The nominal accuracy of electronic belt scales typically ranges from 0.25 to 1.5 levels, with the guaranteed accuracy at the factory usually between 20% to 99.9%. For instance, Remsey Company in the U.S. sets it at 25% to 99.9%; both the draft recommended by the International Organization of Legal Metrology (OIML) and the procedure N075/410/EEC of the European Economic Community (EEC) confirm a guaranteed accuracy range of 20% to 99%, while Chinese belt scale manufacturers generally adhere to a range of 20% to 99.9%.

However, due to the continuous automatic weighing of belt scales under dynamic conditions, especially electronic belt scales used in thermal power plants, they often operate in harsh environments such as dirt, vibration, dust, and electromagnetic interference. Factors like changes in belt tension, uneven material loading, belt misalignment, and variations in temperature and humidity make it difficult to maintain the dynamic accuracy of the electronic belt scales in use within the factory-warranted precision range.

We must enhance our research on the dynamic characteristics of belt scales, enabling them to automatically overcome or minimize the impact of various adverse factors, thereby maintaining high dynamic accuracy for electronic belt scales. Additionally, regular calibration of the belt scales is essential, with frequent adjustments made to improve their dynamic precision.

Currently, there is no unified method or standard for belt scale calibration in our country. At the 6th International Conference on Legal Metrology, the United Kingdom proposed a "50th International Recommendation" (a set of test regulations for belt scales), but the conference did not establish a uniform regulation. The actual calibration methods used domestically and internationally include simulated calibration (such as hanging weight calibration and chain weight calibration, etc.) and material calibration.

The operation of the hanging weight calibration is simple and quick, but the calibration error is relatively large, making it suitable only for general auxiliary verification methods. Calibration using the accumulated value has less error than the instantaneous value, hence, U.S. regulations stipulate: for belt scales of 0.5 grade during hanging weight calibration, the accumulated weight indicator value should be used and at least 500 readings or 20 minutes of accumulation are required; for belt scales of 1.0 grade, at least 200 readings or 10 minutes of accumulation are necessary.

Chain coding calibration has taken a significant leap forward compared to挂牌calibration. Japan has conducted extensive research on chain coding calibration, leading to standardized and serialized products. For example, the KUBOTA Corporation offers a series of roller chains with models like 3, 5, 7, 10, and so on, up to 200, 300, 400, 500 tons/m, totaling 20 different types. Our Chengdu Scientific Instruments Factory mass-produces two types with 8 and 20 tons/m. Due to the length and weight of roller chains, calibration generally requires the use of hoisting and fixation devices. Additionally, it's crucial to ensure the center line of the chain aligns with the belt center line, the belt doesn't deviate, and the chain maintains consistent contact with the belt throughout. These conditions are challenging to achieve, which is why chain coding calibration is rarely used in China.

Proof in Practice: Methods such as code method, chain code method, and other simulation calibration methods can only serve as a means for debugging, inspection, and maintenance of belt scales. The ultimate dynamic accuracy can only be achieved through material calibration. Because it not only includes the basic error of the measuring equipment but also incorporates errors influenced by operation, maintenance, material, and equipment status, providing users with practical and accurate measurement standards. As per error theory, if the error of a higher standard is compared to that of a lower standard or a common measuring instrument at a ratio of 1/5, the former is considered to be the relative true value of the latter. The definition of true value is: the actual value体现ed by the effect of a certain quantity.

Thus, the actual value is a relative true value that can be known and applied in reality. In other words, when the standard instrument is calibrated to a degree of 1/5 or more, it can serve as the standard for the instrument being calibrated. Since the belt scale's stated accuracy can reach 0.5% F.S, and some even reach 0.25% F.S, it is not advisable to use outdated methods like manual handling, sorting, and cumulative measurement for calibrating the material on the belt scale. Not only does this consume a great deal of manpower, but the accumulated error far exceeds the precision required for a standard instrument, leading to false calibration results. Only an electronic hopper scale with an accuracy of 0.1 level or even 0.05 level can be competent for the actual calibration of the belt scale's material.