Understanding the Three-Run Method in Rotor Balancing

Devices

Portable balancer & Vibration analyzer Balanset-1A

€1,975.00 + VAT (if applicable)

Parts

Vibration sensor

€90.00 + VAT (if applicable)

Parts

Optical Sensor (Laser Tachometer)

€124.00 + VAT (if applicable)

Devices

Balanset-4

€6,803.00 + VAT (if applicable)

Parts

Magnetic Stand Insize-60-kgf

€46.00 + VAT (if applicable)

Parts

Reflective tape

€10.00 + VAT (if applicable)

Devices

Dynamic balancer “Balanset-1A” OEM

€1,735.00 + VAT (if applicable)

The three-run method is the most widely used procedure for two-plane (dynamic) balancing. It determines the correction weights needed in two correction planes using exactly three measurement runs: one initial run to establish the baseline unbalance condition, followed by two sequential trial-weight runs — one for each plane. Three runs are the theoretical minimum that still fully describes a two-plane system, which is why this method has become the default for field work.

It strikes an excellent balance between accuracy and efficiency, needing fewer machine starts and stops than the four-run method while still gathering enough data to calculate effective corrections for the great majority of industrial balancing tasks.

1. The Three-Run Procedure, Step by Step

The procedure follows a straightforward, systematic sequence. At every run, vibration is captured as a vector — both amplitude and phase — at each of the two bearings, because both pieces of information are needed to locate, not merely size, the unbalance.

Run 1 — Initial baseline measurement

The machine runs at its balancing speed in its unbalanced, as-found condition. Vibration is measured at both bearing locations (Bearing 1 and Bearing 2), recording amplitude and phase angle. These represent the vibration vectors produced by the original unbalance distribution.

• Measure at Bearing 1: amplitude A₁, phase θ₁
• Measure at Bearing 2: amplitude A₂, phase θ₂
• Purpose: establishes the baseline condition (O₁ and O₂) that must be corrected

Run 2 — Trial weight in Correction Plane 1

The machine is stopped and a known trial weight (T₁) is temporarily fitted at a precisely marked angular position in the first plane (typically near Bearing 1). The machine is restarted at the same speed and vibration is measured again at both bearings.

• Add: trial weight T₁ at angle α₁ in Plane 1
• Measure at Bearing 1: new vector (O₁ + effect of T₁)
• Measure at Bearing 2: new vector (O₂ + effect of T₁)
• Purpose: reveals how a weight in Plane 1 affects vibration at both bearings

The instrument calculates the influence coefficients for Plane 1 by vector-subtracting the initial readings from these new ones.

Run 3 — Trial weight in Correction Plane 2

The first trial weight is removed and a second trial weight (T₂) is fitted at a marked position in the second plane (typically near Bearing 2). A further run again records vibration at both bearings.

• Remove: trial weight T₁ from Plane 1
• Add: trial weight T₂ at angle α₂ in Plane 2
• Measure at Bearing 1: new vector (O₁ + effect of T₂)
• Measure at Bearing 2: new vector (O₂ + effect of T₂)
• Purpose: reveals how a weight in Plane 2 affects vibration at both bearings

The instrument now holds a complete set of four influence coefficients describing how each plane affects each bearing.

2. Calculating the Correction Weights

With the three runs complete, the balancing software performs vector mathematics to solve for the correction weights.

The influence-coefficient matrix

From the three runs, four coefficients are determined:

• α₁₁: how Plane 1 affects Bearing 1 (primary effect)
• α₁₂: how Plane 2 affects Bearing 1 (cross-coupling)
• α₂₁: how Plane 1 affects Bearing 2 (cross-coupling)
• α₂₂: how Plane 2 affects Bearing 2 (primary effect)

Solving the system

The instrument solves two simultaneous vector equations for W₁ (correction for Plane 1) and W₂ (correction for Plane 2):

• α₁₁ · W₁ + α₁₂ · W₂ = −O₁ (to cancel vibration at Bearing 1)
• α₂₁ · W₁ + α₂₂ · W₂ = −O₂ (to cancel vibration at Bearing 2)

The solution yields both the mass and the angular position required for each correction weight. Where the calculated angle falls on an obstruction or between fixed blade seats, the answer can be redistributed onto reachable positions using split correction.

Final steps

1. Remove both trial weights.
2. Install the calculated permanent correction weights in both planes.
3. Run a verification pass to confirm vibration has dropped to acceptable levels.
4. If needed, perform a trim balance to fine-tune the result.

3. Advantages of the Three-Run Method

Several strengths have made three runs the industry standard for two-plane work.

Optimal efficiency

Three runs are the minimum needed to establish four influence coefficients — one baseline plus one trial run per plane. That minimises downtime while still characterising the whole system.

