Data Classification
Numerical technical claims in this knowledge center are classified by type so that a specification value, a measured value and a design assumption are never confused.
The seven data types
| Data Type | Meaning |
|---|---|
| Manufacturer Specification | Value specified by the manufacturer for a product — the defined property range of that product. |
| Typical Value | Representative value under stated conditions. Not a guarantee. |
| Laboratory Test Result | Result obtained under stated laboratory test conditions, with the test method where available. |
| Field Test Result | Result from a specific real application. Not universal product performance. |
| Design Value | Engineering assumption used for design or sizing. |
| Literature Value | Value reported by an external technical source. |
| Derived Value | Value calculated from documented source data. |
Rules of use
- Typical values are not guarantees. A "typical" adsorption capacity does not promise that capacity in every application.
- Field results are not product capability. A single application test shows what happened in that application under those conditions. It does not establish the same result elsewhere.
- Design values are not measurements. A design assumption must not be quoted later as measured performance.
- Literature values stay attributed. A value from an external source is only as good as that source, and is cited as such.
- Derived values show their inputs. A calculated percentage must state the source data it was calculated from.
How to read a classified claim
Example: "An Xuanbao application test recorded CO concentration reduction from 1,499 ppm to 18 ppm (Field Test Result; derived reduction ≈ 98.8%)."
This tells you: what was measured, where the numbers come from, and that the percentage is derived — not a rated performance of the catalyst in general.
Why classification matters
Activated carbon and catalyst performance depends on the operating conditions: temperature, humidity, concentration, flow, residence time, poison content. A number without its type and its conditions cannot support an engineering decision. Classification is the difference between documentation and advertising.