In-depth Maintenance Guide for Abnormal Noises in Industrial Fanless PC Cooling Fans: From Bearing Lubrication to PWM Speed Control

In the final assembly workshop of an automobile manufacturing plant in Qingdao, a flexible production line consisting of 300 devices is operating at a pace of rolling off one car every 90 seconds. At the moment when the robotic arm completes the installation of the car door, a piercing metallic friction sound suddenly emanates from the cooling fan inside the control cabinet—a scene that epitomizes the "cooling crises" unfolding in countless industrial settings. Statistics indicate that 63% of Industrial Fanless PC failures are related to abnormalities in the cooling system, with 41% of these failures stemming from a chain reaction triggered by abnormal fan noises.

1. The Overlooked "Industrial Heartbeat": The Lethality of Cooling Fans

1.1 The Evolution from "Minor Abnormal Noises" to "Major Breakdowns"

The blast furnace monitoring system of a steel enterprise once experienced a situation where the fan bearing seized up, causing the temperature of the PLC controller to soar to 85°C. Due to the lack of a temperature warning mechanism, after running at high temperatures for 2 hours, the core chips suffered irreversible damage, resulting in direct economic losses exceeding 2 million yuan. More insidiously, the metal debris generated by bearing wear may enter the motherboard along with the airflow, causing short-circuit failures—this kind of "secondary damage" often only becomes apparent weeks after the initial failure occurs.

1.2 The Collective Anxiety of Engineers

Surveys show that 78% of industrial automation engineers have encountered equipment failures caused by abnormal fan noises, among which:

• 52% of the failures occurred during the equipment commissioning phase.
• 31% led to production line shutdowns lasting more than 4 hours.
• 19% resulted in the loss of critical data.
  At a 3C manufacturing plant in Shenzhen, Engineer Xiao Zhang's team even developed a "fan phobia"—whenever an abnormal sound came from inside the control cabinet, the entire team would enter a state of high tension.

2. Tracing the Source of Abnormal Noises: Four Core Failure Modes

2.1 Mechanical Noise: The "Death Symphony" of Bearings

When the fan speed reaches 3000 RPM, the contact pressure between the bearing balls and the raceway can reach 500 N/cm². In a case at a chemical enterprise in Qingdao, the oil-impregnated bearing's lubricating oil dried up, causing direct metal contact between the balls and the raceway, producing a piercing abnormal noise of 120 decibels. This type of failure is usually accompanied by the following characteristics:

• The frequency of the abnormal noise is proportional to the fan speed.
• A noticeable vibration can be felt when touching the fan housing.
• After disassembly, a mirror-like polishing phenomenon is observed on the bearing raceway.

2.2 Wind Noise: The "Wild Dance" of Airflow

After replacing the fan in a power monitoring system with a 92mm large-sized model, a design flaw in the air outlet caused the airflow to form eddies in a narrow channel, generating a low-frequency noise of 65 decibels. The typical manifestations of this type of failure include:

• The abnormal noise suddenly appears within a specific speed range.
• The noise frequency is related to the number of fan blades (e.g., a 3-blade fan generates noise at three times the fundamental frequency).
• Changing the installation angle of the fan can alleviate the symptoms.

2.3 Electromagnetic Noise: The "Humming" of the Motor

In a rail transit control system in Wuhan, resonance occurred between the PWM speed control signal and the power supply ripple, causing the motor coil to generate electromagnetic noise at 500 Hz. Although the volume of this noise is relatively low (about 40 decibels), it can interfere with the accuracy of analog signal acquisition, causing fluctuations in sensor data.

2.4 Resonance Noise: The "Domino Effect" of Structures

The case of a semiconductor factory in Shanghai is highly representative: resonance occurred between the fan mounting bracket and the chassis, amplifying the original 55-decibel noise to 78 decibels. The key points for identifying this type of failure include:

• The noise suddenly intensifies within a specific speed range.
• A noticeable vibration can be felt when touching the chassis housing.
• The symptoms disappear after changing the installation position of the fan.

3. Bearing Lubrication: The "Heart Surgery" for Prolonging Service Life

3.1 The Art of Grease Selection

In the practice of a new energy enterprise in Nanjing, after adopting Weineng's high-temperature grease (with an operating temperature range of -40°C to 150°C), the fan's Mean Time Between Failures (MTBF) increased from 20,000 hours to 50,000 hours. When selecting grease, the following aspects should be focused on:

• Base oil viscosity: Low-viscosity oils (ISO VG 5-10) are suitable for high-speed bearings (>5000 RPM).
• Thickener type: Lithium-based greases have excellent mechanical stability.
• Additive formulation: Greases containing molybdenum disulfide can reduce the friction coefficient by 30%.

3.2 Precise Lubrication Operation Guide

Step 1: Cleaning Preparation

• Use a lint-free cloth to wipe the area around the bearing.
• Blow away dust with compressed air (pressure ≤ 0.2 MPa).
• Avoid using solvents such as gasoline to prevent damage to the lubricating film.

Step 2: Quantitative Grease Injection

• Use a 0.1ml quantitative grease gun.
• For oil-impregnated bearings, control the injection amount to 1/3 to 1/2 of the bearing cavity.
• For double-row ball bearings, inject 0.02ml into each ball groove.

