4. Measurement Systems for Balancing Machines
Most amateur manufacturers of balancing machines, who contact LLC “Kinematics”, plan to use the “Balanset” series measurement systems manufactured by our company in their designs. However, there are also some customers who plan to manufacture such measuring systems independently. Therefore, it makes sense to discuss the construction of a measuring system for a balancing machine in more detail. The main requirement for these systems is the need to provide high-precision measurements of the amplitude and phase of the rotational component of the vibrational signal, which appears at the rotation frequency of the balanced rotor. This goal is usually achieved by using a combination of technical solutions, including:

 Use of vibration sensors with a high signal conversion coefficient;
 Use of modern laser phase angle sensors;
 Creation (or use) of hardware that allows for the amplification and digital conversion of sensor signals (primary signal processing);
 Implementation of software processing of the vibrational signal, which should allow for the high-resolution and stable extraction of the rotational component of the vibrational signal, manifesting at the rotation frequency of the balanced rotor (secondary processing). Below, we consider known variants of such technical solutions, implemented in a number of well-known balancing instruments.

4.1. Selection of Vibration Sensors
In the measurement systems of balancing machines, various types of vibration sensors (transducers) can be used, including:

 Vibration acceleration sensors (accelerometers);
 Vibration velocity sensors;
 Vibration displacement sensors;
 Force sensors.

4.1.1. Vibration Acceleration Sensors
Among vibration acceleration sensors, piezo and capacitive (chip) accelerometers are the most widely used, which can be effectively used in Soft Bearing type balancing machines. In practice, it is generally permissible to use vibration acceleration sensors with conversion coefficients (Kpr) ranging from 10 to 30 mV/(m/s²). In balancing machines that require particularly high balancing accuracy, it is advisable to use accelerometers with Kpr reaching levels of 100 mV/(m/s²) and above. As an example of piezo accelerometers that can be used as vibration sensors for balancing machines, Figure 4.1 shows the DN3M1 and DN3M1V6 piezo accelerometers manufactured by LLC “Izmeritel”.

.

                                

                     Figure 4.1. Piezo Accelerometers DN 3M1 and DN 3M1V6

To connect such sensors to vibration measuring instruments and systems, it is necessary to use external or built-in charge amplifiers.

Figure 4.2. Capacitive Accelerometers AD1 Manufactured by LLC “Kinematics”
It should be noted that these sensors, which include widely used market boards of capacitive accelerometers ADXL 345 (see Figure 4.3), have several significant advantages over piezo accelerometers. Specifically, they are 4 to 8 times cheaper with similar technical characteristics. Moreover, they do not require the use of costly and finicky charge amplifiers needed for piezo accelerometers.

In cases where both types of accelerometers are used in the measurement systems of balancing machines, hardware integration (or double integration) of the sensor signals is usually performed.

.

                             

.

                       
Figure 4.2. Capacitive Accelerometers AD 1, assembled.

.

.

                                             

                    Figure 4.3. Capacitive accelerometer board ADXL 345.

In this case, the initial sensor signal, proportional to vibrational acceleration, is accordingly transformed into a signal proportional to vibrational velocity or displacement. The procedure of double integration of the vibration signal is particularly relevant when using accelerometers as part of the measuring systems for low-speed balancing machines, where the lower rotor rotation frequency range during balancing can reach 120 rpm and below. When using capacitive accelerometers in the measuring systems of balancing machines, it should be considered that after integration, their signals may contain low-frequency interference, manifesting in the frequency range from 0.5 to 3 Hz. This may limit the lower frequency range of balancing on machines intended to use these sensors.

