Diagnosing other models
For Westinghouse turbines with an IP1-HP-IP2 arrangement (Figure 4), there are also two values of measurable IP apparent efficiency. The IP1 efficiency is measured from the hot reheat to the IP1 exhaust using vendor-supplied casing pressure and temperature connections in the lower test wells that are somewhat difficult to access on-line. (On some units, there is an extraction location on the IP1 exhaust that can be used for pressure/temperature measurements.)
4. Horse of a different color. A Westinghouse 180-MW IP1-HP-IP2 turbine. Courtesy: Southern Company Generation
IP1 efficiency values account for the effect of leakage of HP exhaust steam leakage to the IP1 inlet through the IP dummy seals. The total IP efficiency, measured from the hot reheat to the LP crossover pipe, includes the effects of leakage through the IP and LP dummies, the bell seals, the HP exhaust piston rings, the hot reheat inlet piston rings, the governor valve stems, and the seal at the end of the governor shaft. Possible sources of nondesign leakage include broken equilibrium pipes, a leaking HP-IP1 inner shell horizontal joint, broken first-stage drain and pressure-sensing lines, and cracks in main steam inlet pipes.
Until recently, the only unusual problem that SCG has experienced on the IP1-HP-IP2 type of turbine was a very high IP dummy leakage on one unit, but not on a sister unit. The problem was diagnosed as "extra" cooling holes found in the IP dummy ring, and it was fixed by reducing in size (welding up) the holes, as the existing IP dummy seals would provide sufficient IP inlet bucket cooling steam. A post-outage test revealed that taking this action substantially reduced both IP dummy leakage and total HP to IP leakage. Including the effect of reduced reheat spray flow, the unit heat rate improvement was about 40 Btu/kWh.
Smaller (100-MW and less) Westinghouse turbines have piston seal rings on the main steam inlet pipes. On these units, excessive leakage will reduce HP turbine efficiency as some or all of the HP blading is bypassed. Total HP-to-IP leakage is not affected by this leakage through the piston seal rings.
On SCG's 100-MW Westinghouse turbines, the LP dummy flow (external pipes) to the IP exhaust can be measured using the vendor-supplied flow measurement/restriction orifice plates. On one occasion, the rotor of one unit was visibly thrusting, causing measured LP dummy flow to cycle. The results of temperature split testing were inconclusive due to the swings in leakage. Eventually, the turbine required an inspection to check for wear of its thrust bearing. The root cause of the thrusting was determined to be worn-out IP dummy seals, which were no longer in a hi-lo configuration (Figure 5).
5. Seals don't lie. A 100-MW Westinghouse turbine with severely worn IP dummy seals that caused noticeable thrusting of its rotor. Courtesy: Southern Company Generation
Later Westinghouse units have a mid-span balance port that can be used for borescope inspections to check for loose HP inner shell bolts and broken equilibrium pipes. On BB44 units, this can also be accomplished by removing an LP crossover pipe.
Allis-Chalmers reheat turbines also are susceptible to several sources of design and nondesign internal leakages. Figure 6 highlights these sources for 75-MW Allis-Chalmers units, of which SCG has three. The sources of design leakage to the IP turbine include the IP dummy, the reheat diaphragm packing, and the LP dummy. Excessive leakage through the IP dummy and reheat diaphragm packing into the IP turbine inlet would produce an increase in measured IP efficiency. By contrast, excessive leakage through the LP dummy into the LP cross-around pipes would cause a decrease in measured IP efficiency. Sources of nondesign leakage include the reheat diaphragm packing housing joint and the horizontal joint.
6. Other sources of internal leakage. The HP, IP, LP, and reheat diaphragm packing of an Allis-Chalmers 75-MW turbine (L) and the HP-IP-LP dummy seals location (R). Courtesy: Southern Company Generation
Nondesign leakages—through the two sets of main steam inlet piston seal rings or the HP inner shell horizontal joint, or due to a broken first-stage pressure-sensing line/drain on the inner cylinder—affect the performance/efficiency of the HP turbine, but not leakage into the IP turbine. SCG has used the Booth-Kautzmann temperature split test to trend the leakage rate before and after a turbine outage on Allis-Chalmers units. We also have temporarily opened the reheat diaphragm (similar to the GE N2 packing blowdown test) and balance piston unloading valves to calculate the reheat diaphragm packing leakage and IP dummy leakage, respectively, based on the change in measured IP efficiency.
Email
Comments
Print
Reuse
