Today I have come across a very strange centrifugal compressor damage. We removed a Natural Gas centrifugal compressor rotor after 3 years of running for routine overhaul, and found a very deep and wide groove on the shaft drive end close to the gas seal. The groove looks like as someone had machined it. Surprisingly there are no signs of damages on any of the stationary parts. We have not found any foreign body in the compressor casing and the gas is very clean.
The white metal lining of gas seals however does not exist. This seal is oil type with Nitrogen as buffer medium. Nearly 3 years ago, due to a process upset, while starting the machine, seals were blown since then compressor was running with pressurized oil console. The oil console is common for seal and lube oil. The online vibration system was out of order for a long time, therefore no vibration, phase and thrust history is available. Direct vibration history from DCS indicates a magnitude of 0.21 mils maximum.
Between March and May 2002, the thrust was changed from –7.1 mils to –3.7 mils. Since then it is gradually changed to –1.9 mils. The rotor speed is 12000 rpm, driven by a steam turbine and connected through a LP casing. The LP rotor is perfect and clean. The maximum gas pressure is 45 bar. Few photographs are attached here.
Has any one seen this type of damage? What could have caused this?
Posts: 8 | Location: Qatar | Registered: 08 March 2005
Originally posted by Arshad: Today I have come across a very strange centrifugal compressor damage. We removed a Natural Gas centrifugal compressor rotor after 3 years of running for routine overhaul, and found a very deep and wide groove on the shaft drive end close to the gas seal. The groove looks like as someone had machined it. Surprisingly there are no signs of damages on any of the stationary parts. We have not found any foreign body in the compressor casing and the gas is very clean.
The white metal lining of gas seals however does not exist. This seal is oil type with Nitrogen as buffer medium. Nearly 3 years ago, due to a process upset, while starting the machine, seals were blown since then compressor was running with pressurized oil console. The oil console is common for seal and lube oil. The online vibration system was out of order for a long time, therefore no vibration, phase and thrust history is available. Direct vibration history from DCS indicates a magnitude of 0.21 mils maximum.
Between March and May 2002, the thrust was changed from –7.1 mils to –3.7 mils. Since then it is gradually changed to –1.9 mils. The rotor speed is 12000 rpm, driven by a steam turbine and connected through a LP casing. The LP rotor is perfect and clean. The maximum gas pressure is 45 bar. Few photographs are attached here.
Has any one seen this type of damage? What could have caused this?
Posts: 8 | Location: Qatar | Registered: 08 March 2005
There was no material of any kind in that groove? No build up? I have seen this kind of damage when there is a build up of an abraisive material in a seal area, and as the shaft wears the material continues to build up to fill in the space and then continues to machine the shaft. In our case the material was a butadiene polymer that built up in a balance piston seal area and wore down the balance piston seal by about 4 inches off the diameter. We never saw any indication that it was happening on the vibration monitoring system.
e-mail me at steven.schultheis at sbcglobal dot net
Posts: 293 | Location: Kuala Lumpur, Malaysia (in transition) | Registered: 21 February 2005
There is no material build up in the groove. It is very clean. The maximum vibration was not more than 0.22 mils, unfortunately our data manager system was out of order for a long time so dynamic data history is missing. Any possibilty of gas shear? is it akin to wind shear phenomina in aircrafts?
Posts: 8 | Location: Qatar | Registered: 08 March 2005
OK, what is the possibility of corrosive gas or liquid being in that area and machining away the shaft? What is the H2S concentration in your gas? I have heard of corrosion damage to shafts in seal areas. If there was nothing there to rub the shaft away, it must have been eaten away. I don't see how any kind of gas shear or dynamics could have caused this unless there is something either corrosive or abraisive in the gas stream. The white metal on the seals being gone says corrosive to me.
e-mail me at steven.schultheis at sbcglobal dot net
Posts: 293 | Location: Kuala Lumpur, Malaysia (in transition) | Registered: 21 February 2005
Gas is very clean, H2S is almost nil. Some sand type of material was reported in the LP casing, where we did not find any kind of damage or unsusual wear. HP casing was free of sand but seal area machined out. Moreover, there are three more compressors, compressing the same gas, vibration and performance history does not reveal any problem with their rotors.
As I mentioned earlier, Buffer seal gas was at more positive pressure for the last three years therfore possiblity of seal area wear due to sand or corrosivity of gas being compressed looks remote. White metal of seals was gone quickly as a result of some process upset three years ago, not due to any other reasons.
I will try to post any new findings soon.
Posts: 8 | Location: Qatar | Registered: 08 March 2005