4.1.2. Vibration Velocity Sensors 4.1.2.1. Inductive Vibration Velocity Sensors. These sensors include an inductive coil and a magnetic core. When the coil vibrates relative to a stationary core (or the core relative to a stationary coil), an EMF is induced in the coil, the voltage of which is directly proportional to the vibration velocity of the movable element of the sensor. The conversion coefficients (Кпр) of inductive sensors are usually quite high, reaching several tens or even hundreds of mV/mm/sec. In particular, the conversion coefficient of the Schenck model T77 sensor is 80 mV/mm/sec, and for the IRD Mechanalysis model 544M sensor, it is 40 mV/mm/sec. In some cases (for example, in Schenck balancing machines), special highly sensitive inductive vibration velocity sensors with a mechanical amplifier are used, where Кпр can exceed 1000 mV/mm/sec. If inductive vibration velocity sensors are used in the measuring systems of balancing machines, hardware integration of the electrical signal proportional to vibration velocity can also be performed, converting it into a signal proportional to vibration displacement.

.

.

                                           

                        Figure 4.4. Model 544M sensor by IRD Mechanalysis.

.

                                               

.

                            Figure 4.5. Model T77 sensor by Schenck It should be noted that due to the labor intensity of their production, inductive vibration velocity sensors are quite scarce and expensive items. Therefore, despite the obvious advantages of these sensors, amateur manufacturers of balancing machines use them very rarely.

4.1.2.2. Vibration Velocity Sensors Based on Piezoelectric Accelerometers. A sensor of this type differs from a standard piezoelectric accelerometer by having a built-in charge amplifier and integrator within its housing, which allows it to output a signal proportional to vibration velocity. For example, piezoelectric vibration velocity sensors manufactured by domestic producers (ZETLAB company and LLC “Vibropribor”) are shown in Figures 4.6 and 4.7.

.

  

                                            

             Figure 4.6. Model AV02 sensor by ZETLAB (Russia)

.

.

                   

.

                Figure 4.7. Model DVST 2 sensor by LLC “Vibropribor” Such sensors are manufactured by various producers (both domestic and foreign) and are currently widely used, especially in portable vibration equipment. The cost of these sensors is quite high and can reach 20,000 to 30,000 rubles each, even from domestic manufacturers.

4.1.3. Displacement Sensors In the measurement systems of balancing machines, non-contact displacement sensors – capacitive or inductive – can also be used. These sensors can operate in static mode, allowing the registration of vibrational processes starting from 0 Hz. Their use can be particularly effective in the case of balancing low-speed rotors with rotation speeds of 120 rpm and below. The conversion coefficients of these sensors can reach 1000 mV/mm and higher, which provides high accuracy and resolution in measuring displacement, even without additional amplification. An obvious advantage of these sensors is their relatively low cost, which for some domestic manufacturers does not exceed 1000 rubles. When using these sensors in balancing machines, it is important to consider that the nominal working gap between the sensor’s sensitive element and the surface of the vibrating object is limited by the diameter of the sensor coil. For example, for the sensor shown in Figure 4.8, model ISAN E41A by “TEKO,” the specified working gap is typically 3.8 to 4 mm, which allows for the measurement of displacement of the vibrating object in the range of ±2.5 mm.

                                                       

                                                  

                  Figure 4.8. Inductive Displacement Sensor Model ISAN E41A by TEKO (Russia)

4.1.4. Force Sensors As previously noted, force sensors are used in the measurement systems installed on Hard Bearing balancing machines. These sensors, particularly due to their simplicity of manufacture and relatively low cost, are commonly piezoelectric force sensors. Examples of such sensors are shown in Figures 4.9 and 4.10.

.

            

       

      

                 Figure 4.9. Force Sensor SD 1 by Kinematika LLC

.

.

                                        

.

            Figure 4.10: Force Sensor for Automotive Balancing Machines, Sold by “STO Market” Strain gauge force sensors, which are manufactured by a wide range of domestic and foreign producers, can also be used to measure relative deformations in the supports of Hard Bearing balancing machines.

4.2. Phase Angle Sensors For synchronizing the vibration measurement process with the rotation angle of the balanced rotor, phase angle sensors, such as laser (photoelectric) or inductive sensors, are used. These sensors are manufactured in various designs by both domestic and international producers. The price range for these sensors can vary significantly, from approximately 40 to 200 dollars. An example of such a device is the phase angle sensor manufactured by “Diamex,” shown in figure 4.11.

.

