Re: Sprinkler System Blamed for Death
One person's thoughts::::
"""""""""""In order to duplicate or test this hypothesis in an appropriate and valid manner, one needs to understand the hypothesis from a more fundamental level.
In order to cause a deflagration or explosion, there needs to be a fuel in the vapor phase that can mix with air to form a vapor mixture within its flammable limits. For this hypothesis, the fuel is identified as glycerin.
The glycerin in this case is initially in the liquid phase, thus, it must change from the liquid phase to the vapor phase.
Atomization can be employed to allow phase change from a liquid to individual droplets that, if small enough can behave as a vapor with respect to combustion of the droplets. Droplet sizes of less than 10 microns tend to behave as a vapor while droplet sizes of greater than 40 microns tend to behave like a spray where a flame surrounds each individual droplet. Droplet sizes associated with fire sprinklers tend to produce droplet distributions greater than 100 microns, so atomization alone of the glycerine may not be responsible for producing a phase change in the glycerine.
Another mechanism for a phase change from liquid to vapor is evaporation associated with the addition of heat. Vaporization of liquid fuels can involve both atomization and heating to produce a phase change from liquid to vapor. In this case, the introduction of the glycerine through a fire sprinkler and the transport of the droplets through a hot upper layer, ceiling jet, or thermal plume may induce a change in phase. This would then be a heat transfer and momentum problem to determine how much vapor could be produced. The amount of vapor produced would also be a function of the concentration and volume of mixture in the system.
In addition, the hypothesis seems to be that the glycerine is mixed with water, thus, it needs to be separated from the water to be an efficient fuel. Thus, the glycerine would need to be dehydrated to separate the liquid glycerine from the liquid water. This could potentially happen since the boiling point of water (100 degrees C) is lower than the boiling point of glycerine (290 degrees C). Again, the rate of heat transfer would determine the amount of vapor produced.
Once the liquid glycerine is separated from the liquid water, it must be further heated to change to the vapor phase. A competing factor to the boiling point is that the latent heat of evaporation is orders of magnitude higher for water than for glycerine. Thus, the water can absorb more energy before converting all of the liquid to a vapor. That is one of the most significant reasons why water is a good fire suppression agent.
The question to be asked is: what is the heat source that could produce such a phase change and is it sufficient to produce vapors of sufficient concentration to cause the damage that was experienced in this explosion/fire.
As the concentration of glycerine increases in the water, the conditions can become more favorable, but still may not be sufficient to case the explosion. A concentration of 100% glycerine might be the most advantageous condition, but most anti-freeze solutions for fire sprinkler systems are purchased in a pre-mixed formulation and pure glycerine has a much higher viscosity when compared to water, thus, it does not flow well as a liquid. Therefore, it unlikely that a 100% concentration of glycerine was added to the sprinkler system.
My initial "engineering intuition" does not provide me with a "warm and fuzzy" that this can happen especially since it does not seem to be a common occurrence. The argument could be that it takes a special set of currumstances to occur that infrequently occur. That is why the right hypothesis testing becomes very important.
While I believe I could construct sufficient conditions to produce such a vapor concentration, I am not sure I could produce it if constrained by the conditions at the time of this specific incident. This is why it is important to test this hypothesis with regard for the facts of the case as best that can be determined. Otherwise, one runs the risk of producing unreliable determinations.
At any rate, this is an interesting discussion and I appreciate the opportunity to try and contribute to the discussion.
Sincerely,
Doug Carpenter
Douglas J. Carpenter, MScFPE, CFEI, PE
Vice President & Principal Engineer
Combustion Science & Engineering, Inc.
8940 Old Annapolis Road, Suite L
Columbia, MD 21045
(410) 884-3266
(410) 884-3267 (fax)
www.csefire.com """"""""""""""
