Jim Sower wrote:

I need a little help here.  I may have got lost back a ways.  Tell

me if

this is

true:
1.  Cavitation happens when the pump can't "suck" (read

"accelerate") the

coolant

into the impeller with the flow rate (velocity) that is required by

output

flow

rate.

2.  The event under discussion took place with the coolant being

drawn

>from and

returned to an open tank with coolant level at some distance

below the

impeller.

If these things are so, it occurs to me that in a closed system,

the coolant

supply to the pump would not be stagnant as in the experiment, but

moving

at the

same flow rate as the output side of the pump, and under steady state

conditions

(no acceleration of the coolant supply) no cavitation would or

could occur.

I intuit that the coolant supply to the pump being a stagnant pool

rather

than a

flowing, closed stream rendered the experiment bogus at the rpm

where the

onset of

cavitation occurred.

Where did I go wrong?           Jim S.
PS finally got my date/time straightened out ;-)

Jeff Spitzer wrote:

Your intuition is correct but the magnitudes may be wrong.  Sucking

>from a

stagnant source will cause the pressure to be lower at the inlet due to
accelerating the fluid.  If the inlet line is of reasonable size this
should be less than 1 psi though.  You are likely to get more pressure

drop

from the fittings to close the loop than from the re-acceleration of the
flow from stagnant.

Jeff Spitzer

Now I'm getting tangled up in intuition.  If pump pumps into block, and

line out of

block to radiator is 1.5" and line from rad to pump is 1.5", and coolant is
incompressible, and flow rate is whatever it is, pushing coolant out of

the pump and

into block is going to result in the very same volume being pushed  into

the pump

inlet.  I intuit that cavitation is a bit of volume with no coolant in it

(maybe this

is where I'm going wrong) and occurs when you pump coolant out of the pump

into the

block, and there is no coolant coming into the inlet side of the pump.

This seems

impossible to me in a closed system.

Could it be that at the temps we are dealing with, just a little 'suction'

or reduced

pressure at the inlet causes the coolant to vaporize locally and behave as

if there

was a vacuum on the inlet side?  That sounds like a bit of a leap.  But

something

doesn't seem to compute.  I can see the impeller spinning and not pumping

anything

out OR drawing anything in, but I can't see coolant pumped out and not being
replaced.
And I thought I had a handle on physics ;-)           Jim S.

you're concern about continuity is valid...what goes out must go in.  But
again, we're talking about magnitudes.  The more flow the higher the
pressures to pump it.  At some point you reach equilibrium where the pump
pressure rise matches the system pressure drop.  The flow rate for this
equilibrium is what changes.  If the pump has localized cavitation and can
only pump so much...thats what you get.  If you fix the cavitation then the
pump does better...more pressure rise...,more flow...new equilibrium point.

Jeff Spitzer

 
The Aircraft Rotary Engine Newsletter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org