Calibration Error Calculator

Input
Reading [?]
Lowest [?]
Highest [?]
Output
Reading [?]
Lowest [?]
Highest [?]
Error
Ideal Output [?]
Error Value [?]
%FS Error [?]

Pressure calibration products

Request info on pressure calibration products for your application.

Contents

User Guide

This calibration error calculator can be used to determine the accuracy of measurement devices which have a linear input and output characteristic. The error is calculated by determining the difference between the actual output measured and the ideal output for a particular input value.

Input

Input Reading

Enter the input value measured/recorded with a calibrator or other reference instrument that has a higher accuracy than the device under test. It is important that all input values are entered in the same engineering units.

Lowest Input

Enter the lowest possible input value that the unit under test was designed to measure. It is important that all input values are entered in the same engineering units.

Highest Input

Enter the highest possible input value that the unit under test was designed to measure. It is important that all input values are entered in the same engineering units.

Output

Output Reading

Enter the output value measured/recorded with a calibrator or other reference instrument that has a higher accuracy than the device under test. If the output of the device under test is a visual display then use the displayed measurement. It is important that all output values are entered in the same engineering units.

Lowest Output

Enter the lowest possible output value that the device under test was designed to produce. It is important that all output values are entered in the same engineering units.

Highest Output

Enter the highest possible output value that the device under test was designed to produce. It is important that all output values are entered in the same engineering units.

Error

Ideal Output

This shows the theoretical output reading that would be produced if the device under test was perfectly precise. This value is compared to the actual “Output Reading” to determine the output “Error Value” of the device under test at a particular calibration point.

Error Value

This represents the difference between the actual “Output Reading” and the “Ideal Output” and is shown in the same engineering units as the output.

%FS Value

This displays the output error as a percentage of the full scale range i.e. the “Error Value “divided by the difference between “Lowest Output” and “Highest Output.”

Help

Defining full scale for compound & bi-directional pressure range errors

I was wondering what full scale you would use for the following pressure sensor ranges:

-1 to +1 bar
800 to 1200 millibar
-1 to +1.5 bar

Method A

In the context of the calibration error calculator, the full scale is determined by subtracting the maximum pressure from the minimum pressure:

2 bar
400 mbar
2.5 bar

Method B

Alternatively some manufacturers will define the accuracy based on the range from zero pressure to the maximum positive or negative pressure, whichever is the greatest, since this represents the base range of the sensor:

1 bar
1200 mbar
1.5 bar

As a further example a -1 to =0.5 bar range might use 1 bar rather than 1.5 bar for defining the accuracy, since the sensing diaphragm will be rated to that pressure.

The absolute error is of course always going to be the same, but when expressed as a percentage error it is important to use the same full scale when comparing your own to the manufacturer’s percentage error values.

Manufacturer’s Full Scale

So you have to be careful when determining the pass/fail error value for the total span of a manufacturer’s specification, because this is not always clearly defined for compound and bidirectional ranges.

If the full scale limits are not clearly defined by the manufacturer, then it is best to assume the manufacturer is using method ‘A’. This will ensure you are using the worst case error for the sensor, when assessing whether it meets your required accuracy. The downside is that you may choose not to use a particular sensor because the assumed error is too large.