Sensitivity and accuracy

Sensitivity refers to the sensitivity of the instrument to the measured parameters, or the ability to respond to the measured quantity change. It is the ratio of the increment of output change to the increment of input variation under steady state:

Sensitivity is sometimes referred to as "magnification ratio" and also the slope of each point of the instrument static characteristics. Increase the magnification can improve the sensitivity of the instrument, simply increase the sensitivity does not change the basic performance of the instrument, that is, the instrument accuracy has not improved, contrary to the oscillation phenomenon sometimes, resulting in unstable output. Instrument sensitivity shall be maintained at the appropriate amount.

But for the instrument users, such as chemical industry instrument, precision instrument is an important index, but in actual use, often more emphasis on stability and reliability of the instrument, because the chemical enterprise detection and process control instrument for measurement of the number, and a large number is used to detect. In addition, the stability and reliability of the measuring instruments used in the process control system are more important than the precision.

Precision instrument precision, also known as accuracy. Accuracy and error can be said to be twin brothers,

Because of the existence of errors, the concept of precision. Instrument accuracy is simply the accuracy of the measured value of the meter to a true value, usually expressed by a relative percent error (also called relative error).

To improve the instrument accuracy, error analysis is necessary. Errors can be divided into neglect error, slowly varying error, systematic error and random error. Negligence error refers to the measurement process caused by human error, one can be overcome, two and the instrument itself does not matter. The slow change error is caused by the aging process of the internal components of the instrument, which can be overcome and eliminated by changing components, components or by constant correction. System error refers to the repeated measurements of the same measured parameters, the numerical error occurred in size or have the same sign, error or change according to certain rules, but has not yet been recognized in the accident caused by the factors, the magnitude and nature are not fixed, it is difficult to estimate, but it can be estimated influence on the test results from the theory by statistical methods. Error sources mainly refer to systematic error and random error. When the accuracy is expressed by error, the sum of random error and system error.

Any instrument has a certain error. Therefore, the instrument must first know the accuracy of the instrument,

In order to estimate the difference between the measured results and the agreed truth values. The accuracy of the instrument is usually used to allow the maximum error of the reference remove percent (%) after the digital measure.

Analog meter accuracy in general should not use absolute error (and the true value of the difference value measurement) and relative error (absolute error and the true value ratio) to said, because the former does not reflect the reasonable requirements of different instrument, which is easy to cause any instrument can not believe the misunderstanding. For example, for a full range of 100mA current meter, measured in zero current, due to mechanical friction that shows the number of hands slightly deviation from zero and 0.2 mA readings, according to the relative error of the algorithm, then the relative error of the point is infinite, it seems that this instrument is completely unable to use but in the engineering staff, the measurement error that is easy to understand, not worth it one may still get excited over a little thing, the more sophisticated instrument! The reasonable accuracy of the analog instrument shall be measured by the maximum absolute error in the measuring range and the ratio of the measurement range of the instrument to the relative (full-scale) percent error.

According to the instrument industry, the accuracy of the instrument is divided into a number of grades, referred to as precision level, such as 0.1, level 0.2, level 0.5, level 1, level 2.5, etc.. Thus, the smaller the accuracy level, the higher the accuracy.