Vibration Analysis with Balanset-1A: A Beginner's Guide to Spectrum Diagnostics

Introduction: From Balancing to Diagnostics — Unlocking the Full Potential of Your Vibration Analyzer

The Balanset-1A device is primarily known as an effective tool for dynamic balancing. However, its capabilities extend far beyond that, making it a powerful and accessible vibration analyzer. Equipped with sensitive sensors and software for Fast Fourier Transform (FFT) spectral analysis, the Balanset-1A is an excellent instrument for comprehensive vibration analysis. This guide bridges the gap left by the official manual, explaining what the vibration data reveals about machine health.

This guide is structured sequentially to lead you from basics to practical application:

• Section 1 will lay the theoretical foundation, simply and clearly explaining what vibration is, how spectral analysis (FFT) works, and what spectral parameters are key for a diagnostician.
• Section 2 will provide step-by-step instructions for obtaining high-quality and reliable vibration spectra using the Balanset-1A device in various modes, focusing on practical nuances not described in the standard instruction.
• Section 3 is the core of the article. Here, the "fingerprints" — characteristic spectral signs of the most common faults: unbalance, misalignment, mechanical looseness, and bearing defects — will be thoroughly analyzed.
• Section 4 will integrate the acquired knowledge into a unified system, offering practical recommendations for implementing monitoring and a simple decision-making algorithm.

By mastering the material in this article, you will be able to use Balanset-1A not only as a balancing device but also as a full-fledged entry-level diagnostic complex, allowing you to identify problems early, prevent costly accidents, and significantly increase the reliability of your operating equipment.

Section 1: Fundamentals of Vibration and Spectral Analysis (FFT)

1.1. What is Vibration and Why is it Important?

Any rotating equipment, whether a pump, fan, or electric motor, creates vibration during operation. Vibration is the mechanical oscillation of a machine or its individual parts relative to their equilibrium position. In an ideal, fully functional state, a machine generates a low and stable level of vibration — this is its normal "operating noise." However, as defects arise and develop, this vibration background begins to change.

Vibration is the response of the mechanism's structure to cyclic exciting forces. The sources of these forces can be very diverse:

• Centrifugal force due to rotor unbalance: Arises from the uneven distribution of mass relative to the axis of rotation. This is the so-called "heavy spot," which, during rotation, creates a force transmitted to the bearings and the machine casing.
• Forces associated with geometric inaccuracies: Misalignment of coupled shafts, shaft bend, errors in gear tooth profiles of the gearbox — all these create cyclic forces causing vibration.
• Aerodynamic and hydrodynamic forces: Occur during the rotation of impellers in fans, smoke extractors, pumps, and turbines.
• Electromagnetic forces: Characteristic of electric motors and generators and can be caused, for example, by winding asymmetry or the presence of shorted turns.

Each of these sources creates vibration with unique characteristics. This is why vibration analysis is such a powerful diagnostic tool. By measuring and analyzing vibration, we can not only say that "the machine vibrates strongly" but also, with a high degree of probability, determine the root cause. This advanced diagnostic capability is essential for any modern maintenance program.

1.2. From Time Signal to Spectrum: A Simple Explanation of FFT

A vibration sensor (accelerometer), installed on the bearing housing, converts mechanical oscillations into an electrical signal. If this signal is displayed on a screen as a function of time, we get a time signal or waveform. This graph shows how vibration amplitude changes at each moment in time.

For a simple case, such as pure unbalance, the time signal will look like a smooth sinusoid. However, in reality, a machine is almost always acted upon by several exciting forces simultaneously. As a result, the time signal is a complex, seemingly chaotic curve, from which it is practically impossible to extract useful diagnostic information.

This is where a mathematical tool comes to the rescue — the Fast Fourier Transform (FFT). It can be imagined as a magical prism for vibration signals.

