Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
D2905-97 – Standard Test Method Technical Guide
ASTM D2905 – 97 (Reapproved 2002) defines a standardized practice for determining the number of specimens required per laboratory sampling unit when testing textiles. While small fixed numbers are generally preferred, this standard provides essential guidance for cases where interlaboratory variability is expected to differ significantly. It offers six distinct texts tailored to specific statistical conditions, ensuring robust and reliable average quality data.
📐 Scope and Applicability
The scope of this practice is to provide a mechanism for calculating the number of specimens per laboratory sampling unit and to provide recommended texts for exceptional cases. It is intended for use when the task group writing a test method anticipates that variability among observations may differ significantly from laboratory to laboratory. Paragraph A14.8 of the Recommendations on Form and Style specifies that statements on the number of specimens be included in all test methods.
The practice covers six recommended texts, chosen based on the purpose of the test and available information. The table below outlines these conditions, each paired with a specific scenario for statistical calculation.
| 🟦 Text | 📏 Condition | 📐 Section |
|---|---|---|
| 1 | Standard Deviation with Two-sided Limits | 10, 11 |
| 2 | Standard Deviation with One-sided Limits | 13, 14 |
| 3 | Coefficient of Variation with Two-sided Limits | 16, 17 |
| 4 | Coefficient of Variation with One-sided Limits | 19, 20 |
| 5 | Variability Known, Fixed Number of Specimens | 22 |
| 6 | Variability Not Known, Fixed Number of Specimens | 24 |
⚙️ Methodology and Variability Assessment
Recommended Texts 1 through 4 include examples derived from interlaboratory test data. Each example illustrates the decisions needed to calculate the numerical values required for inclusion in a specific text. A critical feature of these texts is that they each describe two distinct conditions: (1) the procedure to follow when the user has a reliable estimate of the method’s variability within their own laboratory, and (2) the fixed number of specimens required when the user does not have such an estimate.
Texts 5 and 6 cover fixed-number scenarios where variability is either known or not known. The instructions are specifically applicable to methods based on the measurement of variates, ensuring that the calculations for test precision remain statistically valid.
⚠️ Important Applicability Note: This practice is specifically designed for test methods based on the measurement of variates. It is NOT generally applicable to data based on attributes and is not usually used for methods based on “go, no-go” tests.
💡 Key Calculation Insight: When assessing a lot for acceptance testing, remember to multiply the number of specimens required for each laboratory sampling unit by the total number of such units in the entire lot sample to get the total specimens needed for a pass/fail decision. Ensure instructions for the number of units are in the applicable material specification or agreement.
📊 Relating to Other Standards and Sampling Plans
ASTM D2905 does not specify overall sampling plans; it assumes this is covered elsewhere in the test method. It works in concert with other ASTM standards such as D2906 (Statements on Precision and Bias for Textiles), D4271 (Writing Statements on Sampling in Test Methods for Textiles), and E122 (Choice of Sample Size to Estimate a Measure). The total number of specimens for lot acceptance is derived by multiplying the per-unit requirement by the number of units in the lot sample. The standard provides the framework for writing the specimen count statement within the overall test method structure.
It is crucial to understand that this practice is not suitable for attribute-based “go, no-go” tests. It is exclusively designed for methods that produce measurable, normally distributed data, allowing for the calculation of standard deviations and coefficients of variation.
❓ Frequently Asked Questions
🔍 What is the primary purpose of ASTM D2905?
ASTM D2905 provides a standardized mechanism for calculating the number of specimens needed per laboratory sampling unit for textile test methods. It offers six recommended texts to handle different conditions of variability, helping ensure that test results are statistically sound, especially when variability is expected to differ significantly between laboratories.
💡 How do I calculate the total number of specimens for a lot acceptance test?
According to Section 1.5, to obtain the total number of specimens for accepting or rejecting a lot, you must multiply the number of specimens required for each unit or laboratory sample by the total number of such units in the entire lot sample.
⚡ What is the key difference between Text 5 and Text 6?
Text 5 (Section 22) covers the scenario where “Variability is Known” and a fixed number of specimens can be specified. Text 6 (Section 24) covers the scenario where “Variability is Not Known,” requiring a different fixed number or estimation procedure.
📌 Is ASTM D2905 applicable to “go, no-go” or attribute tests?
No. Section 1.4 explicitly states that the instructions in this practice are specifically applicable to methods based on the measurement of variates. They are not generally applicable to data based on attributes and are not usually used for methods based on “go, no-go” tests.