An earlier article presented information about the magnitude of firework mortar recoil forces. As a starting point, that article used the averages of 136 measurements of the peak internal mortar pressure produced upon firing 3- to 12-inch spherical aerial shells. These peak pressures were then converted to recoil forces using basic principles of physics. However, while the average peak recoil force results are thought to be highly reliable for typical shells, those results were not confirmed by physical measurements of peak recoil forces.

During the 1990s, the Canadian Explosives Research Laboratory (CERL) conducted a large number of measurements of firework mortar recoil forces. Some of the results from the testing of 205-, 255- and 305-millimeter (8-, 10-, and 12-inch) aerial shells have been published. Approximately half of these test shells were produced in the Orient and half were from Europe. The purpose of this short article is to compare the CERL results with the results the earlier article and to draw some conclusions from those comparisons.

This article reports on another batch of comets, imported from China, that tend to malfunction by exploding violently within their mortars as they are being fired. In the last few years, the authors have reported on imported Chinese comets from two other manufacturers, with somewhat similar construction characteristics, that also tended to explode violently upon their firing.[2-6] In each case, it was found that the malfunctioning white (or silver) comets had a high percentage of very fine magnalium (magnesium-aluminum alloy), used potassium perchlorate as the oxidizer, and had internal voids that could function as fire paths.

The comets in question for this article are of two different types: 3-inch solid cylindrical comets, and 3-inch tiger tail comets made by layering comet composition over an internal aerial shell.[1] While some cartons of both types of comets tended to experience the same explosive malfunctions, only the cylindrical comets were carefully studied and will be discussed in this article.

In the 2000 edition of the National Fire Protection Association's Code for Fireworks Display, a chapter was added with requirements for performing displays on barges and floating platforms. One of those requirements for manned displays when fired electrically (or as an alternative fired manually) is to provide a safety shelter for the on-board crew. (In the 2006 edition of the code, a safety shelter is also required for roof-top and limited-access displays, unless the electrical firing crew is located at least 75 feet from the nearest mortar.)

One requirement for the construction of the safety shelter is that it be made using at least the equivalent of 3/4-inch plywood. While it might reasonably be concluded that some number of standard dimensional 2x4's would also be used in the construction that is not specified in the NFPA code. One requirement for the placement of the safety shelter is that it be placed at least 2 feet per inch of mortar diameter away from mortars up to 6-inch in diameter and at least 4 feet per inch of mortar diameter away from mortars larger than 6-inch diameter.

In drafting the 2000 version of the code it was widely accepted by the NFPA's Technical Committee on Pyrotechnics that a safety shelter was a good idea and that the construction and placement requirements were reasonable. However, at that time no one had any test data to offer in support of the concept of a safety shelter, its manner of construction, or its positioning with respect to the mortars in the display. Thus, it seemed worthwhile to the authors to conduct some testing of safety shelters. (Unfortunately, it has taken several years for the authors to find the time to begin to work on this project.)

Probably the most extreme test of NFPA’s required barge safety shelter construction is the direct impact of an aerial shell fired from the minimum distance allowed. To begin this investigation, last fall some initial testing was conducted and the results were reported. In summary, when a typical 3-inch shell was fired directly at the plywood (i.e., perpendicular to its front surface) from a distance of 8 feet, the 3-inch shell was found to be capable of penetrating ¾-inch AC exterior plywood backed with 2×4 dimensional lumber on 2-foot centers. Following this somewhat surprising result, thought was given to relatively easy and inexpensive ways that the shelter’s resistance to shell penetration might be improved. During further trials, it was eventually found that relatively simple and inexpensive modifications could be made that provided substantially improved penetration resistance, ranging to at least 6-inch shells. Descriptions of many of the trials and their results are presented below.

This investigation was commissioned to evaluate a range of various aluminum metal powders for their potential use in compounding flash powders for use in powerfully exploding devices such as so-called M-80s and devices of similar construction.

Often, standard micro-analytical chemistry performed on pyrotechnic residues cannot be expected to provide sufficient information for an investigation of the cause and course of an accident. On those occasions, PRRP analysis generally provides important information that is not otherwise available. For example, there are times when standard micro-analytical chemistry will fail to discriminate sufficiently between materials of pyrotechnic origin and other unrelated substances also present on the items being sampled. In addition, there are times when PRRP analysis can help identify details concerning the cause and course of events that are simply beyond the capability of micro-analytical chemistry.