Calibrating a water quality sensor is a necessary step in creating measurement results that meet the highest standards of accuracy. Water quality instruments that allow user calibration are common. Because calibrations impact the quality of the data retrieved from the instrument it is important to understand the sources of error that may contribute to calibration uncertainty and make procedural adjustments to minimize uncertainty in a calibration result.
One way to frame and visualize the variables that impact the calibration result is a cause and effect diagram, also known as a fishbone diagram (Figure 1). Such a diagram categorizes causes into groups with similarities. A popular way of categorizing causes is by the 6 Ms: machine, method, material, man power, measurement, and Mother Nature. With some modification the 6Ms become a useful set of categories for sources of calibration uncertainty. These are: User, How, Sonde, Weather, Standards, and Location.
The manpower category captures attributes of the user doing the calibration. Experience, training, and temperament are important. The more experience a user has with water quality instrumentation the more likely the user will perform the procedures satisfactorily. However, experience alone is not adequate unless the user has been through training in how to do proper calibrations. Patience and diligence are important elements of the user’s temperament.
The method category captures how the user does the calibration. The user may be following a standard operating procedure (SOP), a supplier method that is recommended by the instrument supplier, or a method dictated by a regulation. Sometimes these are the same, but not always. There may be good reasons to do one over another. In any case, follow the method consistently and document the method that was used.
The machine category captures attributes of the sonde, sensor, or instrument used to collect data. Maintenance is required to make sure the sonde is working the best way. Perform maintenance as needed and as the instrument manufacturer recommends. Instruments that are maintained will stay in good condition longer. However, instruments will eventually wear out and a replacement might be required.
Mother Nature captures the elements of the weather that can impact calibration results. Generally it is best to calibrate indoors where weather will have a limited effect on sensor calibration. However, sometimes an SOP or the type of measurement (example: dissolved oxygen % saturation) require field calibration. Precipitation can contaminate or dilute calibration standards if care is not taken to prevent this. Extreme or varying temperature can take the sensors and calibration standards out of thermal equilibrium. Sunlight can create interference with optical sensors or create unwanted temperature gradients in the calibration standards due to radiative heating.
Material required to do a successful calibration includes the calibration standards used. The chemistry of a calibration standard may vary from source to source. For example, some conductivity standards are made from NaCl and some are made from KCl. Each has slightly different responses to changes in temperature. As another example, formazin and polymer bead turbidity standards have different optical properties and will generate different readings from one turbidity sensor to another. Many calibration standards change properties as they age, so be aware of the expiration date of the standard. Handling may impact a calibration standard too. For example, pH 10 buffer will absorb carbon dioxide from the atmosphere and become less basic. Therefore it is important to keep the pH 10 buffer container closed to minimize the exposure to carbon dioxide from the atmosphere.
When circumstances allow a choice for the location of calibration it is normally better to calibrate indoors in an area dedicated for calibration, such as a laboratory bench. The stability of the environment is normally much more conducive to good calibrations because temperature, humidity, and lighting are more controlled than they would be outside. The comfort of being inside is more likely to encourage the necessary behaviors for doing a precise calibration. And in a dedicated area there is easier access to the tools, materials, and equipment needed to do a successful calibration.
Figure 2 summarizes the factors explained above that impact sensor calibration. When developing a monitoring program consider these factors when developing the processes and methods for calibration.
Minimizing uncertainty is key in the calibration process. The Hydrolab Operating Software available on the Hydrolab HL4 Multiparameter Sonde has built in features to ensure proper calibration practices such as guided and semi-automated calibration routines and options to validate calibration.
For more information about minimizing uncertainty in calibration of Hydrolab multiparameter water quality instruments, visit www.ott.com or contact the OTT Hydromet Technical Support Team at firstname.lastname@example.org.