Corner Load Testing for Platform Scales: Procedure, Tolerances and Troubleshooting

A platform scale may show the expected weight when a pallet is placed in the centre but give a different result when the same pallet is moved closer to one side. This problem can affect stock records, receiving quantities, dispatch weights, production measurements and transactions based on weight. A correct centre reading does not prove that the entire platform is weighing correctly. This happens because the deck, frame, supports, load cells, cables and junction box must transfer and combine force correctly at every loading position. Dirt beneath the platform, contact with a ramp or pit frame, uneven support, a damaged load cell or poor corner balancing can make the indicated weight depend on where the load is placed.

Corner load testing identifies this position-dependent error. The same known test load is placed in prescribed areas of the platform, and every indication is compared with the applicable tolerance. OIML R76 uses the formal term eccentricity test, while corner test, shift test and off-centre load test are also widely used. 

What Is Corner Load Testing for a Platform Scale?

Corner load testing checks whether a platform scale gives acceptable results when the same load is placed in different areas of its load receptor. The test helps separate a position-related fault from a general calibration error that affects the entire weighing range. A passing test does not always mean that every position must display an identical number. The main requirement under OIML R76 is that the indication obtained at each tested position remains within the maximum permissible error for the applied load. Comparing corners with one another is useful, but each result must also be compared with the known test load. 

Corner Test, Shift Test and Eccentricity Test

The terms describe the same general type of performance check, although the exact load and placement method can vary. Eccentricity test is the formal OIML term. Corner load test is commonly used by industrial scale owners and service technicians, while shift test often appears in weights-and-measures guidance. Off-centre load test is a plain-language description of the same principle. The main terminology associated with this subject includes platform scale, floor scale, test weight, quadrant, corner deviation, load cell, junction box, maximum permissible error and mechanical binding.

Why Does Moving the Load Affect the Reading?

Moving a load changes how its force is distributed through the platform. On a four-load-cell floor scale, a load placed near one section applies more force to the nearest support and less force to the other supports. The platform structure and electronic system must combine these different load-cell outputs into one acceptable total weight. If one load cell responds differently, one foot is not supported correctly or the platform touches another structure, the load may follow an unintended force path. The display can then rise or fall as the load moves across the deck, even though the actual weight has not changed.

Does “Corner” Mean the Extreme Outer Edge?

A corner load test does not always require placing the weight at the extreme outer edge of the platform. For an instrument with no more than four support points, OIML R76 divides the load receptor into four roughly equal quarter segments. Each segment is loaded in turn. When one large test weight is used, it should be placed centrally within the selected segment. When several smaller weights are used, they should be spread uniformly across the segment. OIML also states that unnecessary vertical stacking should be avoided. 

Placing a large concentrated load directly at an unsupported edge can create a test condition that differs from the approved procedure. It can also place unnecessary stress on the deck or an individual load cell. The intended segment or support position should therefore be identified before the test begins.

How Do You Select the Correct Test Load and Positions?

The correct test load depends on the applicable standard, maximum capacity, additive tare capacity, number of support points and instrument design. A fixed rule such as “always use 25% of capacity” is not technically correct for every platform scale. The instrument nameplate, type-approval information, manufacturer instructions and applicable legal-metrology procedure should be checked before selecting the test load. The physical size of the platform alone does not provide enough information.

OIML Test Load for a Standard Platform Scale

Unless a different requirement applies, OIML R76 specifies an eccentricity test load equal to one-third of the sum of the maximum capacity and the corresponding maximum additive tare effect:

Test load = ⅓ × (Max + maximum additive tare effect)

Where the scale has no additive tare effect, the calculation can be simplified to:

Test load = Max ÷ 3

OIML requires the indication at every tested position to remain within the maximum permissible error for the applied load.

For example, consider a platform scale with:

• Max = 1,500 kg
• No additive tare effect
• No more than four support points

The test-load calculation is:

1,500 kg ÷ 3 = 500 kg

The same 500 kg test load would be applied to each prescribed quarter segment. This is an OIML-based example, not a universal instruction for every 1,500 kg scale under every inspection or manufacturer procedure.

