SC QUALITY GUILDIM&TE accuracy specification calculator
SCQUALITYGUILD.COM/TOOLS/IMTE-ACCURACY-SPECIFICATION.HTML
Enter an instrument's accuracy specification as the maker writes it, and get the allowed error and the test limits at any reading. Add the readings from a calibration to see which points pass.
ON THE EXAM · CCT II.D, II.C, V.D.1 · FREE · NOTHING YOU TYPE LEAVES YOUR BROWSER
Q-BotClick for an explanation of this tool.For educational use. This page is a study aid. The method behind it has far more depth and detail than one page and one tool can cover, and what you see here is not everything the method offers or everything this site offers. For full understanding and application, use the governing standards and texts, and seek further education, training or qualified advice. Results depend on the data you enter, so check them before relying on them.
How accuracy specifications are written
Instrument makers state accuracy as a sum of terms, each covering a different kind of error. A typical digital multimeter specification reads ±(0.05% of reading + 0.01% of range + 2 counts). The allowed error at a reading is the sum of all the terms at that reading.
- Percent of reading (gain or scale error) grows with the reading. It is zero at zero.
- Percent of range or percent of full scale (offset error) is the same everywhere on the range. It is a percentage of the range or full-scale value, not of the reading.
- Counts or digits are steps of the least significant displayed digit. Two counts on a display with 0.001 V resolution is 0.002 V.
- Fixed terms and floors set a minimum. Some specifications add an absolute term such as + 5 µV; others say "±1% of reading or 0.5 °C, whichever is greater".
Specifications are often given in ppm instead of percent: 10 ppm is 0.001%.
Why the bottom of a range is weak
The range, count and fixed terms do not shrink as the reading gets smaller, so the allowed error as a share of the reading grows near the bottom of a range. In the worked example the meter is good to about ±0.07% of reading at full scale, but only about ±0.25% at a tenth of it. That is why procedures test near the top of each range, and why a reading low on a range is better taken on the next range down.
Qualifiers
A specification only holds under its stated conditions: for a period since calibration (90 days, 1 year), in a temperature band (23 °C ± 5 °C), sometimes after a warm-up and within a humidity limit. Outside those conditions the maker usually adds a temperature coefficient per degree. Record the qualifiers with the specification.
From specification to test limits
In a calibration the reference standard applies a known value and the instrument's reading is compared with it. The limits are the applied value ± the allowed error. The test points table gives the limits for a procedure or data sheet, and, when you add the readings, the error at each point and whether it is in tolerance. A point that uses most of its allowance needs care: once the calibration uncertainty is counted, it may not be a clear pass. The TUR and guard-band calculator handles that decision.
Specification is not uncertainty
A specification is the error the maker promises. The uncertainty of a measurement made with the instrument also includes the reference, the method, the operator and the environment. A specification is often one input to an uncertainty budget, usually as a rectangular distribution.
On the exam
CCT II.D asks you to analyze specification descriptions: percent of full scale, percent of range, percent of reading and number of counts, along with tolerance, modifiers and qualifiers, output, scale and floor terms. II.C covers resolution and readability, and V.D.1 determining conformance status. Expect to calculate the allowed error and the limits at a test point, and to decide whether a reading passes.