After last week’s comparative levity, this week, I am discussing something rather more serious.
When it comes to detecting hydrocarbons, we often don’t have a cylinder of target gas available to perform a straight calibration, so we use a surrogate gas and cross calibrate. This is a problem because pellistor’s give relative responses to different flammable gases at different levels. Hence, with a small molecule gas like methane a pellistor is more sensitive and gives a higher reading than a heavy hydrocarbon like kerosene.
In order to compensate, we use a cross calibration factor (a theoretical response figure for the flammable gas in question), also known as a correction factor. In most cases, manufacturers provide the correction factors† for us along with their devices.
Commonly available gases, such as methane and pentane are usually used for pellistor calibration. The correction factor is simply used as a multiplier to get the correct calibration level for the target gas. For example, hydrogen has a cross calibration factor of 1.2† against methane.
So, if using 50%LEL methane as your calibration gas, you are calibrating the detector to 60%LEL (50*1.2) hydrogen, as illustrated below. The device readout should be adjusted, using the manufacturer’s PC application, to 60%LEL accordingly. Likewise, if you were using pentane and wanted to calibrate for hydrogen, the cross calibration factor is 0.6†; so the calibration value would be 30%LEL (50*0.6) hydrogen if using 50%LEL pentane.
Gas level used x Cross calibration value = Target gas level
methane 50%LEL x 1.2 = 60% LEL hydrogen
pentane 50%LEL x 0.6 = 30% LEL hydrogen
Where the correction factor with methane is greater than 1.2, we normally recommend the use of pentane as the calibration gas (unless the methane gas level is around 20-25%LEL). This is because pentane produces approximately a 50% response with respect to methane.
For example the target gas methyl acetate has cross correction factor of 1.8 for methane and 0.9 for Pentane it is better practice to calibrate for pentane where possible, for these reasons.
Using the lower cross correction factor of 0.9† with pentane, the pellistor can be calibrated at a level of 45%LEL which will avoid issues with zero stability, gain errors, as well as risk of inaccurate readings around the normal alarm levels (20/40%) as part of the calibration/testing process.
Note: not all pellistors respond well to some hydrocarbons and advice should be sort from the supplier before considering cross calibration.
‡These correction factors are not suitable for infrared detectors, which we will talk about in a future blog.
†Correction factors can vary from manufacturer to manufacture and are available from the pellistor suppliers as well as the complete detector manufactures; contact your supplier for more details