Selecting Hydraulic Jacks for Tank Load-Cell Testing

The verification display rises while the vessel indicator falls as you pump. That is the expected direction. The fault is a mismatch between the two changes, not the opposite signs.

Start with the force path and corner loading. A common-pressure manifold does not make four lifting points move together, and an 80-ton nameplate total does not automatically create a 64-ton safe system. Every jack, hose, fitting, valve, support point, and structural member must remain within its own limit under the worst credible load distribution.

Check the force path first

Sketch the complete load path before selecting hydraulics. At each corner, force must pass from the jack through the verification load cell and into a part of the vessel support structure that can accept the concentrated load. The reaction beneath the jack must be equally sound.

The intended measurement is load substitution:

• The verification cells gain compressive load as the jacks take weight.
• The installed vessel cells lose approximately the same total load.
• The vessel only needs enough movement to transfer measurable load. It does not need to clear the mounting kits unless the test method specifically requires that condition.

Rigid process piping, flexible connections at travel limits, bracing, ladders, conduits, and vessel restraints can divert force away from the installed cells. They can also impose horizontal force or bending on the jacks and verification cells. If the summed readings disagree, external load paths are one of the first causes to investigate.

Panel symptom	Likely cause or next check
Reference load rises and vessel indication falls by the same amount	The substitution force is reaching both measurement systems as intended.
One reference cell rises much faster	That corner contacted first, has a different hydraulic area, or is carrying more load because of vessel geometry or center of gravity.
Reference sum rises more than the vessel reading falls	Check indicator calibration, corner summing, mechanical binding, piping reactions, and whether both systems use the same force or mass units.
Pressure rises with little reference-cell response	Look for a closed valve, bottomed jack, blocked coupler, trapped air, side loading, or contact outside the intended force path.
Readings drift after pumping stops	Check leakage, valve creep, hose expansion, structural settling, temperature drift, and load-cell mounting movement.

Calculate the load at each lifting point

Do not size the system by multiplying four jack ratings and applying a blanket 0.8 factor. The limiting condition is the most heavily loaded individual point. Vessel center of gravity, support stiffness, piping reactions, contact sequence, and jack friction can produce unequal corner loads.

For the stated 25,000 kg example, gravitational force is approximately:

F = m × g = 25,000 kg × 9.81 m/s² = 245,250 N

With four perfectly equal support reactions:

Fcorner = 245,250 N / 4 = 61,313 N, approximately 6,250 kgf or 6.25 metric tonne-force per point.

That equal-share result is only the starting case. Measure or calculate the maximum reaction at each vessel support. Select each jack and verification cell from the largest credible reaction at that point, including the applied test load. Keep the selected working load below every component's published working-load limit. Also confirm whether a quoted “ton” rating means metric tonne-force or short ton-force.

If the test unloads only part of the vessel weight, calculate the required transferred force from the desired calibration span. Do not raise the tank farther or transfer more load merely because unused jack capacity exists.

Separate pump pressure from jack capacity

A pump creates pressure; a jack converts that pressure into force. Use the jack manufacturer's effective piston area:

Fjack = P × Aeffective

Rearrange it to select pressure:

Prequired = Frequired / Aeffective

The listed pump output of 600 kgf/cm² is approximately 58.84 MPa. It does not state how much force a jack will produce because the jack piston area is missing.

The 400 cc pump value is oil capacity, not cylinder volume or piston area. The listed 340 mm length is also not a hydraulic-cylinder stroke or internal height. Using those values to calculate a radius produces a fictional piston and explains the unrealistic force result.

Read the effective area, rated operating pressure, oil capacity at the required stroke, and usable stroke from the jack data. Then check both conditions:

• The pump can reach the pressure required for the heaviest jack load without exceeding the working pressure of any component.
• The pump has enough usable oil for all connected jacks: Vrequired = Σ(Aeffective × stroke used), plus the oil needed to fill hoses and fittings and the manufacturer's required reservoir margin.

A high-pressure rating cannot compensate for insufficient reservoir volume. Likewise, a large reservoir cannot compensate for a pump that cannot reach the required pressure.

Control four jacks without assuming equal lift

With identical jacks on one open manifold, each branch sees approximately the same static pressure. Similar pressure and piston area produce similar theoretical force, but they do not produce equal travel. The least-loaded or least-restrained corner normally moves first.

Individual needle valves can help establish contact gradually, but they are flow controls. Do not treat a needle valve as the sole load-holding device unless its manufacturer explicitly rates it for that duty and flow direction. A trapped branch can retain a different pressure after its valve closes, and reopening several branches can cause abrupt pressure redistribution.

Use a manifold arrangement that gives you controlled isolation, rated load holding, and a controlled return path. Fit a pressure gauge where it reads the pressure delivered to the manifold. If isolated branches must be diagnosed separately, provide a way to measure branch pressure rather than assuming the common gauge represents a closed branch.