What If the Platform Has More Than Four Support Points?

OIML uses a different calculation when the load receptor has more than four support points. The test-load fraction is:

1 ÷ (n − 1)

In this formula, n means the number of physical support points. The resulting fraction of maximum capacity plus the maximum additive tare effect is applied at each support point.  For example, a platform with six support points would use the fraction:

1 ÷ (6 − 1) = 1/5

The placement area should correspond to the support being assessed. This n should not be confused with the number of verification scale intervals used in scale accuracy classification.

What Test Weights Should Be Used?

The test load should be assembled from calibrated and traceable mass standards that are suitable for the error limit being assessed. For formal type examination or verification, OIML R76 states that standard weights or masses should meet the relevant OIML R111 requirements and should not contribute an error greater than one-third of the instrument’s maximum permissible error for the applied load. 

Before using the weights, confirm:

• Their nominal and calibrated values
• Identification or serial numbers
• Calibration status
• Traceability
• Suitability for the expected tolerance
• Safe lifting and positioning arrangements
• The load-bearing limits of the platform and surrounding floor

A certificate does not make a weight suitable for every test. Its error, uncertainty and total available mass must support the platform scale and tolerance being tested. Scales4U supplies individual calibration weights and weight sets in a range of denominations, including larger cast-iron weights for industrial-scale work. 

How Is a Platform Scale Corner Load Test Performed?

A reliable corner test requires more than moving weights between four positions. The scale must first be identified, inspected and allowed to stabilise. The loading sequence and all results must then be recorded so the test can be reviewed or repeated. Formal calibration or verification should be completed by suitably qualified personnel. Basic operator checks may reveal a problem, but unauthorised adjustment of controlled settings can affect the scale’s approval or verification status.

Step 1: Identify the Scale and Test Requirements

Begin by recording the scale’s identifying and metrological information. This confirms which procedure, test load and tolerance apply.

Record:

• Manufacturer
• Model
• Serial number
• Installation location
• Platform dimensions
• Maximum capacity, Max
• Minimum capacity, Min
• Verification scale interval, e
• Display interval, d
• Accuracy class
• Number and arrangement of support points
• Trade or non-trade status
• Applicable manufacturer or approval documentation

Scales4U supplies industrial platform scales in several platform sizes and capacity configurations, including trade-approved and non-trade models. This shows why two platforms that look similar may need different test loads and acceptance limits. 

Step 2: Inspect the Platform and Installation

Inspect the complete installation before applying any test load. A scale should not be electronically adjusted while a mechanical obstruction or foundation problem remains.

Check that:

• The deck is empty and clean.
• Nothing touches the side of the platform.
• No dirt or product is trapped underneath.
• Ramps and pit frames have suitable clearance.
• Each foot or mounting point is supported correctly.
• The foundation is stable.
• Cables are free and not pulling against the structure.
• Check rods, bumpers and stops are not binding.
• The deck and frame are not bent.
• The indicator is powered and stable.

Floor-scale manufacturers identify poor foundations, debris, loose support, contact between live and fixed structures, damaged load cells and junction-box faults as possible causes of corner errors or unstable readings. 

Step 3: Mark the Test Positions and Confirm Zero

For a normal rectangular platform with no more than four support points, divide the deck into four approximately equal segments. Mark the centre of each segment and label the positions clearly, such as Position 1, Position 2, Position 3 and Position 4. A clockwise sequence is easy to follow, but the direction is less important than using the same sequence throughout the test. OIML requires the load locations to be shown on a sketch in the test report. 

Before loading:

1. Remove all loads from the platform.
2. Confirm there is no unintended tare value.
3. Allow the display to stabilise.
4. Record the initial no-load indication.
5. Note any zero drift or unstable movement.

OIML’s formal eccentricity procedure also addresses zero error and specifies that automatic zero-setting or zero tracking should not operate during the test. These controls should be handled according to the approved procedure by a qualified technician.

