Maintenance strategy based on Reliabilty Centered Maintenance (RCM) 1

[ahung]

In our power plant we have five (5) combustion turbines burn natural gas on simple cycle. Normally when we do the major outages (Hot gas Path Inspection or Major inspection) in average each three year (24.000 hours)we change all electric motors bearings (for example for lube oil pumps, fans, blowers, etc) for new ones, change some hoses, rebuild (change gaskets kit)regulators valves for instrumentation air system, change some belts for new ones. We don´t care about the bearing, hoses, belts conditions we remove the olds and put new ones.

This criteria can be considered a Reliability Centered Maintenance (RCM) strategy? The final goal is not have to shutdown the CT for change any bearing, hose or belt on any auxiliary equipment for three years until the next major outage. In others words in general terms we want with this strategy reduce the EFOR indicator and increase the availabilty indicator EAF for reach a better performance of the plant.

I appreciate some comments about this maintenance strategy


Regards

Alberto J. Hung C
Caracas Venezuela

 

[IRstuff]

What is the MTBF of a brand-new system? If it's less than 3 yrs, then what???

If the bearing life is 10 yrs, then you're wasting quite a bit of life. That may be worth it to your application, though.

You can't control random failures.

TTFN

 

[nnjunger]

This system is not in my opinion RCM. At best, it is a time based preventative maintenance overhaul. This may or may not be the best startegy for your application. Not only could you be wasting money on unneccesary parts, but you could introduce contaminants to the system, take the risk of infant mortality into the system (whether via parts defects or workmanship), as well as a host of other issues.

I think one of the fundamental questions you need to answer is how effective has this strategy been on achieving the intended results. If you can achieve your results, it is cost effective, and the consequence of failure is higher than the possibility of problems with conducting the repair, then utilize the strategy.

 

[jeez]

ahung: If your company has not performed a formal RCM analysis on these CT assets to identify the overall system's probable failure modes, effects, consequences and identify corrective/containment actions for a PM/PdM approach, then answer is "No". You don't have an RCM strategy just because you use some of the attributes of corrective measures identified during typical RCM analysis. I agree with nnjunger.

You are also, with each "corrective PM action" you take, building in a higher probability of failure by the mere involvement of humans touching the equipment, as nnjunger mentions. For example, during the belt changes your maintenance tech could inadvertently cause misalignment that goes unnoticed..and premature belt failure or sheave bearing failure occurs before your 3 years. Maybe the belt-driven subsystem is Critical, maybe its not.

With a true RCM you would have first, in a series of analysis sessions with the appropriate experts, identified the sources of probable failure or low reliability of the CT systems. And determined then what specific corrective maintenance; preventive maintenance; predictive maintenance; design changes, or run-to-failure actions you should take to ensure the highest overall reliability of the systems longterm. There is little or no "blindness" or "shotgunning" to RCM, if done properly.

But it seems you guys are satisfied with your approach? How long will that last?

just some thoughts...take care.
skm

 

[svanels]

Reliability Centered Maintenance is defining the right maintenance strategy for the right equipment, given the operational/environmental/economical constraints.

Given something complex like 5 CTs in a powerplant, the economical effect of an emergency shutdown for a $50 bearing on an electric motor, would have serious consequences for further employment at the company.

If the bearings are of the sealed type (non greasable), I would change them anyway. If they are not, and performing a small repair does not hamper (redundancy) the operation of the turbine, I could opt for vibration monitoring and change them when convenient.

During a planned outage you have the luxury of a thorough inspection, a running turbine is a different animal, with production demands, log-out tag-out, permit to work, safety precautions, stress!! etc..
Also the 3 years give room to detailed planning and scheduling.

Given the application, the cost of a CT, and the consequence of not running the strategy I consider RCM based.

 

[ahung]

Thanks all for this valuables comments about my question.
In order to clarify with more details the majority of bearings are sealed type and the electric motors are locate in difficult access places. Another factor to be consider is our plant running 24 hours 365 days of the year at base load condition. I agree 100% with svanels point of view because he said something is definitely true: "A RUNNING TURBINE IS A DIFFERENT ANIMAL"


Regards

Alberto J. Hung C
Caracas Venezuela