Imagine that a complex time signal is a beam of white light. It seems unified and indistinguishable to us. But when this beam passes through a glass prism, it breaks down into its constituent colors — red, orange, yellow, and so on, forming a rainbow. FFT does the same with a vibration signal: it takes a complex curve from the time domain and decomposes it into simple sinusoidal components, each of which has its own frequency and amplitude.

The result of this transformation is displayed on a graph called a vibration spectrum. The spectrum is the main working tool for anyone performing vibration analysis. It allows you to see what is hidden in the time signal: what "pure" vibrations make up the machine's overall noise.

1.3. Key Spectrum Parameters to Understand

The vibration spectrum you will see on the Balanset-1A screen in "Vibrometer" or "Charts" modes has two axes, understanding which is absolutely necessary for diagnostics.

Horizontal Axis (X): Frequency

This axis shows how often oscillations occur and is measured in Hertz (Hz). 1 Hz is one complete oscillation per second. Frequency is directly related to the source of vibration. Various mechanical and electrical components of a machine generate vibration at their characteristic, predictable frequencies. Knowing the frequency at which a high vibration peak is observed, we can identify the culprit — a specific unit or defect.

Rotational frequency (1x): This is the most important frequency in all vibration diagnostics. It corresponds to the rotational speed of the machine's shaft. For example, if a motor shaft rotates at 3000 revolutions per minute (rpm), its rotational frequency will be: f = 3000 rpm / 60 s/min = 50 Hz. This frequency is denoted as 1x. It serves as a reference point for identifying many other defects.

Vertical Axis (Y): Amplitude

This axis shows the intensity or strength of vibration at each specific frequency. In the Balanset-1A device, amplitude is measured in millimeters per second (mm/s), which corresponds to the root mean square (RMS) value of vibration velocity. The higher the peak in the spectrum, the more vibration energy is concentrated at that frequency, and, as a rule, the more serious the associated defect.

Harmonics

Harmonics are frequencies that are integer multiples of the fundamental frequency. Most often, the fundamental frequency is the rotational frequency 1x. Thus, its harmonics will be: 2x (second harmonic) = 2×1x, 3x (third harmonic) = 3×1x, 4x (fourth harmonic) = 4×1x, and so on. The presence and relative height of harmonics carry crucial diagnostic information. For example, pure unbalance manifests mainly at 1x with very low harmonics. However, mechanical looseness or shaft misalignment generate a whole "forest" of high harmonics (2x, 3x, 4x,...). By analyzing the ratio of amplitudes between 1x and its harmonics, different types of faults can be distinguished.

Section 2: Obtaining a Vibration Spectrum Using Balanset-1A

The quality of diagnostics directly depends on the quality of the initial data. Incorrect measurements can lead to erroneous conclusions, unnecessary repairs, or, conversely, to missing a developing defect. This section provides a practical guide to collecting accurate and repeatable data using your device.

2.1. Preparation for Measurements: The Key to Accurate Data

Before connecting cables and launching the program, careful attention must be paid to the correct installation of sensors. This is the most important stage, determining the reliability of all subsequent analysis.

Mounting Method: Balanset-1A comes with magnetic sensor bases. This is a convenient and fast mounting method, but for its effectiveness, several rules must be observed. The surface at the measurement point must be:

• Clean: Remove dirt, rust, and peeling paint.
• Flat: The sensor must lie flush with the entire surface of the magnet. Do not install it on rounded surfaces or bolt heads.
• Massive: The measurement point should be part of the machine's load-bearing structure (e.g., bearing housing), not a thin protective cover or cooling fin.

For stationary monitoring or to achieve maximum accuracy at high frequencies, it is recommended to use a threaded connection (stud) if the machine design allows.

Location: Forces arising during rotor operation are transmitted to the machine casing through the bearings. Therefore, the best place to install sensors is the bearing housings. Try to place the sensor as close as possible to the bearing to measure vibration with minimal distortion.

Measurement Direction: Vibration is a three-dimensional process. For a complete picture of the machine's condition, measurements should be taken in three directions:

• Radial horizontal (H): Perpendicular to the shaft axis, in the horizontal plane.
• Radial vertical (V): Perpendicular to the shaft axis, in the vertical plane.
• Axial (A): Parallel to the shaft axis.