Step 4: Apply the Same Load to Each Position

Place the selected test load centrally within Position 1. Avoid dropping the weights or dragging them across the platform because impact and side force can affect the scale and create a safety risk. Wait for the indication to stabilise, then record the displayed value. Move the same load to Position 2 and repeat the process for all remaining segments. The test-load value, footprint and arrangement should remain consistent at every position.

The sequence is:

1. Load Position 1.
2. Record the stable indication.
3. Move the same load to Position 2.
4. Record the indication.
5. Repeat at Positions 3 and 4.
6. Remove the load.
7. Record the final zero indication.

OIML states that loading the eccentric segments is sufficient; a centre load is not required as part of its standard eccentricity sequence. 

Is a Centre Reference Reading Useful?

A centre reading can still be useful as a diagnostic check. If the centre and all four segments show nearly the same error, the scale may have a general span or calibration problem. If the centre is acceptable but one segment shows a large deviation, the fault is more likely related to load position. The centre indication should not automatically be treated as the true reference value. It can also contain an error. Formal compliance should be judged against the known test load and the applicable tolerance, not only against the centre reading.

Step 5: Record Zero Return and Questionable Results

After the test load is removed, observe whether the scale returns to its starting indication. A delayed or incomplete return to zero may point to binding, load-cell creep, platform contact, structural movement or another fault. Repeat any reading that was unstable or clearly different from the others. Do not erase the first result. Record the original reading, repeated reading and any change made between tests.

A useful test record includes:

How Are Corner Load Test Results Calculated?

Corner testing produces several useful values. The indication error shows how far a position is from the known load. The position difference compares a corner with an optional centre reading, while the corner spread shows the total variation across all tested positions. These calculations answer different questions and should not be treated as interchangeable.

Indication Error at Each Position

The basic indication error is:

Indication error = Scale indication − Known test-load value

For example:

• Known test load: 500 kg
• Scale indication: 500.5 kg

The error is:

500.5 kg − 500 kg = +0.5 kg

A positive result means the scale reads high. A negative result means it reads low. A formal test may also account for zero error, digital rounding and the calibrated values of the mass standards. The simplified formula is useful for understanding the result, but it does not replace the complete verification method.

Difference From a Centre Reading

If a centre reading has been taken for diagnostic purposes, calculate:

Position difference = Position indication − Centre indication

For example:

• Centre indication: 500.0 kg
• Position 2 indication: 500.5 kg

The position difference is:

500.5 kg − 500.0 kg = +0.5 kg

This shows how much the reading changed after moving the load. It does not confirm that the centre indication was accurate.

Spread Between the Tested Positions

The corner spread is:

Corner spread = Highest position indication − Lowest position indication

For example:

• Highest indication: 500.5 kg
• Lowest indication: 499.5 kg

The spread is:

500.5 kg − 499.5 kg = 1.0 kg

A large spread clearly shows position-dependent variation. A small spread does not automatically mean the scale passes. Every position could be wrong by a similar amount, so each indication must also be compared with the known test load.

What Tolerance Applies?

The applicable tolerance depends on the instrument’s accuracy class, verification scale interval, applied load and type of assessment. OIML R76 requires each eccentric-position indication to remain within the maximum permissible error for that load. 

There is no universal rule stating that all positions must be:

• Within one display division
• Within 0.1%
• Within a fixed kilogram value
• Identical to the centre reading

The correct limit must be calculated or obtained from the governing procedure.

Worked Class III Platform-Scale Example

Consider a Class III platform scale with the following specification:

• Max = 1,500 kg
• e = 0.5 kg
• No additive tare effect
• Four support points
• OIML eccentricity test load: 500 kg

The applied load in verification intervals is:

500 kg ÷ 0.5 kg = 1,000e

For a Class III instrument, an applied load above 500e and up to 2,000e has an initial-verification MPE of ±1e. In this example:

±1 × 0.5 kg = ±0.5 kg

OIML’s Class III initial-verification error bands are ±0.5e up to 500e, ±1e above 500e up to 2,000e, and ±1.5e above 2,000e up to 10,000e. 

