Northern States Power (NSP) has a 384 megawatt coal-fired power plant in Minneapolis, Minnesota. Its Unit 7 and 8 coal bunkers were built more than 30 years ago. Typical of many coal bunkers built then and even today, little consideration was given during the design stage to the flow properties of the coal. As a result, flow problems can be expected, even when handling what is considered to be relatively free-flowing bituminous coal.

After switching to a low sulfur sub-bituminous coal, NSP experienced several fires in the Unit 7 coal bunker, and one explosion. Immediately, a task force known as "Operation Cease Fire" was set up to investigate and eliminate the spontaneous combustion problem in all of NSP's plants.

The problem was determined to be the funnel flow discharge pattern of sub-bituminous coal through the bunkers. The first-in last-out flow sequence allowed stagnant regions of coal to remain indefinitely. Such stagnant coal is highly susceptible to spontaneous combustion.

Simply stated, the solution to a funnel flow problem is to convert the discharge pattern to mass flow. With mass flow, all of the material is in motion whenever any is withdrawn. This eliminates stagnant regions, thereby providing a uniform residence time for all of the material.

To create a mass flow pattern, the hopper walls of a bin, silo, or bunker must be steep enough and have sufficiently low friction to allow the material to flow along them. Representative samples of the sub-bituminous coal were sent to one of our labs for flow properties tests using the widely accepted Jenike Shear Tester. With cohesive strength and wall friction data now available, our engineers were able to recommend modifications that would provide mass flow to the existing bunkers.

Due to economic constraints, NSP decided to modify only the Unit 8 bunker. The modifications first involved replacing the lower portion of a series of pyramidal hoppers with conical hoppers. We recommended that the valleys in the upper portion of the pyramidal sections that remained, be filled in with flat plates in order to provide a steeper slope.

Headroom was limited, so we knew that the cones alone would not be sufficient to ensure mass flow. Therefore, we recommended a BINSERT® system. Properly designed, such a cone-in-cone design enables mass flow to occur with angles that would not normally be steep enough.

For this application, our tests showed that TIVAR® 88 (UHMW polyethylene) was less frictional than mill finish carbon steel plate and stainless steel sheet with a 2B finish. Therefore, all of the remaining existing hopper sections and modifications were lined with 1/2 in. TIVAR® 88.

After the modifications were implemented and the unit was back on line, an immediate improvement was noticed. The limited live capacity of the original equipment had made it necessary to refill each bunker approximately every 12 hours. After the modifications, refilling was not required for 18 hours. This is a clear indication that the modifications dramatically increased the live capacity of the bunker. Further visual observations revealed that the bunker does in fact operate in a mass flow pattern as intended.

According to Jeri Huggins, Production Engineer at NSP, "It is a definite improvement to go an extra 6 hours between refills, but most important is the fact that no fires or explosions have occurred since the bunker was modified. The testing and design recommendations provided by Jenike & Johanson, as well as the engineering assistance from Poly Hi Solidur on the TIVAR® 88 installation, have been an unqualified success from start to finish."