Hello and welcome to an new Rule of Thumb post in which we have a closer look at understanding Measurement uncertainty and measurement error in plastics.
Back at University during my polymer engineering study, I remember, there was a paper hanging on the wall of the rheology lab which I kept all the years in my mind. On the paper was a quote, if I remind correctly, from David Packard and it stated the following:
"You are always measuring wrong, you just have to know by how much"
That quote captures a fundamental truth in polymer engineering, and physics: perfect measurement does not exist. Every measurement contains some degree of error, and the key to precision is understanding and quantifying that error.
This brings me to the standard deviation (σ) which measures the dispersion or spread of data points around their arithmetic mean. A low σ indicates data clusters tightly near the average, while a high σ shows wide scattering. It carries the same units as the data, making it highly interpretable (Figure 1).
 |
| Figure 1: The importance of mean and standard deviation in plastics testsing and analysis including an example of PP tensile strength measurement. |
Example - Standard deviation σ in polymer analysis
Standard deviation σ in polymer analysis directly quantifies the absolute spread of polymer chain lengths or molecular weights around the mean. It is critical for predicting physical properties like viscosity, tensile strength, and melting point, providing a more precise measure of chain variation than the standard Polydispersity Index (PDI).
What a high σ tells you - more examples
- Plastics testing: Tensile strength of injection molded Polyproyplene (PP) specimen measured under the same conditions. A high σ indicates an incosnistent process with greater part variability and higher risk of outliers (Figure 1).
- Molding inconsistencies: Fluctuations in barrel temperature or cooling rates.
- Operator variance: Poor grip alignment or extensometer slippage during measurment
Ok, and what I can do to have the standard deviation under control?
Here is a quick Troubleshooting Checklist:
- Sample prep: Are specimens conditioned to eliminate moisture variance.
- Calibration: Verify force cells and displacement transducers strictly meet ISO/ASTM requirements.
- Sample size: Ensure you test at least 5 representative specimens for a statistically sound mean.
Thanks for reading & #findoutaboutplastics
Greetings,
Herwig
Literature:
[1] https://onlinelibrary.wiley.com/doi/book/10.1002/0470100427
[2] https://www.hanser-fachbuch.de/Kunststoffpruefung/978-3-446-48105-3
No comments:
Post a Comment