Watch all four reference channels during contact and loading. An electrically paralleled vessel system produces one combined signal; it does not prove that the mechanical load is evenly divided. Corner imbalance can overload one jack or support while the total indication still looks reasonable.

Select jacks for alignment and controlled return

Low duty cycle does not reduce the required load rating, pressure rating, stability, or load-holding performance. Price differences commonly reflect verified ratings, material and seal quality, machining, side-load tolerance, return mechanism, stroke control, serviceability, documentation, and availability of compatible couplers and repair parts.

Check these items before purchase:

• Rated capacity at the specified operating pressure.
• Minimum closed height and sufficient usable stroke for contact, load transfer, and removal.
• Effective piston area and oil volume over the intended stroke.
• Rated return method and a controlled way to retract under the actual orientation and load.
• Permitted mounting orientation and restrictions on side load or off-center loading.
• Saddle geometry that keeps force centered through the verification cell.
• A positive stroke stop or a clear method to prevent overtravel.
• Manufacturer documentation identifying working pressure, inspection criteria, and compatible fluid.

A low-profile shape is useful only if the jack, reference cell, load plates, and alignment hardware fit beneath the vessel with enough clearance to install and remove them without uncontrolled vessel movement.

Match hoses, valves, fittings, and couplers

Build the circuit around one documented maximum working pressure. Every pressure-containing component must have a working-pressure rating at least equal to that circuit limit. Do not compare a hose burst pressure with a pump working pressure.

Match hydraulic fluid compatibility, pressure rating, connection type, thread form, sealing method, flow direction, and coupler series. The pump listing mentions a ZG3/8 connection, but that label alone is not enough to select an adapter. Obtain the manufacturer's thread specification and sealing-face drawing. Similar-looking tapered and parallel threads are not interchangeable.

Keep hoses short enough to limit stored energy and expansion but long enough to avoid tension, sharp bends, abrasion, and trip hazards. Route them away from pinch points and places where vessel movement can crush them. Fit protective caps to disconnected couplers and keep dirt out of every open connection.

Select a gauge with the correct pressure range, accuracy, connection rating, and readable resolution at the intended test pressure. The gauge is a hydraulic diagnostic; the calibrated verification cells remain the force reference because jack output also depends on effective area, friction, and alignment.

Run the substitution test in controlled increments

1. Empty or isolate the process as required by the site's approved test plan. Record vessel contents, total indicated load, piping condition, support condition, and all load-cell readings before applying hydraulic force.
2. Inspect every jack, cell, hose, fitting, coupler, valve, manifold, reaction surface, and load plate. Reject damaged or unidentified pressure components.
3. Center each jack and verification cell in the intended vertical force path. Remove side load and provide rated load-spreading plates where the support surface requires them.
4. Open only the first controlled branch and pump slowly until its verification cell shows stable contact. Stop before lifting that corner appreciably. Isolate it with the rated load-holding arrangement.
5. Repeat the contact step at the remaining points while watching all channels. A change at another point indicates structural interaction or load redistribution.
6. Confirm that every point has contact, adequate remaining stroke, acceptable alignment, and load below its individual limit. Set the test displays to the procedure's starting references only after recording the raw pre-test values.
7. Connect the branches into the intended common-pressure configuration without creating an abrupt redistribution. Increase pressure in small steps.
8. At every step, record manifold pressure, each verification-cell load, the sum of the verification loads, and the vessel-system change. Allow readings to settle before recording them.
9. Compare Σ reference load with |Δ vessel indication|. They should track with opposite signs. Use several increasing and decreasing points to reveal hysteresis, binding, drift, or a single-point coincidence.
10. If a span correction is justified, calculate it from the measured slope. For an indicator whose displayed span scales directly with its calibration factor, new factor = old factor × reference change / indicated change. Confirm the indicator's factor convention before entering the result.
11. Repeat the increasing and decreasing sequence. Then release pressure slowly and evenly, confirm that the vessel returns fully to its original supports, and verify that no branch retains pressure.

Do not work beneath a hydraulically supported vessel. Use engineered mechanical support or cribbing whenever personnel exposure or the procedure requires the load to remain elevated. Hydraulic pressure is not a mechanical restraint.

Verify the result before accepting calibration

Accept the resolving branch only after the corrected system passes the same test without another factor change. Check zero return, repeatability, increasing-versus-decreasing agreement, individual corner behavior, and the relationship between the summed reference load and vessel indication change.

Investigate rather than recalibrate when the error changes with pressure direction, time, or corner sequence. Those patterns point to mechanical binding, external piping loads, structural deflection, hydraulic leakage, or load-cell mounting movement. A calibration factor can correct a stable slope error; it cannot repair a variable force path.

Record the jack and reference-cell positions, component identifiers, starting load, applied load points, pressure readings, raw readings, correction calculation, final verification run, and post-release zero. That record makes the next infrequent test reproducible and exposes changes in the vessel structure or process connections.

FAQ