I have seen problems with multi-stage condensing steam turbines if the seal steam and gland condensors were adjusted incorrectly. In one particular turbine this is what happened. The operators mistakenly opened a vent on the leak-off line that diverts steam from the high pressure end of the turbine to the low pressure end of the turbine during normal operation. This particular turbine is run from full speed down to slow roll several times per day. When the turbine would come down to slow roll, the case pressure would drop and cold air would be sucked in that open vent. When the cold air hit the hot shaft, the rotor would bow and the vibration would spike up very, very high. If there is any chance that vacuum at the gland packing could be drawing air in along the shaft, that can definately cause a vibration spike. For most condensing tubines, seal steam is introduced from an outside system for startup. Once you are up and running, the leakoff of steam from the packing case on the high pressure end supplies the sealing steam to the vacuum end. In some cases, if there is excess, it may be necessary to vent off some of this steam. A slight amount of steam leakage from each end is important to be sure you do not draw in cold atmospheric air along the shaft. By phase angle, I am refering to the referance angle of the vibration. If a vibration is occurring at one times run speed and the phase angle is changing, it could indicate that the balance of the rotor is changing, perhaps from mineral buildup. If the vibration is occurring at a non-synchronous frequency, and the phase angle is shifting, it could indicate some part of the machine is moving in and out of resonance. A rotor resonance could be excited by a light rub. Even a very faint rub in a shaft laby seal can excite a rotor resonance in some machines. I would have thought that this could not exist long term since it seems like the steel shaft would wear away the softer aluminum and the rub would stop. But I have seen long-term vibration phenomenon that were traced back to a light rub in an aluminum laby seal. A rub could be diagnosed by looking at the vibration spectrum and the orbit. A rub might have a frequency at about 1/2 times run speed and the orbit might have an internal loop indicating a rub with a bounce on every second revolution. It is more likely that your tubine has steel laby seals (J-strips or L-strips) or perhaps carbon rings. In the final analyis I would suggest the following:
Make sure that the seal steam and gland condensor systems are adjusted as the manufacturer recomends. Make sure you cannot be drawing in cold air along the shaft.
Examine the vibration spectra to determine the frequency at which the high vibration is occuring. Examine the shart orbit during the event to look for a possible internal rub.
When you say "Flexible Coupling" I assume you mean disk pack, diaphragm, etc. If the main drive coupling is a gear coupling, it could be intermittently locking up and causing the vibration. But I think it is more likely a rotor bow or internal rub. I almost forgot about bearing instability as a possibility. An oil whirl in a radial bearing could be an intermittent problem. This would also tend to show up in the spectra or shaft orbit.
