How do you know cavitation isn't occurring, and how did you establish the fact that the pump/s aren't cavitating?

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.)

you don't know any basics.
NPSHA shall be greater by 1m than NPSHr as minimum.
The 3% head drop is method by which NPSHR is measured either by suction throttling or vacumm supression method.
Its not that it is not applicable for smaller pumps, you have again sum confusion.

NPSHA shall always be greater than NPSH3.

A well phrased question has answers in itself!!!

But why is NPSHr is measured like this?
Because "cavitation" is difficult to accurately define or measure the onset, whereas 3% drop is relatively straightforward and easy. Any air ingress into the water inlet tends to mask the typical noise heard.

Often the onset of cavitation is higher than the NPSHr curve hence the normal requirement to be min 1m above, however if you operate at lower or higher points on the curve than BEP, this margin can vary - see graph below.



Why in fact cavitation may not occur when NPSHa<NPSHr ?

Because you can't hear it or maybe there is air in the fluid which helps mask the noise / prevents cavitation.

And how come smaller pumps are actually less prone to cavitate, even when NPSHa < NPSHr ?
Not something I've come across as a trend, but maybe the external noise is just less relevant. Cavitation is impacted by specific speed so maybe that has something to do with it.

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

Possibly in some smaller pumps,especially in those of low to medium head, the increase in pressure in the impeller doesn't result in such a violet implosion as the cavitation bubble collapse, therefore less noise/ less damage.

Just my first-off thoughts.

Maybe Johhny Pellin will / can add something from experience with the small high pressure pumps in his plant.

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.)

NPSHR has +0% tolerance in every spec and standard that I am aware of. It must be a conservative number so that every time that pump is manufactured, NPSHR is less than or equal to the value that was quoted.

So, if NPSHA<NPSHR, a pump with perfect castings might not cavitate, at least not as measured by a 3% head drop. Small pumps are more sensitive to little variations in the castings (surface condition, vane thickness) so the difference between a "good impeller" and a "not so good impeller" has more of an effect on NPSHR than with larger pumps.

Thank you all for your inputs.