By: Stuart Simmonds – VP of Business Development.
In the area of general technical points to remember when considering analytical instrumentation measurements, one key point of note is how the units of measurements are made and why this then becomes an important factor to consider.
Generally speaking, most gas measurements are analysed volumetrically, typically reported as a percentage (%) or parts per million volume (ppmv) concentration in the sample gas. Specifically, this is then a portion of the entire volume of the gas considered as a whole sample, either within the process (as an insitu measurement) or taken out extractively and passed through an analyzer sample cell.
This can be a significant factor in determining the overall precision and/or accuracy of a measurement, and so, the physical properties of the sample gas must be taken into account in order to provide the best analytical readings.
As we mentioned in our previous blog post on wet versus dry measurement, if the volume of the gas as a whole is altered – in that case by the removal of water vapour – then the volumetric reading of the gas of interest is also affected.
Take for example, a measurement of 5% Carbon Dioxide in a hot wet flue gas with 18% Water Vapour content. If this sample gas is then passed through a chiller and the water vapour is removed (assuming 100% removal) then our Carbon Dioxide reading would now be 5% from a total volume of 82% (rather than 100%) and the reading would be 6.1%.
This is a rather extreme, but real-life situation, of concentration variance due to volumetric changes and one that the industry is familiar with and therefore has procedures in place to account for differences. However, small changes in the physical properties of the sample gas can occur regularly during analytical measurements, which going unnoticed can be assumed to be noise (or drift) in the analyzer itself. Let’s look at the three main factors that can affect the analyzer readings.
- Sample Pressure:
An analyzer is typically calibrated and operated at a fixed pressure. Small variations in sample pressure can be resolved with pressure compensation modules (particularly for insitu analyzers) but in most cases the preference is to fix the sample gas pressure for the analytical measurements. Any variation in the pressure of the gas will result in a change in the reading which does not (necessarily) relate to the volumetric change in the target gas in the sample.
Let’s look at an example of 5% Carbon Dioxide in a sample at 1 Atmosphere of pressure. If we were to double the pressure in the analyzer to 2 Atmospheres, the analyzer would now read 10% Carbon Dioxide. We have effectively doubled the volume of gas in the sample cell, doubling the number of molecules of Carbon Dioxide in the given space and hence the analyzer will read double the concentration. The same logic would apply if we lowered the pressure below 1 atmosphere, we would lower the reading.
- Sample Temperature:
Similar logic can be applied to the sample temperature during measurement, but the significance of the effect is less dramatic. As a gas is heated it starts to move with greater energy and the molecules interact and collide with one another with more frequency. This expanding gas in turn wants to occupy a greater volume. Therefore, in a fixed volume sample cell, as the temperature increases the sample gas reading will decrease, as they cannot occupy the same volume.
In our example of Carbon Dioxide, if the 5% measurement were taken at 25 Deg C then the temperature raised to 45 Deg C, we would expect the reading to lower to approximately 4.8%. These figures are hard to calculate as it depends on a number of factors related to the sample gas, but the significant point here is that changing sample temperature will have an effect on the sample reading.
- Sample Flow Rate:
Finally, for best practice we also need to consider the flow rate of the sample gas. A fluctuating flow rate can affect the reading by similarly affecting the volume of gas in the sample cell. The net result is to cause volume variations in the measurement cell which the analyzer perceives as changes in the concentration.
Conclusion
A process analyzer (unlike its laboratory cousin, which is often designed and operated to determine unknown measurements) is designed to perform a specific measurement under a specific set of conditions. The accuracy of the onsite readings relates directly to the factory set up and calibration performed on the analyzer during manufacture. If the conditions for measurement change at site, or are fluctuating at site, then the readings in turn will be affected.
For best practice, and the best analytical results, we need to always consider the physical properties of the sample gas and ensure they are consistent with the original design criteria of the analyzer and then remain consistent during on site operation.
Novatech’s analytical experts are always on hand to resolve any issues related to sample handling and analytical measurements.