  +44 (0)1424 852744 Resources

| Novatech Assessing Errors

Most performance specifications for loadcells are generally quoted as percentages of rated load. It is important to note that the magnitude of the error may translate throughout the measuring range of the loadcell. The first step in assessing errors is to equate the quoted error to a quantity such as force or loadcell output. This value can then be used to calculate the typical measurement error at a particular force, which is usually the information that is required by most users. The following example demonstrates this for a single error, repeatability. Repeatability is defined in Engineering Sheet E004. The table below lists the parameters required for the calculations.

 Parameter Value Unit Rated load 200 kgf Repeatability ±0.02 % Rated Load Rated output 2.0 mV/V

If we are interested in two measuring points, 10kgf and 50kgf, we can calculate the typical errors. These values show that proportionately the measurement point becomes more erroneous as it progresses down the rated load range. This is why we manufacture loadcells of so many load ratings, to ensure that the errors due to the loadcell are minimised for the required measurement range. The above example deals with just one type of error, it is easy to see that the other errors that may come into play can greatly affect the measurements accuracy due to the cumulative effect.

If we assume that the loadcell is powered by a 10V excitation supply we can calculate the repeatability error in terms of the loadcell output in millivolts. Loadcell rated outputs are normally given in millivolts per volt of excitation supply (mV/V), multiplying this value by the excitation supply voltage gives the output at the rated load. This is useful information when assessing the measurement errors caused by the instrumentation used to measure the loadcell output.

Novatech Measurements Limited

83 Castleham Road, St Leonards on Sea, East Sussex, TN38 9NT, England.
Telephone: +44 (0)1424 852744
Fax: +44 (0)1424 853002 