The systematic error is also called the regular error.
It is that under certain measurement conditions, when the repeated measurement is performed on the same measured size, the magnitude and sign of the error value (positive or negative value) remain unchanged; or when the condition changes, it changes according to a certain law. Error; the average of the results obtained by performing an infinite number of measurements on the same measurement under repeated conditions.
Causes of systematic errors
The systematic error is caused by fixed or factors or factors that change according to certain rules, mainly including the following factors:
1. Instrument and device factors
Deviations between the measured results and the measured true values due to the inaccuracy of the instrument itself, such as the use of uncalibrated or calibrated instruments, measuring instruments, etc., can cause instrument errors.
Or due to the shortcomings of the design principle of the detection instrument and the device structure, such as the error caused by the linear displacement of the gear lever micrometer and the disproportion of the rotation angle; the manufacture and installation of the instrument parts are incorrect, such as the scale deviation of the scale, the dial and the pointer Installation of eccentricity, the length of the arm of the balance varies.
2. Environmental factors
The deviation of the measured value at the actual ambient temperature and the standard ambient temperature, and the deviation of the temperature, humidity and atmospheric pressure to be measured according to a certain rule during the measurement.
3. Factors in the determination method
It is the error caused by the measurement method itself, or the error caused by the test method itself is not perfect, using the approximate measurement method or empirical formula. For example, in the gravimetric analysis, systemic errors in the measurement may occur due to dissolution of the precipitate, coprecipitation, precipitation decomposition or volatilization during burning.
4. Personnel factors
Due to the physiological defects, subjective prejudice, bad habits of the operator, etc., such as personal characteristics or irregular operation, such as when estimating the reading on the scale, it is customary to be biased in a certain direction, the value of the burette is high or low, and the color of the end point is determined. Errors caused by being deep or shallow. Errors due to personnel factors are generally referred to as operational errors.
5. Factors in using reagents
The deviation between the measurement result and the actual result caused by the impurity water used in the test or the impure reagent used.
System error reduction and elimination method
1. Eliminate systematic errors from the root cause of the error
Before the measurement, the system personnel are required to conduct a serious analysis of the systematic errors that may occur during the detection process. It is necessary to foresee as much as possible the sources of possible systematic errors and try to eliminate or minimize the effects.
For example, check the performance of the instrument before the measurement, so that the environmental conditions and installation position of the instrument meet the requirements of the inspection technical requirements; correct adjustment of the instrument before use; strict inspection and analysis of the measurement method is correct to eliminate the instrument, detection The systematic error caused by the method, environment and other factors; in order to prevent the accuracy of the instrument from being reduced due to long-term use, the metering department is sent to the metrology department for periodic verification.
2. Use correction method to eliminate systematic error
This method is to take measurement instruments such as burettes, pipettes, and volumetric flasks, and correct them before measurement to make a calibration curve or error table. After the measurement, the actual measurement value is corrected to avoid or eliminate the result. The resulting system error.
3. Use blank experiment to eliminate systematic error
The blank test refers to the measurement performed under the same operating conditions according to the analytical test method standard or procedure without adding a sample.
The value of the result of the blank test is a blank value. Then, the measured sample is added to the test sample according to the analytical test method standard or the procedure under the same operating conditions to obtain the measured value of the sample, and finally the blank value is subtracted from the measured value of the sample, and a relatively accurate analysis result is obtained. This eliminates systematic errors caused by impurities in the distilled water or impure reagents.
4. Use controlled experiments to eliminate systematic errors
The control test is to use the same analytical method to replace the sample in parallel with the standard under the same conditions. The control results can be corrected by controlled tests to eliminate systematic errors.
5. Unchanged system error elimination method
For fixed system errors in the measurement process, the following elimination methods can be used:
1 exchange method:
According to the cause of the error, some conditions that cause the system error are exchanged, and other conditions remain unchanged, so that the factors that cause the system error have the opposite effect on the measurement result, thereby eliminating the system error.
If weighing with an equal arm balance, there will be a slight difference in the length of the two arms on the left and right sides of the balance, and a constant system error will occur during weighing. If the weighed goods are exchanged with the weight on the scale of the balance, weigh twice, and the average of the two measurements is the final measurement result of the measured object, which can eliminate the balance of the two arms. Systematic error.
2 offset method:
That is, two measurements are required to change certain conditions in the measurement, such as measurement direction, voltage polarity, etc., so that the systematic errors of the two measurements before and after are equal and opposite signs, and the average of the two measurement results can eliminate the system. error.
3 alternative method:
This method is to replace the measured condition with a known standard quantity without changing the measurement condition, and perform the measurement again to obtain the difference between the measured value and the standard value, that is, the measured value is equal to the standard value plus the difference value, thereby achieving Eliminate the purpose of systematic errors.
4 zero indication method:
In order to eliminate the systematic error caused by the misalignment of the meter, the effect of the measured pair on the indicator meter and the known standard quantity are balanced with each other during the measurement process, so that the indicator indicates zero, and the measured value is measured. It is equal to the standard value, which is the zero method. For example, bridge circuits, potentiometers, etc., use this method to eliminate systematic errors caused by inaccurate indications.
6. Change system error correction method
1 half cycle elimination method:
For periodic errors, one measurement can be taken at intervals of half a cycle, and then the arithmetic mean of the two readings is used as the measured value, which effectively eliminates periodic systematic errors.
For example, a pointer meter, if the error caused by the eccentricity of the dial, can be eliminated by averaging the readings marked by one or several pairs of pointers separated by 180°.
2 Symmetrical measurement elimination method:
Symmetrical measurements can effectively eliminate linear system errors over time.
If the voltage is measured with a voltmeter, the voltmeter is calibrated and zeroed before measurement, and then the voltage of the voltage source is measured. As the measurement time changes, the zero point of the voltmeter gradually drifts to produce a linear system error. The difference between the voltage of the voltage source to be tested and the voltage of the standard power source can be measured at equal time intervals, and the difference between the voltage to be measured and the standard voltage shown in the voltmeter is not affected by the systematic error.
We should grasp the characteristics of systematic errors according to the specific experimental conditions, find out the main causes of systematic errors, and take appropriate measures to reduce its impact.
