— A Practical Guide for Engineers, EPCs, and Project Teams

In industrial instrumentation, three terms are frequently used to describe measurement performance: accuracy, repeatability, and resolution. They are often confused, sometimes used interchangeably, and frequently misunderstood. Yet understanding the distinction between them is fundamental to selecting the right instrument for an application—and to interpreting the data it produces.

This guide explains what each term means, how they differ, and why the distinction matters in practical industrial applications.

1. The Core Distinction: A Simple Analogy

Before diving into technical definitions, consider this analogy:

Imagine you are throwing darts at a target. Your performance can be described in three ways:

A dart thrower can have:

• Good repeatability but poor accuracy — all darts cluster together but miss the bullseye

• Good accuracy but poor repeatability — darts average around the bullseye but scatter widely

• Good resolution — can make tiny adjustments to aim

• Poor resolution — cannot make small adjustments

The same principles apply to instruments. An instrument can be highly repeatable but inaccurate, or accurate but with poor resolution.

2. Resolution: The Smallest Detectable Change

Definition: Resolution is the smallest change in the measured variable that the instrument can detect and display.

Key points:

• Resolution is limited by the instrument's design and electronics

• It determines the fineness of the measurement

• It is not the same as accuracy—an instrument with high resolution can still be inaccurate

Example: A digital thermometer with a resolution of 0.1°C can display changes as small as 0.1°C. A thermometer with a resolution of 1.0°C cannot detect changes smaller than a full degree—even if the actual temperature changes by 0.5°C, the display will not change.

Resolution in analogue instruments:

• Determined by the smallest division on the scale

• A pressure gauge with 100 divisions over a 100 psi range has a resolution of 1 psi

Resolution in digital instruments:

• Determined by the number of digits displayed

• A 3½-digit display has a resolution of 1 part in 1999 (approximately 0.05% of full scale)

• A 4½-digit display has a resolution of 1 part in 19999 (approximately 0.005% of full scale)

Why resolution matters: Resolution must be sufficient for the application. If the required control tolerance is ±0.5°C, a sensor with 1.0°C resolution is insufficient—it cannot detect changes within the tolerance band.

3. Repeatability: The Consistency of Measurement

Definition: Repeatability is the closeness of agreement between successive measurements of the same variable under the same conditions. It measures how consistently an instrument gives the same reading for the same input.

Key points:

• Repeatability is expressed as a range or standard deviation

• It is measured under identical conditions (same operator, same instrument, same environment, short time period)

• It is always better than accuracy—an instrument cannot be more accurate than it is repeatable

Example: A pressure transmitter is repeatedly exposed to 100 psi under the same conditions. It reads:

• 99.8 psi, 100.1 psi, 99.9 psi, 100.0 psi, 100.2 psi

• The repeatability is approximately ±0.2 psi (the range of readings)

Why repeatability matters:

• It determines whether the instrument can reliably detect changes in the process

• It affects the ability to control processes precisely

• It indicates the instrument's stability and freedom from random errors

Distinction from reproducibility: Repeatability is under the same conditions; reproducibility is under changed conditions (different operators, different locations, different times).

4. Accuracy: The Closeness to Truth

Definition: Accuracy is the closeness of a measured value to the true value of the measured variable. It tells you how "correct" the measurement is.

Key points:

• Accuracy is expressed as an error band (e.g., ±0.25% of span)

• It includes the combined effects of all errors—linearity, hysteresis, repeatability, temperature effects, etc.

• Accuracy is the most important performance specification for most applications

Example: A pressure transmitter with ±0.25% accuracy on a 100 psi range has a maximum error of ±0.25 psi. A reading of 50 psi could represent a true value between 49.75 psi and 50.25 psi.

Accuracy specifications:

Why accuracy matters:

• Accuracy determines how closely the measurement represents the actual process condition

• It directly affects product quality, safety, and process efficiency

• For custody transfer and billing applications, accuracy has direct financial consequences

5. The Relationship: How They Connect

Resolution, repeatability, and accuracy are related, but not interchangeable:

Resolution is the foundation:

• An instrument cannot detect changes smaller than its resolution

• Resolution limits the achievable accuracy and repeatability

• In digital instruments, accuracy is often a multiple of the resolution

Repeatability limits accuracy:

• An instrument cannot be more accurate than it is repeatable

• If readings vary by ±0.2 psi under identical conditions, the accuracy cannot be better than ±0.2 psi

• Accuracy is always greater than or equal to repeatability

Accuracy is the summary:

• Accuracy combines all error sources—including repeatability, resolution, linearity, hysteresis, temperature effects, etc.

