Do LEL sensors have a cross interference to CO2?

Do LEL sensors have a cross interference to CO2?

Dave Wagner | Friday, September 30, 2011

A career firefighter and great friend of Industrial Scientific shared a video clip with me this week. 

Firefighters in Phoenix, AZ were called to an incident at a fast food restaurant after a worker had collapsed walking up the stairs from the basement. The firefighters, after almost being overcome themselves, were surprised to see high readings on their combustible gas sensors from what they later learned was carbon dioxide. Clearly not a combustible gas.  So, do LEL sensors really have a cross interference to CO2?

Well, not in the way that we are normally used to seeing cross interference on other gas sensors. Typical LEL sensors are made up of two coils of very fine wire. The resistance of these coils will change as they heat and burn gas. The change in resistance due to the change in temperature produces the signal that we measure as the gas concentration. If the thermal conductivity of the atmosphere changes without the presence of any combustible gas, the resistance of the sensor coils will also change. So if there is a gas present in a significant enough quantity to change the thermal conductivity of the atmosphere from what it is in normal air, the resistance of the sensor elements will change and a combustible gas reading will be displayed. It is important to note that the resulting reading from the combustible sensor in this situation may be either positive or negative.

The firefighters in this situation stated that they observed an oxygen reading of 17.3% on their instrument display. From this reading, you can calculate that the carbon dioxide concentration in this instance was on the order of 17.5%. This concentration is certainly high enough to change the thermal conductivity of the atmosphere and produce a change on the LEL sensor.

In this situation, you should also note that because of the high level of carbon dioxide, the oxygen sensor was probably reading higher than the actual gas concentration due to interference with that sensor. The reality of this particular incident was that the actual oxygen concentration was more likely in the 15-16 percent range, the CO2 concentration was more in the 23-28 percent range and both the oxygen concentration and carbon dioxide concentration were presenting very significant hazards.

Understanding the reactions of different sensors in an incident like this is the key to properly assessing and resolving the situation. Outside of going into the basement without a gas detector in the first place when they suspected a problem, these firefighters did it the right way. The bottom line is, when the gas detector alarm sounds or you sense that something is wrong, GET OUT first and ask questions later.

Keep it safe out there!

Dave