Practitioners of Ambient Environmental Sampling invariably work at either 16.67 lpm (1 M3/hr) or 38.33 lpm (2.3 M3/hr). All are required, under various National regulations, to calibrate the flow rate through their sampling devices, with some degree of regularity. Specific Brands and types of flow meter are seldom specified. Some countries are very specific about flow meter performance, but not type. This leaves it to the expert practitioner to read the specifications and interpret what they mean, in order to make a selection. In the USA this is simple. The performance of the Flow Meter must be twice as good as the instrument being tested. Simply put, if the sampler has a required flow rate accuracy of 2 % of the 16.67 lpm point, then the flow meter must be good to 1 % of the point. The exception, of course is temperature range. If the instrument has an operating range of -30° C to +55° C, then so must the calibration/verification instrument.
The real problem, particularly in countries with more flexible regulations, is that Flow Meters are often compared, which have different accuracy specifications. In Europe where the flow rate is typically 38.33 lpm, all manner of flow meters are used which have Specification Accuracies of 0.75%, 2%, 2.5%, 3% and 5%. Some devices require corrections for Barometric Pressure and Temperature and some do not. Invariably, Flow Metering devices are compared and do not agree. They do not agree for a variety of reasons which may involve the effect of the meter upon the device under test. The result of the effect usually arises from issues of resistance to flow or induced pulsation.
The most confusing problem with uninformed comparisons is due to mixing Flow Meters with different accuracies. Consider a sampler nominally set for 38.33 lpm which is tested with a 0.75% device and a 2.5% device. If the sampler is truly running at 38.33 lpm, then you will see the following potential result:
This simple table illustrates that a sampler running at a true flow rate of 38.33 lpm, could be tested with a “fine” instrument and show a flow rate as low as 38.04 lpm. It could then be tested with a “coarse” instrument and show a flow rate as high as 39.29 lpm. An unskilled investigator could decide that the fine instrument “failed by 3.18%.”
Setting aside all of the fine Metrological points concerned with the evaluation of Flow measuring devices it is futile to ignore the Accuracy of a device when attempting comparisons.