Skip to content

Select your region & language

Global

Region

Frequently Asked Questions about Measurement - Part 21: "Calibration of Accelerometers using Vibration Calibrators"

This measurement column addresses frequently asked questions received by our customer support center and provides answers to those questions.

When analyzing acceleration signals recorded by an accelerometer using analysis equipment such as an FFT analyzer or data station, the sensitivity value of the accelerometer may be input directly, or calibration may be performed using a sensitivity calibrator for the accelerometer.

If calibration is performed with incorrect connections or settings, or without noticing equipment malfunctions or broken cables, the displayed values may appear correct, but the actual measurements will not be accurate. This time, we will introduce several examples of incorrect calibration for charge output type accelerometers.

A sensitivity calibrator for an accelerometer is an exciter that vibrates at a fixed magnitude and frequency (e.g., 159.2 Hz, 10 m/s²). By attaching the accelerometer to this exciter and checking the voltage of the signal output from the accelerometer using an analysis device, the sensitivity of the accelerometer can be determined. This verifies the operation of the accelerometer, amplifier, and analysis device, and allows you to obtain a sensitivity value that corrects for sensitivity deviations due to aging and deviations in the input characteristics of the analysis device.

Normal calibration signal

Figure 1 shows an example of connecting a charge-output type accelerometer to an analysis device.

  • Figure 1: Example of connection between a charge-output type accelerometer and an analysis device.
    Figure 1: Example of connection between a charge-output type accelerometer and an analysis device.

Our DS-3000 data station and CF-9000 FFT analyzer cannot be directly connected to charge-output type accelerometers. Products such as our CH-1200A charge amplifier and CH-6130/6140 charge converter must be connected in between to allow the accelerometer to output power.
The load signal needs to be converted into a voltage signal.

If your accelerometer has a built-in preamplifier, you can connect it directly. Our accelerometer
Of the units, the NP-2000 series is a charge output type, and the NP-3000 series is a preamplifier-integrated type (voltage output).
It is a charge output type. The easiest way to distinguish between a charge output type and a preamplifier-integrated type is by the type attached to the accelerometer.
The first step is to check the product specifications sheet. Products with a sensitivity unit of pC/(m/s²) are charge output type. Products with a built-in preamplifier have a sensitivity unit of mV/(m/s²).

Figure 2 shows the results of measuring the calibration signal of the accelerometer sensitivity calibrator at 10 m/s² and 159.2 Hz when our NP-2710 charge output type accelerometer was connected to a DS-3000 data station via a CH-6130 charge converter. The sensitivity of the charge output type accelerometer was confirmed to be 0.297 pC/(m/s²) using the sensitivity calibrator. The gain of the CH-6130 charge converter is 1.0 mV/pC, so the combined sensitivity of the accelerometer and the CH-6130 charge converter is 0.297 mV/(m/s²). The magnitude of the calibration signal is 10 m/s² in RMS and 28.2 m/s² in both amplitudes, so the voltage of the calibration signal is calculated to be 2.97 mV in RMS and 8.40 mV in both amplitudes. The difference between the maximum and minimum values of the voltage waveform (upper panel) is approximately 8.45 mV, indicating they are nearly identical.

The difference between the maximum and minimum values of the acceleration waveform (middle panel) calibrated with the sensitivity calibrator is 28.4 m/s², which is in close agreement with the 28.2 m/s² of both amplitudes of the calibration signal. The 160 Hz component and overall values of the power spectrum (bottom panel) are 9.80 m/s² and 10.0 m/s², respectively, which are in close agreement with the magnitude of the calibration signal.

  • Figure 2. Normal calibration signal (Top: Voltage waveform, Middle: Acceleration waveform, Bottom: Power spectrum)
    Figure 2. Normal calibration signal (Top: Voltage waveform, Middle: Acceleration waveform, Bottom: Power spectrum)

When CCLD (constant current drive) is turned OFF

CCLD (Constant Current Drive) is a mechanism for supplying power from analysis equipment to sensors and preamplifiers. Some manufacturers call it IPC or IEPE, but it's the same mechanism. (CH-6130/6140)
The CCLD converter operates using the CCLD power supply, so when using it, connect the CCLD to the analysis device, etc.
A power supply is required.

Figure 3 shows the calibration signal measured with the CCLD power supply from the DS-3000 turned OFF.
The difference between the maximum and minimum values of the calibration signal voltage waveform (upper panel) is 70.8 μV, which is a very small value compared to the actual value (8.40 mV). However, if the vertical axis of the graph is set to autoscale, a waveform close to a sine wave will be displayed, which could lead to calibration being performed without noticing a setting error.

The acceleration waveform and its power spectrum when calibration with a sensitivity calibrator is performed in this state are shown in the middle and lower sections of Figure 3. The overall sensitivity is 0.297 mV/(m/s²), so the EU value should be around 2.97E-4V/EU, but the actual EU value is 2.58E-5 V/EU. This indicates that the sensitivity of the acceleration detector is only 0.0258 mV/(m/s²), which is an incorrect value.

The voltage waveform (top) and acceleration waveform (middle) have a DC offset (DC component) superimposed on them. The magnitude of the DC offset is -189.2 μV and -7.36 m/s². When 0 V is input, the analysis device should display it as 0 V, but in reality, it displays a value slightly deviating from 0 V. This is the DC offset error. According to Figure 3, the DC offset is 189.2 μV, but due to an incorrect calibration, an acceleration waveform is obtained that makes it appear as if vibrations of 7.36 m/s² are occurring. This is due to a calibration error and the DC offset error, and is not an actual vibration occurring.

