PGI Bulletin No. 45, January, 1985
Back Issues are available from www.pgi.org

THE ANATOMY OF A PYROTECHNICALLY CAUSED FATALITY
by
Bob Winokur

On Thursday September 27th 1984 I received a call from Sparky Bement. He explained that he was shooting a show for Zambelli at Caesar's Palace on Friday, would be in town for several days and would like to get together with me during his stay in Las Vegas. I had met Sparky briefly during the 1981 PG1 convention. On Friday morning Sparky called again and mentioned that he needed some extra smoke generators. These were to be used in a simulated volcano which was being prepared for Friday's large private party which had a Hawaiian theme. I explained that I had some class ''C'' smoke devices which would suit his purposes nicely, and we agreed to meet at 4:30 that afternoon. Sparky said that he would have the aerial portion of the show well prepared by then and would easily have time to meet me.

The following description and explanation of what happened is based on the following sources of information: 1) first hand observation of the mortar rack and other physical objects kept in evidence by the Clark County Fire Department, 2) the official written report generated by the fire department and 3) extensive conversations with Sparky's relatives and fire department officials.

At about 1:45 PM on that Friday (September 28th) Sparky had already unloaded a large number of wooden mortar racks from a truck onto the pavement of one of Caesar's large parking lots and was in the process of loading 3'' shells when the accident occurred. A 3'' shell left a mortar and struck Sparky in the forehead crushing his frontal bone and driving fragments of bone into his brain. He was taken by helicopter to a local hospital and underwent emergency surgery that night. His condition was very grave, but they managed to stabilize him, and he lived for several weeks. He never regained consciousness. The display was cancelled and the fire department undertook an investigation of the events surrounding the accident.

In my opinion the most important thing to be discovered is exactly what caused the ignition. The mortar rack contained ten mortars and looked as if it had not been previously used. Pieces of 1/4*'' wide red tape remained on the sides of the mortars. The tape had been positioned so it would be broken when a shell lifted from the gun. This is a common safety technique which allows a pyrotechnician to identify a mortar containing a live shell after a display. Although electrical squibs were to be used in the display, none had yet been installed. No squibs or evidence of squibs were found in association with any of the racks and it was concluded that the ignition did not have an electrical origin. There is no evidence to support rumors that a staple struck a squib and caused an ignition. Squibs were found packed in the truck but none had apparently been brought out. A Hanson Tacker staple gun was found on the parking lot pavement in several pieces. It apparently had fallen apart as a result of being flipped from Sparky's hand when he was struck by the shell. The entire rack had only two staples protruding from the wood. These two staples were not near each other being adjacent to different mortars. One of them held a small piece of burned match piping. There were no holes or other indications that there were other staples near the two staples which had held the match. There were two large bent nails protruding from the top of the rack, but these also were not close to the staples. The rack itself was well-made and held together by large industrial type staples. There were no staples, large or small, or other metal objects anywhere near the two staples which held the match piping. I looked very carefully at this because of the presence of other metal near staples is always hypothesized as a source of sparks.

Fortunately the Clark County Fire Department collected samples of unburned quickmatch from one of the nearby racks and included it with other materials kept in evidence. I was given a piece of this match which appeared ordinary and probably of Japanese origin. One of my first thoughts was that the match might be made of a potassium chlorate or perchlorate composition which would make it more easily ignited by mechanical action. Using some simple chemical spot tests {Aniline hydrochloride in hydrochloric acid for chlorate, sulfuric acid for chlorate and barium salts, and methylene blue for perchlorate.} I determined that the match did not contain any potassium chlorate or perchlorate or any barium compounds. The tests I used are very sensitive but might not, of course, detect the very smallest trace amounts of these substances. Nevertheless, for reasons given below, it seems very unlikely that the ignition can logically be attributed to the presence of these substances.

With fire department staff members assistance, a severe test was performed on about a one foot length of the match. The piece of match was placed flat on a concrete sidewalk and struck repeatedly with a large claw hammer. This treatment was continued until the sample of match had all but disappeared as tiny fragments. We shrugged and wondered. I took some of the match home and later that day using a somewhat smaller claw hammer tested several two inch long pieces as before. The first piece behaved like the one tested earlier but the second piece produced a snapping sound. Had I heard a tiny explosion? I wasn't convinced. I tested still another piece. Yes -- I heard it again! I smelled the hammer and the familiar acrid odor of sulfur involved in an explosion was subtle but clearly present. The smell is similar to that produced by the caps used in cap pistols, only without any of the phosphorous content. I tried again and again with one inch pieces and repeatedly achieved small distinct explosions. Later, using both American and Japanese match I not only obtained snapping sounds but complete ignition of the match. There was no question in my mind that quickmatch could be ignited by mechanical means.

