The servo tilt sensor LCF-100 from JEWELL (www.jewellinstruments. com) in the United States has many applications in China, and users feel that this sensor is very accurate. There are many indicators of inclination sensors, such as temperature drift, nonlinearity, zero bias, etc. How to analyze these indicators and which indicators are key indicators are not clear to many users and even manufacturers. This article will provide detailed explanations.
Accuracy Analysis of LCF-100 Sensor
The LCF-100 product measures tilt angle based on the principle of force balance servo acceleration. It measures the component of gravity acceleration on the sensitive axis of the accelerometer and converts it into angle data, that is, the inclination value and acceleration value have a sine relationship. The measurement accuracy of LCF-100 is closely related to the following indicators:
1. Sensitivity error: depends on the inherent characteristics of the core sensitive device, but is also associated with frequency response, also known as amplitude frequency characteristics. Sensitivity is the foundation of sensor accuracy, and sensors with low sensitivity will inevitably have poor accuracy. This is a key factor in determining accuracy.
2. Zero bias: Depending on the inherent characteristics of the core sensitive device, it refers to the actual output angle value of a sensor that is non-zero when there is no angle input (such as absolute horizontal plane). This indicator can be eliminated through calibration, so it is not a critical indicator.
3. Nonlinear: It can be corrected through subsequent calibration, depending on the number of calibration points. The more correction points there are, the better the nonlinearity. The better the nonlinearity, the more accurate the measurement angle, so this indicator is the key indicator.
4. Horizontal axis error: refers to the error generated when a sensor is coupled to the output signal of the sensor when a certain acceleration is applied perpendicular to its sensitive axis or tilted at a certain angle. This is a key indicator.
5. Input axis misalignment: refers to the angle between the actual sensitive axis of the sensor and the edge of the base. The tilt angle sensor requires that the tilt direction be parallel or coincident with the designated edge of the sensor during installation. When the sensitive axis of the tilt sensor itself does not coincide with the actual tilt direction, the additional error generated varies sinusoidally as the tilt angle increases. The misalignment of the input axis can generally be eliminated through calibration, so this indicator is not a critical indicator.
6. Repeatability and hysteresis: depend on the inherent characteristics of the core sensitive components and cannot be improved through subsequent corrective measures. This is a key indicator.
7. The influence of temperature on zero point and sensitivity: It also includes the repeatability of drift and temperature curves, which depends on the inherent characteristics of the core sensitive device and cannot be improved through subsequent correction measures. In the case of determined repeatability, subsequent calibration can be carried out depending on the number of calibration points (angle points and temperature points). The more calibration points there are, the better the temperature drift accuracy. This is also a key indicator.
From this, it can be seen that the errors of inclination sensors include sensitivity errors, zero bias, repeatability, and temperature drift repeatability, as well as horizontal axis errors, input axis misalignment, nonlinearity, temperature drift linearity, etc. Some errors are improved through correction and compensation measures.
According to the performance indicators of LCF-100 publicly released by Jewell company, the following error calculations are performed:
From the analysis in the table above, it can be seen that when the temperature is constant and all other error terms have been strictly calibrated, the tilt sensor can achieve an accuracy of 0.0028 degrees, which is a relatively high indicator level. This is much higher than the accuracy index of typical tilt sensor products. At the same time, we see that if the temperature changes, the accuracy drift caused by temperature changes will be a major error term, so maintaining a constant temperature is the key to ensuring sensor accuracy.
We also need to strictly calibrate the error terms that can be eliminated through calibration. If the calibration is not done well enough, it will cause the sensor accuracy to deteriorate rapidly. Generally, the following measures need to be taken to handle the errors:
1. Precision machining of the flatness and parallelism of the sensor bottom surface;
2. Perform software correction on the horizontal axis error;
3. Software correction for nonlinearity;
4. Software correction of errors caused by misalignment of the input shaft;
5. Perform noise filtering processing;
6. Reduce resonance caused by external low-frequency vibrations that affect filtering function;
7. Implant the temperature sensor into the product to detect the actual temperature drift data of the sensor as much as possible;
8. Use high-resolution AD to reduce quantization errors;
9. Adopt good EMC measures to reduce external electromagnetic interference;
10. Use high-precision standard equipment for testing and calibration.
Appendix
The core technological advantages of Huilian Technology's tilt sensor
Huilian Technology's tilt sensor has excellent angle measurement performance, creatively incorporating horizontal axis error and zero offset as important components of measurement accuracy. (Note: Other general tilt sensor products only consider nonlinearity as the measurement accuracy of the sensor). The Huilian tilt angle sensor combines the "horizontal axis error" with the "allowable input axis non coincidence degree", greatly solving the problem of achieving truly accurate tilt measurement at any angle point:
1. The actual tilt of an object cannot occur completely along the strict X and Y orthogonal axes, as factors such as the clearance of mechanical components and the difficulty in determining the actual tilt axis determine that the actual tilt angle is not in the strict X and Y orthogonal directions. When there is a tilt in the X-axis direction, if the horizontal axis error of the sensor is too large, it will cause a change in the tilt angle data in the Y-axis direction, but in reality, the Y-axis direction may not have truly tilted, and vice versa. Therefore, even if the linearity is high, the actual measurement accuracy of an ordinary tilt sensor with a horizontal axis error of 3-5% will still be around 3-5%, rather than linearity data.
2. If the sensor does not have "allowable input axis misalignment" data during actual installation, it is difficult to obtain truly accurate tilt data through simple visual inspection. Excessive deviation angle between the sensitive axis of the sensor and the actual tilt direction of motion will result in an additional "sine error" being added to the output data of the sensor. As the tilt angle increases, the error also gradually increases.
3. Truly achieve full angle range measurement. The Huilian tilt sensor does not have a significant increase in+-90 degree error, and can achieve accurate measurement in all directions and attitudes. High accuracy and good consistency throughout the entire measurement range.
4. Implement dynamic compensation for the tilt sensor to reduce the impact of external acceleration on measurement, without reducing the measurement frequency response, while achieving high-speed and high-precision measurement.
The inclination sensor of Huilian Company has effectively solved the above practical problems and is a product that truly achieves high-precision inclination measurement.
Introduction to Huilian's core technical capabilities in inclination measurement
1. Modal testing of the inclination sensor housing is not simply about installing the sensor core components in any housing and packaging them into a product.
2. Perform modal testing on the tilt sensor PCBA board to fundamentally address the impact of resonance caused by vibration on sensor filtering.
3. Tilt angle sensors that are tested strictly in accordance with relevant domestic and international standards, including GB, GJB, MIL, IEC, ISO, EN, etc.
4. Multiple invention patents, utility model patents, appearance patents, and software copyrights in the field of inclination measurement.
5. An automated testing technology tilt sensor with independent patents reduces the impact of human factors on quality and maintains consistency in quality.
6. The accuracy can reach a comprehensive accuracy of ± 5 arcseconds @ -40~85 ℃ at all temperatures, and the zero stability over 12 months can reach ± 3.6 arcseconds. We can provide the world's most accurate tilt sensor and are still improving it.
7. We have core research and development capabilities, including three PhDs and five masters, and can customize tilt measurement products for customers with special requirements.