As a rule, the stiffness of the structure in the horizontal direction is lower than in the vertical, so the vibration amplitude in the horizontal direction is often the highest. This is why the horizontal direction is often chosen for initial assessment. However, axial vibration carries unique information, critically important for diagnosing defects such as shaft misalignment.

Balanset-1A is a two-channel device, which is primarily considered in the manual from the perspective of two-plane balancing. However, for diagnostics, this opens up much broader possibilities. Instead of measuring vibration on two different bearings, both sensors can be connected to the same bearing unit, but in different directions. For example, sensor channel 1 can be installed radially (horizontally), and sensor channel 2 axially. Simultaneous acquisition of spectra in two directions allows for instant comparison of axial and radial vibration, which is a standard technique in professional diagnostics for reliable misalignment detection. This method significantly expands the diagnostic capabilities of the device, going beyond what is described in the manual.

2.2. Step-by-Step: Using "Vibrometer" Mode (F5) for Quick Assessment

This mode is designed for operational control of main vibration parameters and is ideal for quick "on-site" machine condition assessment. The procedure for obtaining a spectrum in this mode is as follows:

1. Connect sensors: Install vibration sensors at selected points and connect them to the X1 and X2 inputs of the measuring unit. Connect the laser tachometer to the X3 input and attach a reflective marker to the shaft.
2. Start the program: In the main Balanset-1A program window, click the "F5 - Vibration Meter" button.
3. The working window will open (Fig. 7.4 in the manual). Its upper part will display digital values: overall vibration (V1s), vibration at rotational frequency (V1o), phase (F1), and rotational speed (N rev).
4. Start measurement: Click the "F9 - Run" button. The program will start collecting and displaying data in real time.
5. Analyze the spectrum: At the bottom of the window is the "Vibration spectrum-channel 1&2 (mm/s)" graph. This is the vibration spectrum. The horizontal axis shows frequency in Hz, and the vertical axis shows amplitude in mm/s.

This mode allows for the first, most important diagnostic check, recommended even in the balancing manual. Compare the values of V1s (overall vibration) and V1o (vibration at rotational frequency 1x).

• If V1s≈V1o, it means that most of the vibrational energy is concentrated at the rotational frequency. The main cause of vibration is most likely unbalance.
• If V1s≫V1o, it indicates that a significant portion of the vibration is caused by other sources (misalignment, looseness, bearing defects, etc.). In this case, simple balancing will not solve the problem, and a deeper analysis of the spectrum is necessary.

2.3. Step-by-Step: Using "Charts" Mode (F8) for Detailed Analysis

For serious diagnostics requiring a more detailed examination of the spectrum, "Charts" mode is significantly better. It provides a larger and more informative graph, which facilitates the identification of peaks and the analysis of their structure. The procedure for obtaining a spectrum in this mode:

1. Connect sensors in the same way as for "Vibrometer" mode.
2. Start mode: In the main program window, click the "F8 - Charts" button.
3. Select chart type: In the opened window (Fig. 7.19 in the manual), there will be a row of buttons at the top. Click "F5-Spectrum (Hz)".
4. The spectrum analysis window will open (Fig. 7.23 in the manual). The upper part will display the time signal, and the lower, main part will display the vibration spectrum.
5. Start measurement: Click the "F9-Run" button. The device will perform a measurement and build detailed graphs.

The spectrum obtained in this mode is much more convenient for analysis. You can more clearly see peaks at different frequencies, evaluate their height, and identify harmonic series. This mode is recommended for diagnosing faults described in the next section.

Section 3: Diagnostics of Typical Faults by Vibration Spectra (up to 1000 Hz)

This section is the practical core of the guide. Here we will learn to read spectra and correlate them with specific mechanical problems. For convenience and quick orientation in the field, the main diagnostic indicators are summarized in a consolidated table. It will serve as a quick reference when analyzing real data.