Step 3: Operation for Grease Diffusion

• Manually rotate the fan shaft core 10 times.
• Run the fan at 500 RPM for 30 minutes.
• Remove excess grease (to prevent high-temperature coking).

3.3 Maintenance Cycle Management

4. PWM Speed Control: The "Nerve Center" of Intelligent Temperature Control

4.1 Limitations of Traditional Speed Control

A voltage speed control scheme was once adopted in the control system of an injection molding machine, resulting in:

• A speed control accuracy of only ±15%.
• A 40% increase in motor heat generation.
• A limited speed control range (only 30%-100%).

4.2 Technical Advantages of PWM Speed Control

In the practice of a smart factory in Hangzhou, after adopting PWM speed control:

• The speed control accuracy was improved to ±2%.
• The motor efficiency increased by 25%.
• The speed control range was expanded to 10%-100%.
• The noise was reduced by 8 decibels (at 50% load).

EG628

Linux OSFlexibly ExpandRich Interface

4.3 Design of Intelligent Speed Control Strategies

Taking the USR-EG628 Industrial Fanless PC as an example, its built-in WukongEdge platform supports three speed control modes:

Mode 1: Temperature-Speed Linear Mapping

• When T < 40°C → Speed = 800 RPM
• When 40°C ≤ T < 55°C → Speed = 800 + (T - 40) * 40 RPM
• When 55°C ≤ T < 70°C → Speed = 1400 + (T - 55) * 60 RPM
• When T ≥ 70°C → Speed = 3000 RPM (maximum speed)

Mode 2: Predictive Speed Control

By integrating AI algorithms, it predicts the temperature trend in the next 10 minutes based on historical temperature data and adjusts the fan speed in advance. In the application of a data center in Shenzhen, this mode reduced the CPU temperature fluctuation range by 60%.

Mode 3: Multi-Fan Collaborative Control

When multiple PWM fans are arranged inside the chassis, the speed synchronization is achieved through a CAN bus. When an abnormal temperature is detected in a certain area, the system can directionally increase the air volume in that area while reducing the speed of fans in other areas to save energy.

5. USR-EG628: The "Intelligent Partner" for Cooling Maintenance

When addressing the issue of abnormal noises in cooling fans, it is crucial to choose an Industrial Fanless PC with intelligent management capabilities. The USR-EG628 Industrial Fanless PC excels in this regard:

• Intelligent Temperature Control System: Equipped with high-precision temperature sensors, it can monitor the temperature of key components in real time and dynamically adjust the fan speed through PWM speed control technology, reducing noise while ensuring cooling efficiency.
• Protocol Compatibility: It supports mainstream industrial protocols such as Profinet, Modbus TCP, and OPC UA, allowing seamless integration into existing industrial networks and facilitating remote monitoring and fault diagnosis by engineers.
• Edge Computing Capability: With an NPU featuring 1 TOPS of AI computing power, it can run localized vibration analysis algorithms to predict fan bearing wear in advance, transforming maintenance from "passive response" to "active prevention".
• Industrial-Grade Design: It has three-level surge protection and electrostatic protection, and can operate stably in a wide temperature range of -20°C to 70°C, adapting to various harsh industrial scenarios.

6. Maintenance Practice: The Entire Process from Fault to Repair

6.1 Four-Step Fault Diagnosis Method

• Listening for Sound Location: Use a stethoscope to locate the source of the abnormal noise (distinguish between bearing noise, wind noise, and electromagnetic noise).
• Data Collection: Read the fan speed, temperature, and current data through the monitoring interface of the USR-EG628.
• Spectrum Analysis: Use a Fluke 438-II power quality analyzer to capture the noise frequency spectrum.
• Root Cause Location: Combine vibration analysis (using an acceleration sensor) and thermal imaging (using a FLIR E86) to determine the fault point.

6.2 Repair Operation Specifications

Case: Bearing Replacement for a Photovoltaic Inverter Fan

• Power Disconnection and Verification: Use a multimeter to confirm that the 48V DC power supply is disconnected.
• Fan Disassembly: Use an ESD anti-static tool kit to disassemble the fan (torque controlled at 0.5 N·m).
• Bearing Replacement: Select an NSK 608ZZ double-sealed bearing (precision grade ABEC-5).
• Lubrication Treatment: Inject 0.05 ml of Klüber ISOFLEX NBU 15 grease.
• Installation and Commissioning: Run the fan at 200 RPM for 1 hour and monitor that the vibration value is ≤ 2.5 mm/s.
• Performance Verification: Continuously run the fan for 72 hours at 45°C and confirm that the temperature stabilizes at 68°C ± 2°C.

6.3 Preventive Maintenance Plan

Contact us to find out more about what you want !

Talk to our experts

Making the Cooling System a Reliable "Industrial Guardian"

In the case of a heavy industry enterprise in Jinan, by implementing the above maintenance strategies, the equipment failure rate dropped from 2.3 times per month to 0.5 times, saving 1.2 million yuan in maintenance costs annually. More importantly, engineers have got rid of the "fan phobia" and can devote more energy to process optimization.

The abnormal noises of cooling fans should not be the "background music" in industrial settings but rather the "alarm signals" that initiate intelligent maintenance. Through precise management of bearing lubrication, intelligent control of PWM speed control, and the assistance of intelligent devices such as the USR-EG628, we can fully upgrade the cooling system from "passive maintenance" to "active health management", safeguarding efficient production in the era of Industry 4.0.