Six Hundred Variations

When the fire alarm went up in Baltimore on the morning of Sunday 7 February 1904, the Baltimore and Ohio Railroad put a train together and ran the Washington fire brigade up the line in thirty one minutes, which the railway believed to be the fastest run it had ever made.

Chief Belt and his men got off at Camden Station, laid out their hose, and found that the couplings would not screw onto Baltimore's hydrants. The threads were a different pitch.

Why the threads were different

This was not an oversight. The first fire hydrant was designed in 1817 by George Smith, and every design that followed, including the threads on its hose connections, was patented by the company that made it. The dimensions of the connection were not an engineering detail. They were the mechanism by which a manufacturer kept a city's fire service tied to its own hoses, its own spares and its own fittings.

A city that bought from one supplier ended up with a water system that a neighbouring city's equipment could not use. A survey carried out at the National Bureau of Standards by Albert Merrill, a few months before all this, found roughly 600 sizes and variations of fire hose coupling in use across the country.

The National Fire Protection Association and the National Board of Fire Underwriters had already argued for a national thread standard before the Baltimore fire. It received little support.

Thirty hours

The fire is believed to have started with a cigar or a cigarette falling into the basement of the John Hurst and Company building. A cold front was moving in, and it burned for thirty hours behind winds gusting to something over forty miles an hour.

Companies came from Altoona, Annapolis, Chester, Harrisburg, New York, Philadelphia, Wilmington and York, as well as Washington. By the end there were 1,231 firefighters at work with 57 engines, nine trucks, two hose companies, a fireboat and a police boat.

The fire took 1,526 buildings across seventy city blocks and 2,500 businesses. One person died: James McGlennen, a firefighter, a few days afterwards, of injuries he took fighting it.

What the story usually leaves out

The version everybody knows is that the out of town crews stood useless on the pavement while the city burned. The government's own account is more careful than that. Some of the responding companies' hoses fitted Baltimore's hydrants and some did not. The National Institute of Standards and Technology, writing the centenary report, says the lack of uniform threads is commonly cited as a factor in the destruction, which is not the same as saying it was the cause. There was also a gale, a wooden-floored business district and thirty hours.

What is not in doubt is that a significant part of the mutual aid that arrived by train could not be connected to the water.

The bureau that could not couple its own hoses

The point was made again, almost immediately, in the most embarrassing way available.

In late 1904 a National Bureau of Standards employee named Franklin Durston was dealing with a small fire on the bureau's own premises. He discovered that the hoses in the bureau's North Building and its South Building had different threads and could not be coupled to each other. The national standards body could not join up its own two hoses.

The standard

In 1905 a committee of the National Fire Protection Association settled the question. A fire hydrant would carry hose connection nozzles of 2.5 inches with seven and a half threads to the inch, and a pumper connection of 4.5 inches with four threads to the inch. It was adopted by the NFPA, the National Board of Fire Underwriters, the Associated Factory Mutual Fire Insurance Companies, the International Association of Fire Chiefs, the American Water Works Association and others.

Those numbers are still the standard. That part of the story is a success.

The part that is not

Adoption was another matter, and the figures are worth reading slowly. By 1914, nine years after every relevant body in American firefighting had agreed on it, 287 of the 8,000 cities and towns in the United States had conforming couplings and hydrant outlets. By 1917, 897 cities had agreed to adopt the standard and 390 were actually using it. By 1924 the number in use had reached 700.

Then the report says something in a flat, administrative voice that is worse than any adjective. Some of the cities made the change only after they experienced their own major fire.

Oakland, 1991

On 20 October 1991 fire came down into the Oakland hills on a dry east wind. Engine companies were called in from the surrounding departments immediately, and among the problems their commanders ran into was that engines carrying the standard two and a half inch couplings could not connect to the three inch couplings on Oakland's hydrants.

Twenty five people died, including a police officer and a firefighter. Around 150 were injured. Some 2,450 houses and 437 flats were destroyed, at a cost of roughly 1.5 billion dollars.

After the fire, Oakland replaced all of its fire hydrants with hydrants built to the national standard.

Where it stands

In 2004, a hundred years after Baltimore, two NIST researchers went looking for the current position and found that no central record of American hydrant specifications exists. They had to assemble it from fire departments, fire marshals, local authorities and, in the end, mostly from the manufacturers.

Of the 48 most populated cities in the United States, 18 had hydrants meeting the national standard on both the hose connection and the pumper connection. About 40 had at least the hose connection right, which means an out of town engine can usually get water but its pump may not be able to take a supply line.

The thread was settled in 1905 by unanimous agreement among people who had just watched a city burn. Most of the largest cities in the country still have not finished fitting it.

Sources

• Momar D. Seck and David D. Evans, National Institute of Standards and Technology, "Major U.S. Cities Using National Standard Fire Hydrants, One Century After the Great Baltimore Fire", NISTIR 7158, August 2004 (the patenting of hydrant designs from George Smith's of 1817 and the use of connection dimensions to protect manufacturers from competition; Albert Merrill's survey finding about 600 sizes and variations of coupling; the earlier advocacy of a standard by the NFPA and the National Board of Fire Underwriters receiving little support; the fire starting on 7 February 1904 from a cigar or cigarette in the John Hurst and Company basement and burning for thirty hours; the companies arriving from Washington, Altoona, Annapolis, Chester, Harrisburg, New York, Philadelphia, Wilmington and York, some of whose hoses fitted and some of which did not; the 1,231 firefighters, 57 engines, nine trucks, two hose companies, fireboat and police boat; the 1,526 buildings across seventy blocks, the 2,500 businesses, and the death of firefighter James McGlennen; Franklin Durston's discovery that the bureau's own North and South Building hoses could not be coupled; the 1905 NFPA standard of 2.5 inch hose nozzles at seven and a half threads per inch and 4.5 inch pumper connections at four, and the bodies that adopted it; the adoption figures of 287 conforming towns by 1914, 897 agreeing and 390 using by 1917 and 700 by 1924, and the observation that some cities changed only after their own major fire; the Oakland hills fire of 20 October 1991, the two and a half inch couplings that would not fit Oakland's three inch hydrants, the 25 deaths including a police officer and a firefighter, the roughly 150 injured, the 2,450 houses and 437 flats destroyed at about 1.5 billion dollars, and the replacement of the city's hydrants afterwards; the absence of any central record of hydrant specifications; and the finding that 18 of the 48 most populated cities have both connections to standard while about 40 have at least the hose connection).
• National Institute of Standards and Technology, "History of fire research at NIST" (NIST's involvement in fire safety beginning with a 1904 complaint about incompatible shipboard hose couplings; the Baltimore fire highlighting the problem; and the nationwide survey of coupling threads on which the 1905 NFPA standard was based).
• IEEE Industry Standards and Technology Organization, "Standards, Fire Hydrants, and the Great Baltimore Fire" (the alarm at 10.45 in the morning on 7 February 1904 and the thirty hour duration; the cold front and the sustained winds of ten to twenty knots gusting to thirty six; the roughly 600 variations in coupling and hydrant outlets and the note that differences in diameter and thread protected incumbents against competition; and the slow adoption of the 1905 standard).
• Library of Congress, "Baltimore fire, 1904: Chief Belt, Marshall Bieber and D.C. firemen who went to Balto" (the photograph of the Washington company, from the National Photo Company Collection).