Proven reliability

Decades of field experience show that three runs supply sufficient data for dependable balancing across the vast majority of industrial machines.

Time and cost savings

Compared with the four-run method, dropping one trial run cuts balancing time by roughly 20%, translating directly into less downtime and lower labour cost.

Simpler execution

Fewer runs mean less trial-weight handling, fewer chances for error, and simpler data management.

Adequate for most applications

For typical machinery with moderate cross-coupling and reasonable balancing tolerances, three runs consistently deliver successful results.

4. When to Use the Three-Run Method

The three-run method suits:

• Routine industrial balancing: motors, fans, pumps, blowers — the bulk of rotating equipment.
• Moderate precision requirements: balance quality grades from G 2.5 to G 16, defined under the modern ISO 21940-11 (which superseded the long-familiar ISO 1940-1).
• Field balancing applications: in-situ balancing where minimising downtime matters.
• Stable mechanical systems: equipment in good condition with a linear response.
• Standard rotor geometries: rigid rotors of typical length-to-diameter ratio.

5. Limitations and When Not to Use It

Three runs can fall short in certain cases.

When the four-run method is preferred

• High precision: very tight tolerances (G 0.4 to G 1.0) where the extra linearity check of a fourth run is valuable.
• Strong cross-coupling: planes very close together, or highly asymmetric stiffness.
• Unknown system characteristics: first-time balancing of unusual or custom equipment.
• Problem machinery: equipment showing signs of non-linear behaviour or mechanical faults.

When single-plane might suffice

• Narrow, disc-type rotors where dynamic unbalance is minimal.
• Cases where only one bearing location shows significant vibration.

6. Comparison with Other Methods

Three-run vs four-run method

Aspect	Three-Run	Four-Run
Number of runs	3 (initial + 2 trials)	4 (initial + 2 trials + combined)
Time required	Shorter	~20% longer
Linearity check	No	Yes (Run 4 verifies)
Typical applications	Routine industrial work	High-precision, critical equipment
Accuracy	Good	Excellent
Complexity	Lower	Higher

Three-run vs single-plane method

The three-run method is fundamentally different from single-plane balancing, which uses only two runs (initial plus one trial) but can correct only one plane and cannot address couple unbalance. Whenever a rotor is long enough that its two ends can carry unbalance independently, two-plane work — and therefore the three-run method — is required.

7. Best Practices for Success

Trial-weight selection

• Choose trial weights that produce a 25–50% change in vibration amplitude.
• Too small: poor signal-to-noise ratio and calculation errors.
• Too large: risk of non-linear response or unsafe vibration levels.
• Use similar sizes in both planes for consistent measurement quality. A trial weight calculator gives a sound first estimate from rotor mass and speed.

Operational consistency

• Hold exactly the same speed for all three runs.
• Allow thermal stabilisation between runs where needed.
• Keep process conditions — flow, pressure, temperature — consistent.
• Use identical sensor locations and mounting methods.

Data quality

• Take several readings per run and average them.
• Confirm phase measurements are consistent and repeatable.
• Check that the trial weights produce clearly measurable changes.
• Watch for anomalies that hint at measurement error.

Installation precision

• Carefully mark and verify trial-weight angular positions.
• Make sure trial weights are secure and will not shift during runs.
• Install final correction weights with the same care.
• Double-check masses and angles before the verification run.

8. Troubleshooting Common Issues

Poor results after correction

Possible causes:

• Correction weights installed at the wrong angles or with wrong masses.
• Operating conditions changed between trial runs and correction installation.
• Mechanical problems — looseness, misalignment — not addressed before balancing.
• Non-linear system response.

Trial weights produce a small response

Solutions:

• Use larger trial weights or place them at a greater radius.
• Check sensor mounting and signal quality.
• Verify the operating speed is correct.
• Consider whether the system has very high damping or low response sensitivity.

Inconsistent measurements

Solutions:

• Allow more time for thermal and mechanical stabilisation.
• Improve sensor mounting — studs rather than magnets.
• Isolate from external vibration sources.
• Address mechanical issues causing variable behaviour.

9. The Three-Run Method in the Field

Because it needs no balancing machine and only a handful of starts, the three-run method is the natural fit for on-site work with a portable instrument. A two-channel analyser such as the Balanset-1A reads amplitude and phase at both bearings through one run per plane, computes the influence coefficients automatically, and returns the mass and angle for each correction weight — then verifies the residual unbalance against the chosen ISO 21940-11 grade once the weights are fitted. Working in the machine’s own bearings at operating speed, it captures the true running condition the rotor will actually see, which is exactly what makes the three-run method so dependable in field balancing.

---

← Back to Main Index