Table 3.1: Summary of Diagnostic Indicators

Fault	Primary Spectral Signature	Typical Harmonics	Notes
Unbalance	High amplitude at 1× rotational frequency	Low	Radial vibration dominates. Amplitude increases quadratically with speed.
Misalignment	High amplitude at 2× rotational frequency	1×, 3×, 4×	Often accompanied by axial vibration.
Mechanical looseness	Multiple harmonics 1× ("forest" of harmonics)	1×, 2×, 3×, 4×, 5×...	Subharmonics (0.5×, 1.5×) may appear at 1/2x, 3/2x, etc. due to cracks.
Bearing defect	Peaks at non-synchronous frequencies (BPFO, BPFI, etc.)	Multiple harmonics of defect frequencies	Often visible as sidebands around peaks. Sounds like "noise" in the high-frequency range.
Gear mesh defect	High frequency of gear mesh (GMF) and its harmonics	Sidebands around GMF at 1x	Indicates wear, tooth damage, or eccentricity.

Next, we will break down each of these defects in detail.

3.1. Unbalance: The Most Common Problem

Physical Cause: Unbalance occurs when the center of mass of a rotating part (rotor) does not coincide with its geometric axis of rotation. This creates a "heavy spot" which, during rotation, generates a centrifugal force acting in the radial direction and transmitted to the bearings and foundation.

Spectral Signatures: The main sign is a high amplitude peak strictly at the rotational frequency (1x). Vibration is predominantly radial. There are two main types of unbalance:

Static Unbalance (One-plane)

Spectrum Description: The spectrum is entirely dominated by a single peak at the fundamental rotational frequency (1x). The vibration is sinusoidal, with minimal energy at other frequencies.

Brief Description of Spectral Components: Primarily a strong 1x rotational frequency component. Little to no higher harmonics (a pure 1x tone).

Key Feature: Large 1x amplitude in all radial directions. Vibration at both bearings is in phase (no phase difference between the two ends). Approximately 90° phase shift is often observed between horizontal and vertical measurements at the same bearing.

Dynamic Unbalance (Two-plane / Couple)

Spectrum Description: The spectrum also shows a dominant once-per-revolution frequency (1x) peak, similar to static unbalance. Vibration is at the rotation speed, with no significant higher-frequency content if unbalance is the only issue.

Brief Description of Spectral Components: Dominant 1x RPM component (often with a "sway" or wobble of the rotor). Higher harmonics are generally absent unless other faults are present.

Key Feature: 1x vibration at each bearing is out of phase — there is about a 180° phase difference between vibration at the two ends of the rotor (indicating a couple unbalance). The strong 1x peak with this phase relationship is the signature of dynamic unbalance.

What to do: If the spectrum indicates unbalance, a balancing procedure must be performed. For static unbalance, single-plane balancing is sufficient (manual section 7.4), for dynamic unbalance — two-plane balancing (manual section 7.5).

3.2. Shaft Misalignment: A Hidden Threat

Physical Cause: Misalignment occurs when the axes of rotation of two coupled shafts (e.g., motor shaft and pump shaft) do not coincide. When misaligned shafts rotate, cyclic forces arise in the coupling and bearings, causing vibration.

Parallel Misalignment (Offset Shafts)

Spectrum Description: The vibration spectrum exhibits elevated energy at the fundamental (1x) and its harmonics 2x and 3x, especially in the radial direction. Typically, the 1x component is dominant with misalignment present, accompanied by a notable 2x component.

Brief Description of Spectral Components: Contains significant peaks at 1x, 2x, and 3x shaft rotational frequencies. These appear predominantly in radial vibration measurements (perpendicular to the shaft).

Key Feature: High 1x and 2x vibration in the radial direction are indicative. A 180° phase difference between radial vibration measurements on opposite sides of the coupling is often observed, distinguishing it from pure unbalance.

