As a basis you should start from the manufacturers recommended service interval so that one pump is being serviced / maintained whilst the other one starts its operating time / cycle.
Anything less than that is usually just something that develops over time in a company so that each pump is operated for equal periods over say a year.
So long as the standby is properly isolated / drained then there should be no bad effects on having a standby as, well, a standby.
My motto: Learn something new every day
Also: There's usually a good reason why everyone does it that way
Littleinch wrote:
"So long as the standby is properly isolated / drained then there should be no bad effects on having a standby"
In my experience the pump on "Stand-By" was not isolated and drained (unless it was in fact to be worked on or removed). The switchover frequency for a Service/Stand-by was part of the Start/Stop/Auto operation of the vast majority of the pumps in the Units. They were not taken Off line, they were only cycled off and left idle while the companion pump operated and maintained the system service. To do this the idle pump comes on and gets up to speed before the first pump is shut down. This results in a slight overlap of pump operation but no loss of production.
I agree that that is how it often done, my point was that leaving the standby pump full of product and sometimes with the inlet valve open with just the discharge valve actuated was that you could get seal damage and leakage and potential for corrosion, settling of any solids and other issues associated with long term idleness. In that type of operation you probably want to look at quite regular change over of the duty pump - say at least once a week?
Depending if the service is continuous or intermittent will determine if the change over needs to be "bumpless" or not.
My motto: Learn something new every day
Also: There's usually a good reason why everyone does it that way
"I still wonder what is the appropriate procedure with regard to the centrifugal pumps switchover frequency?"
I assumed that from the posts received so far that it would have been reasonably clear there is no "appropriate procedure", as it comes down to the application, the likely deterioration during service and plant history/ experience. If the pumps in question are in continuous operation and have a service life of 2 years between overhauls / service there wouldn't be much sense in rotating them every other day - however if the pumps run on and off all day there probably isn't any problem in rotating then each week or even each day or each start-up if it doesn't cause any start / operation problems- but entirely up to you, it's your plant and you have the information (or should have)on which to base any decisions.
It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)