Furthermore, although not noticeable in the waveform shown in Figure 3, if measurements are taken with the CCLD power supply turned OFF, a waveform with added noise may be obtained.

When performing calibration using a sensitivity calibrator, you can notice such setting errors by checking the voltage waveform before calibration, as its amplitude will be abnormally small.

  • Figure 3 Calibration signal when CCLD is OFF (Top: Voltage waveform, Middle: Acceleration waveform, Bottom: Power spectrum)
    Figure 3 Calibration signal when CCLD is OFF (Top: Voltage waveform, Middle: Acceleration waveform, Bottom: Power spectrum)

Similarly, when using an accelerometer with a built-in preamplifier, the analysis device must supply CCLD power. As in the case above, if calibration and measurement are performed with the CCLD power supply turned OFF, the voltage will be very low, resulting in a signal with DC offset and noise. In such cases, please check whether your accelerometer requires CCLD power and then change the settings accordingly.

When a charge output type accelerometer is directly connected

Our DS-3000 data station and CF-9000 FFT analyzer cannot be directly connected to charge output type accelerometers. A CH-6130/6140 charge converter or CH-1200A charge amplifier must be used in between.

Charge converters and charge amplifiers are amplifiers that convert the charge signal output from an accelerometer into a voltage signal. The input side is a miniature connector (No. 10-32 UNF) and the output side is a BNC connector. On the other hand, there are also products such as the NP-0021 Miniature/BNC Conversion Adapter that simply convert the connector of the sensor cable. The NP-0021 Miniature/BNC Conversion Adapter looks similar to a charge converter, but it is not a substitute for a charge converter.

  • Figure 4 shows the CH-6130 charge converter (left) and the NP-0021 miniature/BNC conversion adapter (right).
    Figure 4 shows the CH-6130 charge converter (left) and the NP-0021 miniature/BNC conversion adapter (right).

Our NP-2710 charge output type accelerometer can be used with a miniature/BNC adapter in the DS-3000.
Connect to the accelerometer station and input the calibration signal of 10 m/s², 159.2 Hz for the accelerometer sensitivity calibrator.
The measurement results are shown in Figure 5. This is not the correct connection or measurement method.

When connected correctly, the amplitude values of both voltage signals should be 8.40 mV.
The difference between the maximum and minimum values is 6.37mV, which appears to be a normal value.

The sensitivity of the charge-output type accelerometer used is 0.297 pC/(m/s²), so the EU value should be around 2.97E-4 V/EU. If calibration is performed using a sensitivity calibrator in this state, the EU value becomes 2.26E-4 V/EU. This corresponds to the accelerometer's sensitivity of 0.226 mV/(m/s²). Although the units of sensitivity are different, if you only focus on the numerical value, you will get a value that is roughly in the same order of magnitude. Therefore, if you do not realize that the connection is not correct, you may proceed with calibration and measurement in this state.

 

  • Figure 5. Calibration signal when a charge output type is directly connected (Top: Voltage waveform, Middle: Acceleration waveform, Bottom: Power spectrum)
    Figure 5 Calibration signal when the charge output type is directly connected
    (Top: Voltage waveform, Middle: Acceleration waveform, Bottom: Power spectrum)

Regarding the waveform displayed when a charge output type accelerometer is directly connected.

When a charge-output type accelerometer is connected directly to an analysis device without going through a charge converter or charge amplifier, a waveform that appears correct at first glance may be displayed.

The time waveform of a charge signal with amplitude Q[C] and frequency f is given by equation 1, where t is time.
It's in between.

q(t) = Q × sin (2πft) Equation 1

Differentiating the charge signal yields the current signal i(t) as shown in Equation 2.

  • Regarding the waveform displayed when a charge output type accelerometer is directly connected.

formula 2

The voltage signal v(t) obtained by inputting this current signal into an analyzer with input impedance R[Ω] is shown in Equation 3.

v(t) = R × i(t) = 2πf × R × Q × cos (2πf) Equation 3

If the input impedance R is 1 MΩ and the frequency f is 159.2 Hz, then equation 4 is obtained.

v(t) = 2π × 159.2 × 1000000 = 10 9 × Q × cos (2πf) Equation 4

When a charge signal of magnitude Q = 1 pC is input, the magnitude of the voltage signal observed by the analysis device is
This becomes 1 mV, which is 109 times that value. Therefore, the value obtained by performing such an incorrect calibration is...
The sensitivity value will be almost identical to the value obtained by simply changing the unit from pC/(m/s²) to mV/(m/s²) while keeping the numerical value the same.

However, when converting the values from pC to mV, the values match only when the vibration frequency is 159.2 Hz; they do not match at other frequencies. Measuring the actual acceleration signal in this state will not yield accurate results, so it is necessary to connect it correctly using a charge converter or similar device.

summary

When analyzing acceleration signals recorded by an acceleration detector using analysis equipment such as an FFT analyzer or data station, calibration may be performed using a sensitivity calibrator for the acceleration detector.
If calibration is performed with incorrect settings or connections, a seemingly correct acceleration signal may be observed, but the actual measurement will not yield accurate results.

This time, we introduced the types of calibration signals that can be obtained when an accelerometer fails to provide accurate measurement results, specifically when the CCLD power supply is not supplied and when a charge output type accelerometer is connected without using a charge converter. If calibration or measurement fails even after correct connection and settings, the cause may be a broken cable, a malfunction of the accelerometer or analysis equipment, or electrical noise, so please check each of these possibilities one by one.

(Excerpt from the email newsletter issued on December 19, 2018)