Angular Misalignment (Inclined Shafts)

Spectrum Description: The frequency spectrum shows strong harmonics of the shaft speed, notably a prominent 2x running speed component in addition to the 1x. Vibration at 1x, 2x (and often 3x) appears, with axial (along-shaft) vibration being significant.

Brief Description of Spectral Components: Notable peaks at 1x and 2x (and sometimes 3x) of running speed. The 2x component is often as large as or larger than the 1x. These frequencies are pronounced in the axial vibration spectrum (along the machine's axis).

Key Feature: Relatively high second harmonic (2x) amplitude compared to 1x, combined with strong axial vibration. Axial measurements on either side of the coupling are 180° out of phase, a hallmark of angular misalignment.

What to do: Balancing will not help here. Stop the unit and perform a shaft alignment procedure using specialized tools.

3.3. Mechanical Looseness: "Rattling" in the Machine

Physical Cause: This defect is associated with a loss of stiffness in structural connections: loose bolts, cracks in the foundation, increased clearances in bearing seats. Due to clearances, impacts occur, forming a characteristic vibration pattern.

Mechanical Looseness (Component Looseness)

Description: The spectrum is rich in frequency components of the rotational speed. A wide range of integer multiples of 1x (from 1x to high orders, such as ~10x) with significant amplitudes appears. In some cases, subharmonic frequencies (e.g., 0.5x) may also appear.

Spectral Components: Dominant are multiple frequency components of the rotational speed (1x, 2x, 3x ... up to ~10x). Sometimes fractional (half-integer) frequency components may also be present at 1/2x, 3/2x, etc. due to repeated impacts.

Key Feature: The distinctive "series of peaks" in the spectrum — numerous evenly spaced peaks at frequencies that are integer multiples of the rotational speed. This indicates a loss of stiffness or improper assembly of parts causing repeated impacts. The presence of many harmonics (and possibly half-integer subharmonics) is a key indicator.

Structural Looseness (Base/Mounting Looseness)

Description: In the vibration spectrum, vibration at the fundamental or double rotational frequency often dominates. Usually, a peak appears at 1x and/or 2x. Higher harmonics (above 2x) usually have much smaller amplitudes compared to these main ones.

Spectral Components: Predominantly shows frequency components at 1x and 2x speeds of the shaft. Other harmonics (3x, 4x, etc.) are usually absent or insignificant. The component 1x or 2x may dominate depending on the type of looseness (e.g., one impact per revolution or two impacts per revolution).

Key Feature: Noticeably high peaks at 1x or 2x (or both) relative to the rest of the spectrum, indicating looseness of bearings or structure. The vibration is stronger in the vertical direction if the machine is loosely mounted. One or two low-order dominant peaks with a small number of high-order harmonics are characteristic of structural or foundation looseness.

What to do: A thorough inspection of the unit is necessary. Check all accessible fastening bolts (bearings, housing). Inspect the frame and foundation for cracks. If there is internal looseness (e.g., bearing seat), disassembly of the unit may be required.

3.4. Rolling Bearing Defects: Early Warning

Physical Cause: The occurrence of defects (pits, spalls, wear) on the rolling surfaces (inner ring, outer ring, rolling elements) or on the cage. Each time a rolling element rolls over a defect, a short impact impulse occurs. These impulses repeat at a specific frequency characteristic of each bearing element.

Spectral Signatures: Bearing defects appear as peaks at non-synchronous frequencies, i.e., at frequencies that are not integer multiples of the rotational frequency (1x). These frequencies (BPFO - outer race defect frequency, BPFI - inner race, BSF - rolling element, FTF - cage) depend on the bearing geometry and rotational speed. For a beginner diagnostician, it is not necessary to calculate their exact values. The main thing is to learn to recognize their presence in the spectrum.

Outer Race Defect

Spectrum Description: The vibration spectrum exhibits a series of peaks corresponding to the outer race defect frequency and its harmonics. These peaks are usually at higher frequencies (not integer multiples of shaft rotation) and indicate each time a rolling element passes over the outer race flaw.