Every individual result is within ±0.5 kg, so the results meet the stated illustrative initial-verification limit. The total spread is 1.0 kg, but the pass decision is based on each position meeting the applicable MPE. The calculation is a simplified educational example. A formal test may require correction for zero and indication rounding, and South African acceptance should follow the approved instrument documentation and applicable legal-metrology procedure.

Why Does a Platform Scale Fail a Corner Load Test?

A failed corner test does not automatically mean that the nearest load cell must be replaced. Position-dependent errors can come from the deck, foundation, mounting system, wiring, junction box, indicator settings or load cells. The fault should be investigated in a logical order, beginning with visible mechanical conditions before any electronic balancing is attempted.

Debris or External Contact

Material trapped under or around the platform can restrict movement or transfer part of the load into a fixed structure. The scale may then read differently depending on which side of the deck is loaded.

Common obstructions include:

• Stones and gravel
• Packaging material
• Grain or product residue
• Hardened concrete or dirt
• Metal fragments
• Cables caught under the frame
• A ramp touching the platform
• Insufficient clearance around a pit frame

Cleaning should be followed by a complete retest. A scale may appear correct immediately after zeroing but still fail as soon as a load presses the deck against the obstruction.

Foundation, Feet or Frame Problems

Every support point must sit correctly on a stable foundation. If one foot is loose, one corner is unsupported or the frame has twisted, the force distribution changes as the load moves.

Possible causes include:

• Uneven flooring
• Foundation settlement
• Incorrect shimming
• Loose anchors
• A rocking platform
• Bent structural members
• Cracked welds
• Damaged mounting brackets
• Tight check rods or overload stops
• Side loading on a load-cell mount

Official manufacturer documentation warns that poor site conditions, unsupported corners, frame interference and foreign material can produce shift errors, zero changes and unstable indications. 

Load-Cell Faults

A multi-load-cell platform depends on each sensor producing a suitable output. A damaged or mismatched load cell can make one area read consistently high or low.

Possible load-cell problems include:

• Shock or overload damage
• Moisture ingress
• Damaged cables
• Insulation failure
• Zero-balance shift
• Sensitivity mismatch
• Incorrect replacement model
• Loose mounting
• Temperature-related instability

Scales4U supplies several industrial load-cell types used in platform scales and other weighing systems, including single-point and shear-beam models. The replacement must match the scale design, capacity and approval requirements.

Junction-Box or Wiring Problems

A four-load-cell platform often combines the individual sensor signals through a junction or summing box. Loose connections, moisture, corrosion or previous incorrect adjustment can change the contribution of one channel.

Check for:

• Loose terminal connections
• Corroded wiring
• Water inside the enclosure
• Damaged trim potentiometers
• Broken cable screens
• Incorrect channel wiring
• Unauthorised adjustment
• An unstable individual sensor signal

The junction box should not be adjusted until the platform, mounts and load cells have been checked. Electronic trimming cannot correctly compensate for a bent frame or a platform touching a fixed object.

Incorrect Calibration or Configuration

A scale may also fail after a component replacement or an incomplete service procedure. Installing a new load cell without matching and balancing it, changing indicator settings or skipping final span calibration can leave the scale position-dependent.

Common mistakes include:

• Trimming before removing mechanical interference
• Using the wrong load-cell capacity
• Installing a load cell with different sensitivity
• Entering incorrect digital channel values
• Failing to repeat span calibration
• Adjusting one corner without retesting the others
• Ignoring zero return after adjustment

How Should a Failed Corner Test Be Corrected?

The safest correction process starts with the mechanical installation and moves to electronic adjustment only after physical faults have been removed. Adjusting a junction box first may reduce one visible error while leaving the real cause in place. A failed result should also be treated according to its business risk. If the scale is used for commercial transactions or critical production measurements, it should not continue to provide official results while it is outside its permitted tolerance.

