When a supplier answers “±1% accuracy,” the next question is essential: 1% of full scale, or 1% of the current reading? The same number can imply very different performance at the normal operating point.
1. Two different percentage bases
% FS — percentage of full scale: the absolute error limit is calculated from the upper limit of the calibrated range. Under the stated specification conditions, the same absolute limit applies across the specified range.
% RD — percentage of reading: the error limit scales with the current reading. Specifications may also include a full-scale component or a lower operating limit, so read the complete definition.
For a 10 SLPM range with ±1% FS accuracy:
| Operating flow | Absolute error limit | Error relative to reading |
|---|---|---|
| 10 SLPM (full scale) | ±0.1 SLPM | 1% |
| 5 SLPM | ±0.1 SLPM | 2% |
| 1 SLPM | ±0.1 SLPM | 10% |
At lower flow, the same absolute error is a larger proportion of the reading. Oversizing is therefore an accuracy issue, not just a price issue. With an FS specification, a larger calibrated range also increases the absolute error allowance.
2. Accuracy, linearity and repeatability are different
| Property | What it asks | Why it matters |
|---|---|---|
| Accuracy | How close is the result to the reference value? | Confidence in the measured quantity |
| Linearity | How far does the response depart from the specified linear relationship? | Behavior across operating points |
| Repeatability | How consistent are repeated results under the same conditions? | Reproducibility of processes and mixing ratios |
For a multi-channel recipe, inconsistent results from one run to the next can change the mixture even when a fixed bias could be corrected by calibration. The original example compares a ±1% FS analog FL-802 accuracy option with a ±0.25% repeatability value; confirm the repeatability definition and conditions separately in the model documentation.
3. Four selection conclusions
Keep the normal point around 20–80% of full scale when practical. A normal 3 SLPM point on a 5 SLPM range is at 60%. At ±1% FS, the error allowance is ±0.05 SLPM, about 1.7% of the reading. A 100 SLPM range used at 3 SLPM would have a much larger relative allowance.
Check the usable control range. A controller may specify 1–100% of full scale. Below the stated range, performance is not assured. A usable lower limit is not a promise that relative accuracy at that limit equals the FS percentage.
Confirm the complete basis. Ask whether accuracy is FS or RD, which calibrated range it applies to, the test conditions, and the separate repeatability specification. Current Flowmethod accuracy options shown on this website are 1% / 0.5% / 0.1% FS, with the available options depending on the model; analog -A pages show 1% FS.
Buy the performance the process needs. Tighter accuracy or a reading-based specification can increase cost and calibration effort. Routine process monitoring may not justify specifications that the application cannot use. If a special metrological requirement exists, confirm it explicitly rather than assuming a standard process instrument meets it.
Consider accuracy basis, calibrated range and repeatability together. Give the engineer the normal flow, gas, and inlet/outlet pressure ranges to obtain a meaningful recommendation.
For selection, installation or technical enquiries, contact Flowmethod: Mr. Dou, +86 180 3818 3381; doufeijin@hotmail.com. See Contact for the full contact list.