Brief Description of Spectral Components: Multiple harmonics of the outer race ball-pass frequency (BPFO) are present. Typically, 8–10 harmonics of BPFO can be observed in the spectrum for a pronounced outer race fault. The spacing between these peaks is equal to the BPFO (a characteristic frequency determined by bearing geometry and speed).

Key Feature: A distinct train of peaks at the BPFO and its successive harmonics is the signature. The presence of numerous evenly spaced high-frequency peaks (BPFO, 2xBPFO, 3xBPFO, ...) clearly points to an outer race bearing defect.

Inner Race Defect

Spectrum Description: The spectrum for an inner race fault shows several prominent peaks at the inner race defect frequency and its harmonics. In addition, each of these fault frequency peaks is typically accompanied by sideband peaks spaced at the running speed (1x) frequency.

Brief Description of Spectral Components: Contains multiple harmonics of the inner race ball-pass frequency (BPFI), often on the order of 8–10 harmonics. Characteristically, these BPFI peaks are modulated by sidebands at ±1x RPM — meaning beside each BPFI harmonic, smaller side peaks appear, separated from the main peak by an amount equal to the shaft rotation frequency.

Key Feature: The telltale sign is the presence of the inner race defect frequency (BPFI) harmonics with a sideband pattern. The sidebands spaced at the shaft speed around the BPFI harmonics indicate that the inner race defect is being loaded once per revolution, confirming an inner race issue rather than outer race.

Rolling Element Defect (Ball/Roller)

Spectrum Description: A rolling element (ball or roller) defect produces vibration at the rolling element spin frequency and its harmonics. The spectrum will show a series of peaks that are not integer multiples of shaft speed, but rather multiples of the ball/roller spin frequency (BSF). One of these harmonic peaks is often significantly larger than the others, reflecting how many rolling elements are damaged.

Brief Description of Spectral Components: Peaks at the fundamental rolling element defect frequency (BSF) and its harmonics. For example, BSF, 2xBSF, 3xBSF, etc., will appear. Notably, the amplitude pattern of these peaks can indicate the number of damaged elements — e.g., if the second harmonic is largest, it might suggest two balls/rollers have spalls. Often, some vibration at the race fault frequencies accompanies this, as rolling element damage commonly leads to race damage as well.

Key Feature: The presence of a series of peaks spaced by the BSF (bearing element spin frequency) rather than by the shaft rotation frequency identifies a rolling element defect. A particularly high amplitude of the Nth harmonic of BSF often implies N elements are damaged (e.g., a very high 2xBSF peak might indicate two balls with defects).

Cage Defect (Bearing Cage / FTF)

Spectrum Description: A cage (separator) defect in a rolling bearing yields vibration at the cage rotational frequency – the Fundamental Train Frequency (FTF) – and its harmonics. These frequencies are usually sub-synchronous (below the shaft speed). The spectrum will show peaks at FTF, 2xFTF, 3xFTF, etc., and often some interaction with other bearing frequencies due to modulation.

Brief Description of Spectral Components: Low-frequency peaks corresponding to the cage's rotational frequency (FTF) and integer multiples of it. For instance, if FTF ≈ 0.4x shaft speed, you may see peaks at ~0.4x, ~0.8x, ~1.2x etc. In many cases, a cage defect coexists with race defects, so the FTF may modulate race defect signals, producing sum/difference frequencies (sidebands around race frequencies).

Key Feature: One or more sub-harmonic peaks (below 1x) that align with the bearing cage rotation rate (FTF) are indicative of a cage problem. This often appears alongside other bearing fault indications. The key signature is the presence of FTF and its harmonics in the spectrum, which is otherwise uncommon unless the cage is failing.

What to do: The appearance of bearing frequencies is a call to action. It is necessary to intensify monitoring of this unit, check the lubrication condition, and start planning bearing replacement at the earliest opportunity.

3.5. Gear Faults

Gear Eccentricity / Bent Shaft

Spectrum Description: This fault causes modulation of the gear mesh vibration. In the spectrum, the gear mesh frequency (GMF) peak is surrounded by sideband peaks spaced at the gear's shaft rotational frequency (1x gear RPM). Often, the gear's own 1x running speed vibration is also elevated due to the unbalance-like effect of eccentricity.