Follow a Mechanical-First Repair Sequence

A practical troubleshooting order is:

1. Remove the scale from critical use where necessary.
2. Clean under and around the platform.
3. Check deck, ramp and pit-frame clearance.
4. Confirm that all feet and mounts are supported.
5. Inspect the foundation, frame and restraints.
6. Examine load-cell cables and junction-box connections.
7. Compare individual load-cell outputs.
8. Repair or replace damaged parts.
9. Complete corner balancing if required.
10. Recalibrate zero and span.
11. Repeat the full eccentricity test.
12. Arrange reverification where legally required.

This order prevents a technician from using electronic adjustment to hide a mechanical defect.

Analog and Digital Corner Adjustment

Analog multi-load-cell platforms may use potentiometers or trim resistors in a junction box. A technician identifies the channel linked to an outlying position and makes small adjustments, repeating the complete loading sequence after each meaningful change. Digital systems may use individual load-cell addresses, channel gains or software coefficients. They can provide useful diagnostic information about each sensor, but digital compensation still cannot repair deck contact, frame damage or an unstable foundation. Model-specific instructions must be followed. Turning the wrong adjustment or changing protected settings can make the scale less accurate and may affect verification seals.

Which Tests Should Be Repeated After Repair?

A corner adjustment can change the scale’s zero or span. The final assessment should therefore include more than one successful corner reading.

After repair or balancing, repeat:

• Zero check
• Complete eccentricity test
• Span or weighing-performance test
• Repeatability test
• Return-to-zero check
• Linearity testing where required
• Verification where required for trade use

If unequal readings remain after basic cleaning and clearance checks.

When Should Corner Load Testing Be Performed?

Corner testing should be triggered by changes, faults and service events rather than by one universal calendar interval. A lightly used platform in a clean indoor location does not experience the same risks as a floor scale exposed to forklifts, washdown, dust and repeated impact loads.

A corner test should be considered after:

• Initial installation or commissioning
• Moving or reinstalling the scale
• Changes to the floor or foundation
• Platform or frame repair
• Load-cell replacement
• Junction-box replacement or adjustment
• Indicator replacement or reconfiguration
• A severe overload or impact
• Flooding or moisture entry
• Persistent zero drift
• Reports of different readings across the deck
• Failed routine checks
• Scheduled calibration
• Legal verification or reverification

The appropriate service frequency depends on use, environment, fault history, measurement risk, manufacturer guidance and quality-system requirements. Scales4U offers service agreements with visit frequencies selected according to the customer’s equipment and maintenance needs. 

What Does Corner Testing Mean for Legal-for-Trade Scales in South Africa?

Corner testing is especially important where a platform scale is used to determine a selling price, purchase quantity, payment or another prescribed measurement. However, passing a corner test alone does not establish that the scale is legally compliant. South Africa’s NRCS defines verification as the process of determining whether an instrument complies with its type-approval requirements and remains accurate within the tolerances prescribed by the Legal Metrology Act. 

Is a Corner Test the Same as Legal Verification?

No. A corner or eccentricity test checks one part of the scale’s performance. Legal verification considers the instrument’s complete compliance with its approved type and applicable requirements.

A scale can pass its corner test and still fail because of:

• Incorrect span performance
• Poor repeatability
• Zero-return error
• Missing or incorrect markings
• Unapproved components
• Broken or missing seals
• Incorrect software or settings
• A configuration that differs from the approved type

Corner testing should therefore be treated as one important part of a wider calibration or verification process.

Final Takeaway: Test the Whole Platform, Not Only the Centre

A correct centre reading cannot confirm that a platform scale performs correctly across its full deck. Corner load testing exposes position-dependent errors by moving the same known load through prescribed platform segments and checking every indication against the correct tolerance. For a standard OIML platform with no more than four support points, the normal eccentricity load is one-third of maximum capacity plus any applicable additive tare effect. The load is placed centrally within each quarter segment, not automatically at the extreme outer corners. Each position must meet the maximum permissible error for the applied load. 

When a scale fails, inspect debris, deck clearance, supports, foundation, frame, load cells and wiring before changing corner adjustments. After any repair or trimming, repeat the full corner test and confirm zero, span and repeatability. If your platform scale changes reading as a pallet or load moves across the deck, contact Scales4U for fault diagnosis, corner testing, calibration, repair or verification support.