Up Time Magazine - December 2008/January 2009 - (Page 59) to the shaft and damaged it. If nothing else, maybe a few simple tests would eliminate the possibility of a bad shaft. The first thing was a simple bump test. To everyone’s surprise, it rang like a bell around running speed and at two times running speed, (Figure 6). This natural frequency had not always been present. Where did it come from? What had changed? Either the mass or stiffness had to change in order to shift the natural frequency. A crack in the shaft could possibly explain a change in stiffness resulting in a change of the natural frequency. However, neither dye penetrant nor ultrasonic testing uncovered any evidence of a crack in the exposed portion of the shaft. If there were a crack, it was where these tests could not detect it. Any further testing would require pressing the iron off of the shaft. Besides being expensive, this could easily ruin the shaft. Well, either the shaft was already ruined, or it was going to be ruined in order to find out if it was ruined. We decided to stop doing any more testing and not take excessive care to save the shaft (see Figure 7). Instead, we would just build a new shaft. ceptance testing before placing them in service. This establishes a chain of quality custody and a new baseline data for future comparison. If the motor does not pass acceptance, either something happened during transit, the motor base has a problem, or there is some on-site assembly problem because the shop tests prove the motor to be fine. Well, as you might have already guessed, when this motor was run uncoupled for acceptance, it failed. Normally, a slight balance adjustment will allow a rebuilt motor to meet the acceptance criteria. This motor did not respond predictably to weight movements., and further diagnosis discovered symptoms other than imbalance. There was still some 2x and a little 3x in the spectrum, with the majority of vibration at running speed. Figure 8 shows the motor being tested on its own base and foundation. Figure 8 – Test run of motor on its own foundation and base Figure 7 – Shaft removal Measurements were taken from the old shaft and a new one was turned from similar materials. The new one was bump tested before the rotor iron was pressed on, and again after the assembly was complete. These tests showed no natural frequencies near running speed or running speed harmonics. The rotor was balanced. The motor was assembled, once more placed on the test pad, and, after a slight trim balance, the motor passed acceptance testing. The motor was finally placed on a truck, the rotor blocked to prevent any movement, and shipped back to the Muskogee Power Plant to be placed in service. Motors are routinely run uncoupled for acwww.uptimemagazine.com There was no obvious shipping damage. The motor was uncoupled, so there was no misalignment. Imbalance had been ruled out by process of elimination. The base and/or the foundation were really all that were left. A visual inspection of the base revealed that a couple of nuts were not even touching the base. In fact, the blade of a pocket knife could fit in between the nut and the base. The grouting was also cracked and deteriorating. A machinist’s level placed on the base determined a severe out of level condition, so 40 mils of shims were required to level the base. After this temporary repair, the motor was once more run uncoupled. Figure 9 on the next page shows the shim pack under the bottom right hand corner of the base plate. The vibration was greatly reduced, but not yet within expected readings for an unloaded motor. Most of the vibration was now at running speed, and phase analysis indicated a 59 http://www.reliability.com.au/sapimplementation http://www.reliability.com.au/sapimplementation http://www.uptimemagazine.com
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.