Kala Scale

Weighing Accuracy

By scientific definition, the accuracy of a measurement system is how close a result comes to the true value or a standard.

In our daily life, when we see a measurement like a speed of 25 kmph on a speedometer, or a weight of 12.2 kg on a weighing scale, we regard this value as correct without thinking about the errors those values may have. It is a common perception that ‘what we SEE is the CORRECT & ACCURATE value’.

This perception is further strengthened in today’s ‘digital’ indication age. Where instruments produce a direct readout of numeric values. This removes the ambiguity of analogue indication such as needle/pointer of earlier times. However, the fact remains that each measurement, whether done by an analog or a digital instrument, has an error. We cannot say how accurate a measurement is unless we know the true value to compare it with.

Weighing equipment as a ‘system’

A measurement system can have many components. However, there is always at least one critical component which dictates and limits the overall accuracy of the measuring system.

In modern day electronic weighing systems, this most critical measuring component is the transducer. This converts applied load into a proportionate electrical signal. The majority of weighing machines used for commercial purposes have strain-gauge based Load Cells as the transducer. The voltage signal generated is processed and converted into digital form by the weighing instrumentation, commonly known as ‘digitizer’, for displaying the weight and further use. The weighing system needs to be calibrated using standard weights before putting to use.

A weighing system cannot have accuracy more than the accuracy of load cells used in it.

The role of least count

All measuring instruments have a calibrated range known as ‘span’, with a Min. and a Max. limit. This range or span is a graduated scale. The minimum value of displayed graduation is the ‘least count’ or ‘resolution’ of the instrument.

For example: a weighing scale with a least count of 10 kg will show weight only in steps of 10 kg, i.e. if the weight of an object is measured as 1016 kg, the scale can show it either as 1010 kg or 1020 kg. Here it does not have any relation whether the measurement of 1016 kg was correct or not. It is only about displaying the result. The least count of a scale can only be 1, 2, 5, 10 and their multiples.

The least-count/resolution relates more to readability of a weighing instrument, rather than accuracy.

The Legal Metrology perspective

The majority of scales which we see and use everyday, such as weighbridges, platform scales, bench / counter scales etc, are classified as non-automatic weighing machines by Vietnamese Legal Metrology (TCVN). These are further classified into four accuracy classes – I, II, III and IV depending on permissible errors for a measurement, with class I being most accurate and class IV the least.

Most of the scales used for ‘legal for trade’ purpose are certified for min. class III.

All weighing machines used for trade purposes are required to be verified and stamped every year by Legal Metrology. This is according to their accuracy class.

The accuracy classes for scales

One important specification valid for class I and II only is that the accuracy of the scale can be 1, 2, 5 or 10 times the least count of the scale. For example: a scale of 10 kg x 0.1 g can have accuracy 10 times the resolution which is 1 g (10 x0.1 g) i.e. a reading of 5000.1 g may have an error of up to 1 g.

This specification is not applicable to Class III and Class IV machines. For these machines, the accuracy of readings is between 0.5x to 1.5x of the scale resolution, or 1x as an average (simplified for ease of understanding). For example: in a scale with 50000 kg x 10 kg, a reading of 25050 may have maximum error of 10 kg, i.e. the object’s true weight could be anywhere from 25040 to 25060 kg.

For class III machines, the accuracy is generally considered as +/- 1 division (least count).

International standards

OIML (International Organization of Legal Metrology) is the most followed standard internationally, subscribed officially by more than 120 countries including Viet Nam. Additionally there is NTEP (National Type Evaluation Program) standard which is primarily followed by & Canada. (least count).

OIML has defined accuracy classes for load cells (OIML R-60) as well as weighing instruments (OIML R-76) with their relationship as below –

OIML stipulations attribute 50% of the error in a weighing system to the error in the load cells.

For a weighing instrument, the OIML compatibility checks between load cells and weighing scale include:

 a) No. of certified load cell divisions are >= the weighing scale divisions

 b) Load cell accuracy class corresponds to above table or higher

The international guidelines on weighing accuracy stress on importance to verify load cell accuracy to ensure that desired overall accuracy is achievable.

The practical application and ethical aspects

By now it should be amply clear that the load cell(s) used in a weighing scale should be able to deliver the accuracy in terms number of divisions, equal to or better than weighing scale divisions, to achieve desired accuracy.

Taking an example for a weighbridge of 50 t x 10 kg (i.e. 50000/10 = 5000 divisions), the load cell should be at least accurate and certified to 5000 divisions (OIML R-60 C5) or higher accuracy class. Similarly a platform scale of 60 kg x 20 g (i.e. 60000/20 = 3000 divisions) should be using a load cell certified for 3000 divisions (OIML R-60 C3) or higher.

Unfortunately the Indian Legal metrology, despite following OIML guidelines, is yet to specify rules for load cell certification. As a result, the market is brimming with cheap weighing scales using unapproved load cells and the least count of scale being often projected as ‘accuracy’.

However, companies in India with global standards and ethical practices, use OIML approved load cells for weighbridge applications corresponding to accuracy classes of scales manufactured by them.

Load cells are highly accurate devices but have different accuracy classes, these classes can give more information on precisely how accurate the load cell is which in turn can help with selection for certain applications.

Load Cell Accuracy

D1 – C2 – this load cell accuracy class is relatively low. It is sufficient for applications in simple building materials scales which are used to weigh cement, sand or water as well as non-trade applications.

C3 (0.0230%) – accuracy class C3 is the most common found in machine construction. Typical applications include; belt scales, platform scales and other electronic weighing devices. Of the load cells we provide, C3 is the most common accuracy class.

C4 – C5 (0.0174% – 0.0140%) – this accuracy class is higher and used for scales used for shop counter scales, filling machines, dynamic check weighing as well as platform and belt scales which require higher accuracy.

C6 (0.0116%) – C6 load cell accuracy class offer a very high level of accuracy, they are used for high precision check weighing and vehicle testing

OIML – OIML approval ensures the load cell is suitable and “legal for trade” meaning that they conform to a common agreed set f requirements. To gain OIML approval the load cell must gain an R60 certification.

Load cell accuracy classes continue into the As and Bs although these classes are not within our portfolio.

From: Internet

kala@kalascale.com