Fair question. How would you respond to such a query, if pressed by an employee? In fact, this sort of thing did happen, back in the day.
In the early days of instrumentation-based toxic gas detection, the steel industry was a major part of the business, and carbon monoxide was the biggie. Large amounts of CO are produced by blast furnaces, converters, and coke ovens. While certain old-school instruments were installed, based on the use of Hopcalite® catalysts for example, much work was done with detector tubes.
Often, such work involved employees pulling detector tube samples, every few minutes, for their entire shift.
To prove our mettle, Interscan CO monitors were installed close to these employees, and to demonstrate that they were working, CO calibration gas would be plumbed into the units or a planned basis, thus causing a faux alarm. As you can imagine, considerable effort was put into notifying nearby workers that this “alarm” was really an alarm-test exercise. After several weeks of testing, the union was satisfied that our instruments would do do the job.
So, the easiest way to prove that an instrument works is to challenge it with a known gas standard, making sure that everyone knows that this is an alarm-test exercise. But, what do you do when you don’t have a known gas standard on hand? After all, our SENSOR EXPRESS® program was created to free customers from the burden of performing gas calibration. The program is essential for those compounds such as hydrogen peroxide and peracetic acid for which no commercial calibration standard even exists.
Here are two suggestions:
1. Appeal to authority—If you are on the SENSOR EXPRESS® program, to the greatest extent possible, our calibration data is NIST-traceable. However, NIST does not have calibration methods for all gases. As such, our methods for hydrogen peroxide and peracetic acid were developed in-house, and ultimately refer to long established and traceable wet chemistry procedures, and traceable fluid-handling components.
Frankly, though, this traceability covers only the sensor, and cannot guarantee that the rest of the instrument is performing correctly. In general, a properly functioning span and zero control, along with a properly functioning alarm setting functionality—all of which can easily be tested by the end user—serve to indicate proper instrument performance.
2. CAREFUL scheduled release of the target compound—This should only be done using a very small quantity of the substance, in close proximity to the instrument’s inlet port. Of course, employees in the area should be warned in advance of such testing, and data logging records must be scrubbed of this “false” alarm. A good technique is to employ a cotton tipped applicator, dipped into the compound in question.
That is, pour some of the substance into a small beaker, and wet the applicator. Hold the applicator around 12 inches (30 cm) from the instrument’s inlet. But be wary! The concentration on the applicator could be orders of magnitude higher than your typical monitoring concentration. Be prepared to withdraw the applicator immediately if the meter rises rapidly.
Contact us for additional suggestions.