Karl carried a roll of drawings and a sheaf of paper into George’s office. “I just checked that bracket for Eagle Falls. It is way overstressed and will fail,” Karl said.
George was surprised by this report. As was the custom at this consulting company, Karl was verifying the stresses in a steel bracket supporting a 200-mw hydro generator. This was one of two turbine-generating units in the powerhouse. The company routinely checked all critical components of manufacturer’s equipment before commissioning a new unit. Errors were rare, but this check helped guard against failures that would result in a catastrophe during commissioning.
The bracket Karl had checked supported a 420-ton generator rotor, vertical shaft, and Francis turbine runner. The bracket was shaped like a spider, with a central hollow cylinder. Around the cylinder were two rings, which were welded to the cylinder and to the flanges of six I-beam arms extending outward to rest on the powerhouse concrete plinth, just below the generator.
“Karl, you are dealing with a well-known generator manufacturer who has been in business almost since Edison invented the light bulb,” replied George. “You must have a decimal error in your calculations. Check your numbers and come back.”
When Karl returned the next day, he indicated his result was the same: an overstressed bracket. After reviewing the calculations with Karl, George agreed the results were correct. The overstress was at the junction between the central cylinder and the radial arms, where the six arm flanges were welded to the two rings around the cylinder. From a casual look at the bracket, the dimensions appeared to be reasonable, but from experience the steel thickness seemed minimal.
George phoned Ed, the chief electrical engineer at the consulting company, to apprise him of the situation. Ed reacted as George initially had, saying the calculations had to be wrong. George eventually convinced Ed that the manufacturer had to be informed of this situation. Ed gave George ten minutes at an upcoming meeting to discuss this topic with the manufacturer. At that meeting, George made a short presentation to the manufacturer’s electrical engineers, outlining the problem. Unfortunately, these engineers could not understand the details of the situation. They were incredulous but eventually assured George they would pass his concerns on to the engineering department.
Six months went by before George received a response. During that time, work on Eagle Falls was at a standstill. The response was a new drawing from the manufacturer that showed a much reinforced and deeper bracket, with the steel thickness more than doubled in several critical areas. When Karl checked the stresses for the new design, he found them to be satisfactory.
Before proceeding with the new bracket design, George wanted to know what had gone wrong with the original design. At the next meeting, George asked a few pointed questions. One engineer told George that the error had caused consternation in the manufacturer’s engineering department. The company determined a senior technical draftsman who had been with the company for many years had “drawn” all of its generator brackets. The manufacturer had never undertaken stress analysis on this design. Over time, as generators became larger, the same draftsman had simply extrapolated that design to produce a bracket that “looked right.”
As a result of this error, the manufacturer had to engage the services of a structural engineering consultant to review the design of all recently manufactured generator brackets. This task required many months. This consultant redesigned the Eagle Falls bracket, and the manufacturer reached an agreement with the consultant to provide structural services for future units at the manufacturer’s design office. Very fortunately, only one other generator bracket on the units manufactured by this company was found to be slightly overstressed. This situation was corrected by adding reinforcing plates.
At Eagle Falls, the 500-ton rotating weight supported by the bracket was more than twice as heavy as any previous bracket the manufacturer had provided. In addition, the Eagle Falls bracket was about 40 percent larger and the bending moment in the bracket was four times larger than anything built previously. The draftsman simply had not realized the extent of extra steel required.
Where failure of components would result in a significant delay or additional cost to the development, consultants or owners’ engineers should verify the design of all critical components provided by contractors. These components would include crane beams (both on the crane bridge and in the powerhouse superstructure steel), the lower structural beams in gates, tainter gate arms, and gate hoist superstructure steel.
For generator brackets, the deflection also should be determined because there is a limit imposed by the turbine runner seals. If Karl had not discovered the error at Eagle Falls, the bracket would have deflected so much that either the turbine runner would have settled to rest on the draft tube cone or the generator would have dropped onto the brake rim. In either case, considerable delay would have occurred while the bracket was being reinforced.
By James L. Gordon, B.Sc., hydropower consultant. Mr. Gordon may be contacted at 102 St.-John’s Boulevard, Pointe Claire, Québec H9S 4Z1 Canada; (1) 514-695-2884; E-mail: [email protected].