Brief Description of Spectral Components: Notable increase in amplitude at the gear mesh frequency and its lower harmonics (e.g., 1x, 2x, 3x GMF). Clear sidebands appear around the GMF (and sometimes around its harmonics) at intervals equal to 1x the rotation rate of the affected gear. The presence of these sidebands indicates amplitude modulation of the mesh frequency by the gear's rotation.

Key Feature: Gear mesh frequency with pronounced sidebands at 1x gear frequency is the signature feature. This sideband pattern (peaks equally spaced around GMF by the running speed) strongly indicates gear eccentricity or a bent gear shaft. Additionally, the gear's fundamental (1x) vibration may be higher than normal.

Gear Tooth Wear or Damage

Spectrum Description: Gear tooth faults (such as worn or broken teeth) produce an increase in vibration at the gear mesh frequency and its harmonics. The spectrum often shows multiple GMF peaks (1xGMF, 2xGMF, etc.) of high amplitude. Additionally, numerous sideband frequencies appear around these GMF peaks, spaced by the shaft rotational frequency. In some cases, the excitation of gear natural frequencies (resonances) with sidebands can also be observed.

Brief Description of Spectral Components: Elevated peaks at the gear mesh frequency (tooth-meshing frequency) and its harmonics (for example, 2xGMF). Around each major GMF harmonic, there are sideband peaks separated by 1x running speed. The number and size of sidebands around the 1x, 2x, 3x GMF components tend to increase with the severity of tooth damage. In severe cases, additional peaks corresponding to the gear's resonance frequencies (with their own sidebands) may appear.

Key Feature: Multiple high-amplitude gear mesh frequency harmonics accompanied by dense sideband patterns are the hallmark. This indicates irregular tooth passing due to wear or a broken tooth. A heavily worn or damaged gear will show extensive sidebands (at 1x gear speed intervals) around the mesh frequency peaks, distinguishing it from a healthy gear (which would have a cleaner spectrum concentrated at GMF).

What to do: The appearance of frequencies related to gear trains requires closer attention. It is recommended to check the oil condition in the gearbox for metallic particles and to schedule an inspection of the gearbox to assess tooth wear or damage.

It is important to understand that in real-world conditions, machines rarely suffer from only one fault. Very often, the spectrum is a combination of signs of several defects, such as unbalance and misalignment. This can be confusing for a beginner diagnostician. In such cases, a simple rule applies: address the problem corresponding to the peak with the largest amplitude first. Often, one serious fault (e.g., severe misalignment) causes secondary problems, such as increased bearing wear or loosening of fasteners. By eliminating the root cause, you can significantly reduce the manifestation of secondary defects.

Section 4: Practical Recommendations and Next Steps

Having mastered the basics of spectrum interpretation, you have taken the first and most important step. Now it is necessary to integrate this knowledge into your daily maintenance practice. This section is dedicated to how to move from one-time measurements to a systematic approach and how to use the obtained data to make informed decisions.

4.1. From Single Measurement to Monitoring: The Power of Trends

A single spectrum is just a "snapshot" of the machine's condition at a given moment in time. It can be very informative, but its true value is revealed when compared with previous measurements. This process is called condition monitoring or trend analysis.

The idea is very simple: instead of judging the machine's condition by absolute vibration values ("good" or "bad"), you track how these values change over time. A slow, gradual increase in amplitude at a certain frequency indicates systematic wear, while a sudden jump is an alarm signal indicating the rapid development of a defect.

