At about 9:30 in the morning on October 15, 1891, Marcellus Gould, a superintendent at the Amoskeag Mills in Manchester, N.H., went hunting for engineer Samuel Bunker. What was going on with the steam power at the mill, Gould wanted to know? The delivery of power from the great shaft that ran mills 4, 5, 7 and 8 had been erratic all morning.
Bunker replied: “There is nothing that I know of, but I’ll see.” With that he headed for the pit that housed the Amoskeag’s power plant on the ground floor of mill No. 7. Those words were the last he ever uttered to another soul. Moments later the mill would explode.
The Corliss Engine
The Amoskeag Mills Nos. 4, 5, 7 and 8 were powered by a Corliss steam engine. In the 1876 Centennial Exhibition in Pennsylvania, the Corliss had been the star of the show. The exhibition kicked off when U.S. President Ulysses Grant and Brazilian Emperor Pedro II jointly started a massive Corliss built expressly for the event. The engine delivered power to virtually the whole exhibition, using shafts of more than a mile in length.
It was an engineering marvel, and the exhibition made the Corliss engine virtually a household name and cemented its status as the leading industrial engine design in America. The Corliss got its name from Frank Corliss of Providence, R.I., who patented several of its features.
It was only natural that the Amoskeag would order one. Installed in 1883, Amoskeag’s Corliss was the largest in the state, and probably New England. It could deliver 2,000 horsepower to the mills.
At Manchester, the factories were powered through an arrangement whereby water from the Merrimack River could be diverted to turn the shafts that drove the equipment. The river worked in conjunction with the Corliss engine. Lift gates were closed or opened to keep the right proportion of river power flowing so the Corliss didn’t run too fast. And at low water, the engine could run exclusively on fuel.
The belts on the flywheel on the engine could be slackened or tightened to keep the engine from spinning out of control or overheating.
Following Gould’s inquiry, engineer Samuel Bunker descended into the pit in the bottom of mill 7 to see what was amiss. Both Gould and Bunker had seen that the belts were overheating, but the reason wasn’t apparent.
Amoskeag Mill Disaster
Marcellus Gould was a troubleshooting mill manager who had worked at facilities from New Hampshire to Connecticut before the Amoskeag mill disaster. Gould was a one-time state senator in New Hampshire. He had an eye for constructing facilities and a knack for making mills profitable. He worked to generate sales of the latest fabrics while managing the revenue streams from worker housing and keeping wages in check.
After asking Bunker to check out the power plant, Gould stepped into a tunnel that ran between the mills to return to mill 5, where the inconsistency of the power had been slowing the work of carders. He was halfway along the tunnel when the Corliss blew apart.
Seconds before the failure, an assistant engineer had looked through a window into the Corliss engine room and noticed the flywheel turning at an excessive speed. Sparks were flying from the 42-inch-wide main belt around the wheel. The assistant was cutting power to the Corliss when it failed.
In the moments before the Amoskeag mill disaster, the power supply was getting more and more unstable. Carding machines to the south of the power plant were stopping from inadequate power. At the same time, looms operating on the northern side of the Corliss were operating too fast. The looms shut down as they reached dangerous velocity. The shutoff mechanism on one loom failed, and it sped along at a frantic pace that no one ever recreated – either before or after the failure.
The impact of the disaster could be felt throughout the mills. First came what sounded like an enormous explosion as the flywheel on the Corliss flew apart. That was followed by several seconds of cacophony as timbers fell, parts of the mill buildings collapsed and steam jetted out of the failed engine.
Pieces of steel were flung hundreds of feet. One enormous piece of metal debris was shot 200 feet into the air and crashed back down through the roof of the number 7. At neighboring buildings, metal shot through windows, showering workers with glass.
A crew of 11 was at work on the second floor of the number 7 mill. When the flywheel gave way it threw pieces of metal upward. The second floor collapsed, dropping the workers to the ground story amid falling beams and equipment.
Deaths in the Debris
The noise of the Amoskeag mill disaster immediately drew workers running to the number 7 from nearby mills. Many workers had family members and friends working in other parts of the Amoskeag, and the workers were frantic to know who was hurt.
The 1891 Amoskeag mill disaster was not the first for the Corliss. In 1888 William Venable, another engineer, was caught between the flywheel and the pit in which it sat. He was mangled to death. This time when searchers got to the pit that housed the massive, 64-ton flywheel, they found that engineer Bunker’s demise was far quicker. His head had been pulverized, and he died instantly.
Two other workers died in the disaster. Ida Cram suffered numerous injuries and broken bones. A head injury would kill her after she reached Elliot Hospital. Mary Kane, just 21, died from a gruesome injury to her scalp.
Other injuries included cuts to a 12-year-old boy. (Eighteen years later, Lewis Hine would photograph child laborers at the Amoskeag mills.)
In all, eight other workers were treated for broken bones and serious injuries resulting from the Amoskeag mill disaster.
Lemuel Levick told of a miraculous escape from death. He had been working on the second story of No. 7 when the floor gave way. He dropped straight down, and the cupboard he was working next to dropped beside him. As beams and debris rained down, the cupboard provided shelter, forming a cave in the debris that kept him safe.
Outside his safe haven, steam spewed into the air. Levick spotted a mill girl being scalded by the steam and was able to draw her to safety in his cave until rescuers extricated them.
Cause of the Amoskeag Mill Disaster?
No one ever figured out exactly what caused the Corliss engine to fail. The engineers on the scene never had time to assess what was causing the irregular flow of power to the plant – unless Bunker was able to spot the problem before he died.
A coroner’s jury heard testimony about the Amoskeag mill disaster and found no wrongdoing on the part of the mill or the engineers operating the machinery.
A panel of engineers reviewed the incident. They examined the remnants of the 64-ton flywheel. They found it contained irregularities in the metal – gaps left behind from its casting that would have unbalanced it slightly. Since the Corliss had operated for eight years without trouble, they doubted the irregularities could be the cause of the Amoskeag mill disaster, though they did weaken the flywheel making it easier for it to fly apart.
The engineers concluded that the failure was likely caused by the mechanism that connected the belts from the flywheel to the shaft that delivered power to the mill buildings. If that mechanism failed gradually, it might result in the belts slacking off – resulting in a heat buildup. As more power was applied to speed up the belts, the mechanism could grab once again resulting in a speeding up of the wheel, requiring the men managing the engine to slow it down.
The result would be a cycle of increasing and decreasing power from the shafts, which was noted on the factory floor. The shaft delivered power to buildings both to the north and the south, with separate belts on both sides. With the mechanism failing on only one side, the result could explain the decreasing speed of machinery to the south and increasing power to the north.
As the mechanism continued to wear, it could finally lock up with greater force and add so much pressure to the belts that something would have to give. In this case, it was the enormous flywheel, which shattered.
Lacking full evidence, the engineers concluded this was the most likely cause of the Amoskeag mill disaster of 1891.
This story about the Amoskeag mill disaster was updated in 2017.