Vibration and Temperature Rise

But we learned to care. You see, rotor fouling is a double-edged sword. It cuts two ways. One aspect of rotor fouling is loss of adiabatic compressor efficiency, which wastes work and reduces flow. The other edge of the blade is vibration. 

Eventually, the fouling deposits on the rotor will become so thick that they start to break off, especially if you shut the compressor down for a few hours for minor repairs to the lube-oil system. When the compressor is put back on line, bits and pieces of grayish salt break off, and unbalance the rotor. At 8000 rpm, the high-vibration trip cuts off the fuel to the gas turbine, and the machine is taken off line for repair. 

The compressor is disassembled. I get the opportunity to accompany the rotor to Dallas, in the back of a van that needed a new suspension. Once there, I watched the manufacturer s machinist crew clean and rebalance the wheels. I noticed that the salt deposits were thickest on the middle wheel. The last wheel was only slightly encrusted while the first wheel was clean, except for some waxy grease. 

Why this sort of salt distribution I reasoned that the entrained brine did not dry out until it reached the middle wheel. But, by the time it reached the last wheel, all the salt deposits that were going to accumulate in the compressor had done so. 

This gave me an idea. Suppose we injected a liquid spray into the front end of the compressor (we eventually used a heavy aromatic naphtha, obtained from a local refinery). This could prevent the deposits from sticking to the spinning wheels. We tried it, and it worked. Rotor fouling and the consequent vibrations, and loss of capacity, became far less frequent. 

---