In our previous blog we learned about ways to mitigate particle losses in transport tubing from the sampling inlet near a critical point in a process, to the particle counter. Here, we’ll learn about the ways in which some losses are there to stay.
Acceptable Particle Loss
This section is the most difficult to define, as any loss can be deemed to be unacceptable. However, let us look at the uncertainties, the acceptable errors for particle counting.
The below table shows a comparison of various standards associated with airborne particle counting. Although not fully adopted, they outline the errors and how error stack-up can very quickly show particle counters to be quite different from each other.
From the table, we have a spread of errors from 35% to 81% making an assumed average error of approximately 50%. If we also add to this an acceptable loss of an isokinetic sampling of 5% for counts, then the errors for particle transportation tends to be relatively small.
In qualifying the particle losses of any system, whether it be 28.3 LPM (1 CFM) or a manifold based system, the ability to systematic errors is very easy and biasing of data is dependent on probe orientation, distance apart, air flow patterns, number of particles being sampled, duration of test, base line variables between counters, optical variances, etc. Therefore, defining an absolute allowance is difficult.
| Error analysis for various counting standards | |||
| JIS B9921:1997 | ISO-13323-2 | ASTM f328-98 | |
| Particle Size Accuracy | 5% | ||
| Particle CV | 5% | ||
| Sizing Accuracy | 5% | 10% | |
| Counting Efficiency | 20% | 20% | 10% |
| Sample Volume | 10% | 10% | 5% |
| Resolution | 10% | 10% | |
| Coincidence Level | 5% | 10% | |
| Flow Rate Accuracy | 5% | 5% | 10% |
| Time Accuracy | 1% | 1% | |
| Volume Accuracy | 5% | ||
| Total Error % | +/- 46% | +/- 81% | +/- 35% |
Taking a personal rule of thumb of ensuring that any sampling errors are at least 50% of the maximum permitted counting error (50%), then any sample that tests an environment at least 25% efficiency of transportation loss the total error should not be affected significantly.
Therefore, providing our estimated particle counting losses qualification exercise does exceed 25% loss; equivalent to 2.0 m for a 28.3 LPM particle counter (Lasair®III, Airnet® II 510) or approximately 7.0 m for a manifold device (Aerosol Manifold II-16) then acceptable values for 5.0 μm particles can be determined.
Where do I go from here?
In this blog series from Particle Measuring Systems (PMS) , we learned about what kinds of physical forces act on particles, how to mitigate them, and what sorts of particle losses are acceptable. With this knowledge, you can understand your particle monitoring data better, and how to ensure you’re counting accurately.
Ultimately, tubing should be avoided where possible. When needed, tubing length should be kept as short as is reasonable.
Still want additional guidance on these topics? We’ve got experience on our side. Consider asking our industry experts for help if you’re unsure how to streamline your specific application process. Ask an expert.