                                

                   Figure 4.11: Phase Angle Sensor by “Diamex”

As another example, Figure 4.12 shows a model implemented by LLC “Kinematics”, which uses laser tachometers of the DT 2234C model made in China as phase angle sensors. The obvious advantages of this sensor include:

 A wide operating range, allowing measurement of rotor rotation frequency from 2.5 to 99,999 revolutions per minute, with a resolution of no less than one revolution;
 Digital display;
 Ease of setting up the tachometer for measurements;
 Affordability and low market cost;
 Relative simplicity of modification for integration into the measuring system of a balancing machine.

.

.

                                     https://images.ua.prom.st/114027425_w640_h2048_4702725083.jpg?PIMAGE_ID=114027425

                      Figure 4.12: Laser Tachometer Model DT 2234C

In some cases, when the use of optical laser sensors is undesirable for any reason, they can be replaced with inductive non-contact displacement sensors, such as the previously mentioned ISAN E41A model or similar products from other manufacturers.

.

          

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

      4.3. Signal Processing Features in Vibration Sensors For precise measurement of amplitude and phase of the rotational component of the vibration signal in balancing equipment, a combination of hardware and software processing tools is typically used. These tools enable:

Broadband hardware filtering of the sensor’s analog signal;
Amplification of the sensor’s analog signal;
Integration and/or double integration (if necessary) of the analog signal;
Narrowband filtering of the analog signal using a tracking filter;
Analog-to-digital conversion of the signal;
Synchronous filtering of the digital signal;
Harmonic analysis of the digital signal.

4.3.1. Broadband Signal Filtering This procedure is essential for cleansing the vibration sensor signal of potential interferences that may occur at both the lower and upper bounds of the device’s frequency range. It is advisable for the measuring device of a balancing machine to set the lower limit of the band-pass filter to 2-3 Hz and the upper limit to 50 (100) Hz. “Lower” filtering helps suppress low-frequency noises which may appear at the output of various types of sensor measuring amplifiers. “Upper” filtering eliminates the possibility of interference due to combination frequencies and potential resonant vibrations of individual mechanical components of the machine.

4.3.2. Amplification of the Analog Signal from the Sensor If there is a need to increase the sensitivity of the balancing machine’s measuring system, the signals from the vibration sensors to the input of the measuring unit can be amplified. Both standard amplifiers with a constant gain and multistage amplifiers, whose gain can be programmatically changed depending on the real signal level from the sensor, can be used. An example of a programmable multistage amplifier includes amplifiers implemented in voltage measurement converters like E154 or E14-140 by LLC “L-Card”.

4.3.3. Integration As noted earlier, hardware integration and/or double integration of vibration sensor signals are recommended in the measuring systems of balancing machines. Thus, the initial accelerometer signal, proportional to vibro-acceleration, can be transformed into a signal proportional to vibro-speed (integration) or vibro-displacement (double integration). Similarly, the vibro-speed sensor signal after integration can be transformed into a signal proportional to vibro-displacement.

4.3.4. Narrowband Filtering of the Analog Signal Using a Tracking Filter To reduce interference and improve the quality of vibration signal processing in the measuring systems of balancing machines, narrowband tracking filters can be used. The central frequency of these filters is automatically tuned to the rotation frequency of the balanced rotor using the rotor’s revolution sensor signal. Modern integrated circuits, such as MAX263, MAX264, MAX267, MAX268 by “MAXIM”, can be used to create such filters.

4.3.5. Analog-to-Digital Conversion of Signals Analog-to-digital conversion is a crucial procedure that ensures the possibility of improving the quality of vibration signal processing during the measurement of amplitude and phase. This procedure is implemented in all modern measuring systems of balancing machines. An example of effective implementation of such ADCs includes the voltage measurement converters type E154 or E14-140 by LLC “L-Card”, used in several measuring systems of balancing machines manufactured by LLC “Kinematics”. Additionally, LLC “Kinematics” has experience using cheaper microprocessor systems based on “Arduino” controllers, the PIC18F4620 microcontroller by “Microchip”, and similar devices.

Author of the article : Feldman Valery Davidovich
Editor and translation : Nikolai Andreevich Shelkovenko

I apologize for possible translation errors.

.

en_USEnglish