Practical Tip:

• Create a Baseline Spectrum: Conduct a thorough measurement on new, newly repaired, or known-good equipment. Save this data (spectra and numerical values) in the Balanset-1A program archive. This is your "health benchmark" for this machine.
• Establish Periodicity: Determine how often you will perform control measurements. For critically important equipment, this may be once every two weeks; for auxiliary equipment, once a month or quarter.
• Ensure Repeatability: Each time, perform measurements at the same points, in the same directions, and, if possible, under the same operating conditions of the machine (load, temperature).
• Compare and Analyze: After each new measurement, compare the obtained spectrum with the baseline and previous ones. Pay attention not only to the appearance of new peaks but also to the increase in amplitude of existing ones. A sharp increase in the amplitude of any peak (e.g., twice compared to the previous measurement) is a reliable signal of a developing defect, even if the absolute vibration value is still within acceptable limits according to ISO standards.

4.2. When to Balance and When to Look for Another Cause?

The ultimate goal of diagnostics is not just to find a defect, but to make the right decision about the necessary actions. Based on spectrum analysis, a simple and effective decision-making algorithm can be built.

Action Algorithm based on Spectrum Analysis:

1. Obtain a high-quality spectrum using Balanset-1A, preferably in "Charts" mode (F8), by taking measurements in both radial and axial directions.
2. Identify the peak with the largest amplitude. It indicates the dominant problem that should be addressed first.
3. Determine the type of fault by the frequency of this peak:
   • If the 1x peak dominates: The most probable cause is unbalance.
     Action: Perform a dynamic balancing procedure using the Balanset-1A device's functionality.
   • If the 2x peak dominates (especially if it is high in the axial direction): The most probable cause is shaft misalignment.
     Action: Balancing is ineffective. It is necessary to stop the unit and perform shaft alignment.
   • If a "forest" of many harmonics (1x, 2x, 3x,...) is observed: The most probable cause is mechanical looseness.
     Action: Conduct a visual inspection. Check and tighten all mounting bolts. Inspect the frame and foundation for cracks.
   • If non-synchronous peaks dominate in the mid- or high-frequency range: The most probable cause is a rolling bearing defect.
     Action: Check the lubrication in the bearing unit. Start planning bearing replacement. Increase the frequency of monitoring this unit to track the rate of defect development.
   • If the gear mesh frequency (GMF) with sidebands dominates: The most probable cause is a gear defect.
     Action: Check the oil condition in the gearbox. Schedule a gearbox inspection to assess tooth wear or damage.

This simple algorithm allows transitioning from abstract analysis to concrete, targeted maintenance actions, which is the ultimate goal of all diagnostic work.

Conclusion

The Balanset-1A device, originally designed as a specialized tool for balancing, has significantly greater potential. The ability to obtain and display vibration spectra transforms it into a powerful entry-level vibration analyzer. This article was intended to be a bridge between the operational capabilities of the device described in the manual and the fundamental knowledge necessary for interpreting the obtained data from your vibration analysis sessions.

Mastering basic spectrum analysis skills is not just about studying theory, but acquiring a practical tool to increase the efficiency of your work. Understanding how various faults — unbalance, misalignment, looseness, and bearing defects — manifest as unique "fingerprints" on the vibration spectrum allows you to look inside a running machine without disassembling it.

Key takeaways from this guide:

• Vibration is information. Each peak in the spectrum carries information about a specific process occurring in the mechanism.
• FFT is your translator. Fast Fourier Transform translates the complex and chaotic language of vibration into the simple and understandable language of frequencies and amplitudes.
• Diagnostics is pattern recognition. By learning to identify characteristic spectral patterns for major defects, you can quickly and accurately determine the root cause of increased vibration.
• Trends are more important than absolute values. Regular monitoring and comparison of current data with baseline data are the basis of a predictive approach, allowing problems to be identified at the earliest stage.

The path to becoming a confident and competent vibration analyst requires time and practice. Do not be afraid to experiment, collect data from various equipment, and create your own library of "health spectra" and "disease spectra." This guide has provided you with a map and compass. Use Balanset-1A not only to "treat" symptoms by balancing but also to make an accurate "diagnosis." This approach will allow you to significantly increase the reliability of your equipment, reduce the number of emergency shutdowns, and move to a qualitatively new level of maintenance.