether the gauge works, and it is the one 1910.147(d)(6) is asking for when it requires the authorized employee to verify that isolation and de-energization have been accomplished.

Worked walkthrough

A clamp machine with one bladder accumulator at the power unit. Shell volume one gallon, system maximum 3,000 psig, precharge specified 1,800 psig at a 70 F reference. The circuit feeds a 3 inch bore cylinder with 7.07 square inches of piston area. The machine has been running all morning and the oil is at 120 F.

Steps 1 to 3. Clamp opened at the operator station, cylinder retracted. Disconnect opened, personal lock and tag applied, start pressed with no response. Accumulator isolation valve closed and noted.

Step 5, and what the gauge does on the way down. Bleed valve cracked a quarter turn with eye protection on and the technician standing to one side. The manifold gauge falls steadily from 3,000 psig over about 35 seconds, then at roughly 1,620 psig the rate visibly changes and the needle drops the rest of the way to zero in under two seconds.

That knee is worth understanding rather than ignoring. Above precharge the accumulator is pushing out real volume, on this unit 92 cubic inches of oil, so the fall is slow. At precharge the bladder has fully expanded and the poppet closes, so all that remains is the compressibility of the trapped oil and a little hose stretch, which is a very small volume and empties immediately. The pressure at the knee is approximately the precharge, read for free, with no nitrogen gauge and no charging assembly.

Correcting that reading. It was taken on a machine at 120 F, so it is not comparable to a specification written at 70 F. Gas at fixed volume follows absolute pressure over absolute temperature, using degrees F plus 460:

• Observed at the knee: 1,620 psig, which is 1,634.7 psia.
• Corrected to 70 F: 1,634.7 times (530 divided by 580), which is 1,493.6 psia, or 1,478.9 psig.
• Against the specified 1,800 psig, the unit is 321 psi low, 17.8 percent below specification.

That does not change anything about this procedure, and it does go on the ticket, because a precharge that low costs the machine roughly the same 17.8 percent of the oil the accumulator was sized to deliver, and the reference card on what an accumulator stores works that through.

Step 6. With the bleed open, both directional valves cycled through both positions and back to center. No motion, no flow noise, gauge stays at zero. Nobody standing in the clamp travel.

Step 8. The gas valve is left alone. A gauge on it now would read about 1,620 psig, which is normal and is not a reason to do anything.

What the shortcut would have looked like. The version that goes wrong reads the pump-outlet gauge, sees zero because it sits upstream of the charging check valve, and starts on a hose fitting at the manifold with 92 cubic inches of oil behind it at 3,000 psi. The first thread to lose engagement releases it as a jet. Anyone struck goes to an emergency department immediately and uses the words "high pressure fluid injection injury" rather than describing a cut; the card on what a hydraulic system is trading away carries why that wording matters.

Return to service

• Close the bleeds, remove the tags you applied, and reopen the accumulator isolation valve before the locks come off, so the circuit is complete when the pump first turns.
• Each authorized employee removes their own lock. Nobody removes anybody else's, and the group lockout device comes off last.
• Start with a person watching the joints that were opened, from outside the spray path, with the machine at its lowest available pressure setting where one exists. A joint that was disturbed leaks on the first pressurization or not at all.
• Recheck precharge at the next scheduled opportunity if the knee reading came in low, using a charging assembly rated above the precharge, with the oil side bled to zero and nobody's face over the valve. Nitrogen only, never compressed air or oxygen.
• The written procedure itself gets reviewed. 29 CFR 1910.147(c)(4)(i) requires energy control procedures to be developed, documented and used, and (c)(6)(i) requires a periodic inspection of the procedure at least annually. An accumulator added to a machine after the procedure was written is the single most common way a documented procedure quietly stops describing the machine.

References

• 29 CFR 1910.147, the control of hazardous energy: (c)(4)(i) documented procedures, (c)(6)(i) annual periodic inspection, (d)(5)(i) relief of stored and residual energy, (d)(6) verification of isolation, (f)(3) group lockout
• 29 CFR 1910.147(a)(1)(ii)(C), the exclusion that sends work on energized conductors to 29 CFR 1910.333(b)(2) in general industry and 29 CFR 1926.417 in construction
• Accumulator manufacturer documentation for precharge specification and reference temperature, charging assembly requirements and any disassembly procedure
• Safety data sheet for the hydraulic fluid, for burn and skin handling and for what an emergency department needs after an injection injury
• See related: What an Accumulator Stores and Why It Survives a Lockout; What a Hydraulic System Is Actually Trading Away