wrote:

Kind of a follow up to my pump RPM question.

I found a 1200 RPM 5 hp motor and an old Hitachi SJ200 5hp. vfd in the spare
stockroom (er junkpile)

Any reason I can't spin this combo at 3600 RPM? Still have 5 horse?



Speed.

Whenever a problem involves safety there is always one or
more ultra conservative health and safety fan who will paint a
picture of remotely possible extreme dangers arising from
deviation from manufacturers rated conditions.

Certainly motor operation outside normal manufacturers
published rating carries with it a degree of risk. This has to be
a personal decision and this is the way I look at it.

In the 50/60Hz induction motors I have handled I have not
seen mechanical design changes related to the different operating
speeds of 2, 4 and 6 pole machines. Manufacturers appear to use
the same rotor and bearing construction throughout - only the
stator changes. The rotor construction is extremely strong and
the same arrangement is used inaircraft machines operating at
24,000 RPM.

A possible but unlikely problem is the degree of mechanical
unbalance in the rotor. Rotors are routinely trimmed for
mechanical balance. The forces arising from the remaining
residual unbalance rise as the square of the speed. If the
balancing is poor it will become evident as excessive vibration.

On this basis I would have no hesitation in operating a 4
or 6 pole motor at 2 pole speed.

With a suitable VFD I would also be prepared to go well
beyond this.However, in this case, I would first test the motor
in a position wheremechnical failure was not a safety hazard and
never subsequently use it at more than 2/3 test speed.

Motor electrical behaviour.

The copper behaviour is roughly independent of frequency so
the heating resulting from copper losses remains directly
proportional to current squared. This is easily checked and,
provided it remains below manufacturers rated full load current,
there is no problem.

Iron losses are significant. Dependent on design, total iron
loss may be comparable with total copper loss. When operating at
a different frequency the iron losses change. Iron losses are non
linear but, for a first approximation, at flux densities and
frequencies appropriate for our motors, losses rise as the square
of the flux density and directly with frequency.

Although common VFDs can increase the motor drive frequency
up to and beyond several times supply frequency they cannot
correspondinly increase the motor voltage above supply voltage.
Since this means that the motor flux density drops (and
correspondingly reduces the torque per amp of current), the
reduced flux density loss more than compensates for the loss from
increased frequency so the total iron losses reduce.

This meas that there is no overheating problem resulting
from increased frequency at constant motor voltage. With a near
ideal motor the available output power would stay at rated power
as the speed is increased

However with real motors not all of the flux generated by the
ampere turns of the stator windings couples directly to the
rotor. This failure to couple is leakage inductance and because
it effectively appears as an inductance in series with the
windings it reduces the useful voltage reaching the motor. This
effect is directly proportional to frequency and is very
dependent on the layout of the stator windings. Most motors will
behave pretty well as as expected up to at least 1 1/2 times
supply frequency. Good operation is possible well beyond this but
much depends on the detail design of the windings.

Your proposal to treble the supply frequency is certainly
well into the uncertain area. I think there's a good chance of
getting a large fraction of 5HP at the higher speed but you need
to watch for possible large RPM drops. Normal operation and the
above comments assume that the motor is operating within a
hundred or so RPM of synchronous speed. If the rotor speed drops
well below synchronous speed the motor efficiency drops and a
large fraction of the input power is dissipated in the rotor
bars. This excess power loss is directly proportional to slip
speed (RPM below synchronous) and can result in excessive motor
dissipation even if the supply current remains below rated
current.

One other possible problem is fan loading - fan power
absorption rise as the cube of the speed. The power absorption of
any fan attached to your motor rises 27 times!!

Test results on your setup wuld be very interesting -
keep us posted.

Jim