But Sparky wasn't using a hammer on cement. The staple from the Hanson Tacker (a rather small long staple -- ½'' long and 1/8'' wide) was driven into wood not concrete. Could such a staple being driven into match ignite it? The answer seems to me to be clearly yes it could. Several aspects come to mind. Match is commonly manufactured out-of-doors where it is easily contaminated by sand or grit. A small grain of siliceous material might cause a spark when struck by a metal staple. Perhaps the staple gun itself caused a spark which somehow followed the staple through the match piping and caused the ignition. In fact is is unnecessary to hypothesize about sand in the match or staple gun sparks since the tiny chisel pointed staples (which I closely examined) would result in the full impact of the staple gun's powerful spring being brought down onto a very tiny surface and thereby producing enormous force on the match composition. Pyrotechnists have come to think of potassium nitrate compositions such as black powder as ''safe'' and not ignited by friction or impact. (After all, aren't fountains, gerbs and rockets routinely rammed?) Shimizu in his Fireworks, The Art, Science and Technique (1981), page 319, indicates that various mixtures of only potassium nitrate and sulfur are quite capable of being ignited from impact in a drop test. Ignition or explosion sometimes occurred if a two kilogram weight fell more than 38 centimeters (a little over 15 inches) onto the mixture. Shimizu in discussing black powder on page 174 states: ''Attention must be paid to the fact that it can ignite or explode from strong mechanical action.'' Davis, (The Chemistry of Powder and Explosives, 1943, page 21) indicates that black powder can be caused to ignite with a two kilogram drop test of 70 to 80 centimeters (27 1/2 to 31 1/2 inches).

Many experienced pyrotechnists know that staple guns can and have caused ignitions when used to attach match to lances. But in that case we have a staple striking a gunpowder type prime and continuing on into a lance composition containing potassium chlorate or potassium perchlorate. That such ignitions might occur is not surprising, but stapling match into wood with no chlorate or perchlorate present is hardly the same thing. Friends and correspondents of mine assured me that it was very unlikely that match would ignite from such treatment. I then re-thought about my first experience of hammering on pieces of match until they were destroyed without achieving an ignition. The real significance of the experimental hammering on match is not that an ignition can be achieved (the literature cited above clearly shows this should be possible), it is that achieving ignition of blackpowder with mechanical action is a relatively rare event -- conditions apparently must be just right and the chances of causing an ignition is a matter of probabilities.

The evidence from the racks at the scene of the accident suggests that Sparky was not using a staple gun to hold down match as a routine method. People close to him attested to his speed and skill in using twine to match together finales. The fact that there were only two staples in the rack suggests that he was using staples only as an accessory technique, perhaps because of some now difficult to appreciate motivation (possibly a special arrangement of finale racks). The important point here is that even if this was the very first time Sparky ever stapled match to a rack, it was one time to many. Statistical events do not occur as if there is some score keeper keeping track of how many times a dangerous technique is used. Each staple flying at high speed into match composition is an independent event, and the probability of it igniting the first time a person does it is the same as the ten thousandth time. The lesson all pyrotechnists must learn is clear. Treat black powder compositions with care. Do not rain any compositions (even such seemingly well understood items as spolette fuses) in your workshop in the presence of quanti ties of pyrotechnic compositions, do not use staple guns, metal hammers or other mechanical devices which put compositions under great mechanical forces, and never place any part of your body over a loaded mortar. There is also a lesson to be learned here for fire departments and other public authorities which become involved in investigating pyrotechnic accidents. The Clark Count Fire Department did an excellent and thorough job in collecting evidence and preserving materials useful in coming to an understanding of what happened. Without their meticulous investigation, people would continue to claim, as some did at the outset, that the match was contaminated with chlorate or perchlorate or that the staple struck a squib or a nail which produced a spark. It seems to me that while it is still not impossible that these or other causes could be at fault, it is less likely then the simple and scientifically acknowledged fact that black powder can and will be ignited from mechanical action. 

Every pyrotechnically caused fire or explosion in which there is a death, serious injury or significant property loss should be investigated with the same thoroughness as that of a murder. Only with such data will we be able to regard our endeavor as one in which the risks are under control. Hopefully, we will have learned from Sparky's tragic death. If we haven't, we are doomed to repeat this kind of accident and if this seems to be a cliché or an exaggeration -- think about it again.

Follow-on emails from Kurt Medlin (personal addresses and information edited out):


Dan, 


... you can quote me directly if you like regarding the alternate explanation on staple gun danger, but the caveat is I'm going by memory. There was a lot of discussion about the use of staple guns in the few years after Bob's article (leading to things like the NFPA restriction on their use), but that was 20 years ago, so I don't quite remember all the details. 
 
Kurt 
 
Dan, 

You have my permission to reprint with attribution to Bob and the PGI, but I'm copying Bob for two reasons (Bob, this is regarding the article on Sparky Bement's fatal accident): First, as a courtesy to Bob on the off chance he may object;  Second, Bob may want to comment further. I believe subsequent to that article, the conventional wisdom regarding the probable cause of ignitions from staple guns changed from the impact energy, to the fact that staple guns can produce a spark when the staple is hit with the internal metal striker that drives the staple.  Anecdotally, I have heard that you can observe a small spark emanating from some models by firing them in a dark room; I haven't tried that with any of my staple guns, so I can't confirm. 


 I've also heard that there are special staple guns (or was it staples?) designed to not produce sparks in this manner for use in industrial applications around flammable vapors or some such. Either way the messages are the same - 1) don't use staple guns (!) and BP comps can be made to ignite from impact - I have witnessed first hand two such ignitions at Winter Blasts, both involving hand ramming (one involving a wheel driver and one involving a 1-pound BP rocket). Yours, Kurt

Kurt