You are correct in that according to witnesses the K-Mart fire originated in the area where the carburetor cleaner was located. That the cause of the incident was reported to be ignition of those vapors, after being released from the container, by an electric forklift that was parked in the isle next to the cleaner. The failure mechanism releasing the vapors was a falling container struck an object that caused a failure in the container. There growth of the fire was contributed to a number of different items, on of which was the amount of product all in one location. No witness ever said, nor was it ever determined that the aerosol rocketing contained the cleaner.

In discussing the use of LPG as a propellant, what is the average amount of this gas in an aerosol container? Most people do not know that much of the product in aerosols is being dispensed through the use of carbon dioxide and not LPG. In these cases the propellant is no longer a potential hazard. Based on your comment if there is no LPG vapor there is no path for the flames to follow to the displaced liquid.

I agree fugitive products that can be called an accelerant can be a problem on any fire scene. What the investigator must do is evaluate all products in the area, and working with other experts reach a conclusion as to the viability of this material coming from another source during the incident or was placed there intentionally to accelerate the fire’s growth.

In one area I must agree with you that our ideas are different. I though the idea behind 921 and other research was to clarify information being used by investigators in reaching their conclusions. In this case, rocketing aerosols, we have 10 years of research conducted by Factory Mutual. In all the burn scenarios conducted during this period that never had a rocket aerosol set a secondary fire. I will agree with Dan in that this in itself is not proof it will never take place, but it does give one food for though as to the possibilities of it taking place. So what I am asking for some one to produce is a rebuttal to the Factory Mutual test. Presently, I have not been able to find this type of documentation. If someone can show me flaws in the research conducted by Factory Mutual, I am open to reconsideration of the facts.

Here you say, “the burden properly lies with the investigator to conduct any research necessary to positively eliminate spray can effects before declaring multiple origins.” I say the burden lies with the investigator to prove it can take place before declaring this is what happened during the fire.

To so extent I believe we both are correct. You stated, “The problem, of course, is that if an investigator has already summarily dismissed aerosol cans in some previous investigation, it may be difficult to persuade him that he may have been to hasty.” Could you think of this in reverse? I would say, “The problem, of course, is that if an investigator has already summarily concluded rocketing aerosol cans caused secondary fires in some previous investigation, it may be difficult to persuade him that he may have been to hasty in reaching this conclusion before establishing supporting evidence for the possibility of this event taking place.”