Rule of Thumb for Running Idle Pumps / Compressor

[MPP-LTD]

HI,

I am looking to get feedback from operators / maintenance personnel from process facilities on how often to jog starting pumps, compresor and fans that are used infrequently (e.g. less than once per year if ever) but must be available for emergencies. I understand that each equipment requriement should be specific to the equipment, site conditions, maintenance policy, etc. however I I have found vendor literature only covers preventative maintenace for equipment in regular operation or long term storage and nothing for emergency equipment that is rarely used but must be available.

If any further information is required I am happy to provide.

Thanks,

 

[LittleInch]

Hmmmm, I would guess this comes from a RAM analysis or just good practice.

I suppose you start at once a week ( basically using the NFPA fire pump testing frequency ) and go from there depending on criticality, past performance of the unit, input from vendors and plain common sense.

I wouldn't go beyond 6 months for anything which is needed for an emergency as things like batteries fail or lose charge, fuel deteriorates and systems can become solidified / gum up etc.

Some idea of what this equipment is and what its criticality to your process / requirement is might help to get some more inputs?



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MPP-LTD]

The equipment is for a power station. One Gas Turbine has equipment that allows it to be started in the event that there is no power exertnal power from the grid or internal power generated (other than batteries and diesel generator). There is an air compressor, closed cooling water pump and some fin fan coolers.

The equipment is probably less critical than a fire pump, but it is still very important. In my head I thought once a month would be ok.

 

[LittleInch]

Once a month could be fairly onerous and depends on how much time / effort / cost is involved.

Every two to three months sounds more reasonable from a long distance, but you know what this actually involves more than me.

Also will any test actually stress or use any equipment to 80-90% of duty? there is a risk that simply spinning any equipment won't actually tell you whether the item will work for an extended period or could actually damage pumps etc if there is no flow.

How about a test every 3 months and a full scale test / activation / turbine start every 12 or 24 months to make sure it does actually do what it says it will do??

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MPP-LTD]

That is a good approach to split up the preventative maintenace vs functional testing of the system as a whole. I think that the time frames make sense as well.

Thanks a lot for your suggestion