• It is the most comprehensive performance specification

• It represents the overall measurement quality

The hierarchy:

text

复制

下载

Resolution ≤ Repeatability ≤ Accuracy

Important: Resolution and repeatability are "free"—they can be high even when accuracy is poor. Accuracy is the specification that matters most for knowing how close the reading is to the actual process value.

6. Practical Examples

Example 1: Temperature Measurement

Analysis:

• The resolution is finer than the repeatability, so it is not the limiting factor

• The repeatability is better than the accuracy, as expected

• A reading of 100.0°C means the true temperature is between 99.5°C and 100.5°C

Example 2: Pressure Measurement

Analysis:

• The resolution (0.01 psi) is much finer than needed for the repeatability and accuracy

• The repeatability (±0.05 psi) is within the accuracy band

• This is a well-designed, high-performance instrument

Example 3: Level Measurement (Poor Repeatability)

Analysis:

• The resolution (1 mm) is fine, but the repeatability is poor

• The poor repeatability limits the usefulness of the instrument for precise control

• The accuracy is limited by the repeatability, as expected

7. Common Misunderstandings

Misunderstanding 1: "High resolution means high accuracy"

A thermometer with 0.01°C resolution can display temperature to 0.01°C, but if it has an accuracy of ±1°C, the reading is still only within one degree of the true value.

Reality: Resolution does not guarantee accuracy. An instrument with 0.01°C resolution and ±1°C accuracy is no more accurate than one with 0.1°C resolution and ±0.1°C accuracy.

Misunderstanding 2: "Good repeatability means good accuracy"

A pressure transmitter can produce highly repeatable readings that are consistently 2 psi too high. The repeatability is excellent, but the accuracy is poor.

Reality: Repeatability is necessary but not sufficient for accuracy. An instrument must be repeatable to be accurate, but repeatability alone does not ensure accuracy.

Misunderstanding 3: "Accuracy is always ±% of full scale"

Some specifications state accuracy as "±0.1% of reading"—which is significantly different from "±0.1% of full scale."

Reality: The basis of the accuracy specification matters. At low readings, "% of reading" accuracy is much more demanding than "% of full scale" accuracy.

Misunderstanding 4: "Resolution and repeatability can be assumed from accuracy"

They cannot. Accuracy is a summary that includes resolution and repeatability, but it does not tell you their individual values.

Reality: If an application requires detecting small changes (resolution) or consistent measurements (repeatability), these specifications must be checked separately.

8. Selecting Instruments: Which Specification Matters Most?

The priority depends on the application:

General advice:

1. Start with accuracy — it is the most comprehensive specification and usually the most important

2. Check resolution — ensure the instrument can detect changes as small as your process requires

3. Verify repeatability — ensure the instrument is stable and consistent

4. Always read the fine print — understand the basis of each specification (% of span, % of reading, etc.)

9. Why This Matters for EPC Projects

10. Why Choose Anhui Tiankang for Instrumentation?

Anhui Tiankang (Group) Co., Ltd. has been manufacturing industrial instruments for nearly five decades. Our instruments are designed, tested, and documented with clear performance specifications:

• Pressure transmitters: Accuracy up to ±0.075% FS, resolution as low as 0.01%, repeatability ≤0.05%

• Temperature sensors: RTD Class A (±0.15°C), thermocouple Class I (±1.5°C)

• Level instruments: Radar accuracy ±1 mm, repeatability ±0.5 mm

• Flow instruments: Coriolis accuracy ±0.05% of mass flow, repeatability ±0.02%

All performance specifications are verified in our CNAS-accredited laboratory before shipment.

11. Conclusion

The key takeaway: Resolution, repeatability, and accuracy are not interchangeable. Resolution tells you what the instrument can see. Repeatability tells you how consistently it sees it. Accuracy tells you how close the measurement is to the truth.

Remember: An instrument with high resolution and high repeatability can still be inaccurate. An instrument with good accuracy must have good repeatability—but the reverse is not true.

Contact Us

For instrumentation selection advice, technical documentation, or project quotations, please contact:

Yin Shuangjie
International Sales Manager
📧 Email: [email protected]
📱 WhatsApp / Zalo: +86 17856068126
🌐 Website: http://www.tiankang-global.com/

Anhui Tiankang – Your partner for reliable instrumentation solutions.