Real life episode - a recently overhauled open circuit hydraulic pump
(A10VO45DFLR), equipped with pressure compensator, load sensing and
torque limiter control, was about to be tested on a bench. The client
had requested the torque limiter to be set to 15 KW (20HP) at 1450
rpms, which was the nominal power/speed of the electric motor
driving the pump. The request was made due to the plant's electrician
advice to limit the motor current consumption to 30 amperes. In fact,
the exact words were - "adjust the pump to 15 KW or 30 amperes".
The test bench used to test and adjust the pump was
of a very simple kind - a common three phase electric motor (thank you
very much for these, mr. Tesla), simple mechanical coupling, an oil
tank, a needle restrictor and a flow-meter - very basic yet functional
and reliable rig. The technician that was testing the pump took even
the trouble of finding an average AC motor current table (like this one),
and indeed confirmed that the current draw of a 15 kw motor (three
phase, 380 volts, 50hz) rounded 30 amps. To adjust the torque limiter
he proceeded exactly in the way he had adjusted many similar controls
on pumps driven by electric motors before - by reading the phase
current and adjusting the control so that the current doesn't pass a
certain value.
The "adjusting maneuver" was performed flawlessly in
a "by the book" fashion - the current draw in one of the phases of the
test rig motor was being monitored, the torque limiter was set to start
cutting the displacement at 30 amps, and then to maintain the current
roughly at the 30 amps level up until the pressure compensator kicked
in. However, when the actual hydraulic output (pressure times flow)
power was calculated, it turned out to be in the neighborhood of 7kw
(10 hp) - after several tests a conclusion was made that the pump's
overall efficiency was unacceptably low (50 percent). As the pump
didn't have excessive case leakage, the fault was attributed to low
quality spares causing "excessive mechanical drag", the unit was
rendered "faulty beyond repair", furthermore it was decided that the
pump needed to be replaced with a new "original" one, and the client
once again had to face expensive downtime costs until the new pump
arrived.
What is wrong with this picture? If I were asked
this question, I'd say that before answering it I'd have to know how big
the test bench electric motor is. Why? Because of the little thing
called "power factor", which is a strictly AC power related unit, and
in simple turns is the ratio of the real power (the power your motor is
consuming to deliver the demanded torque) to the apparent power (line
current times line voltage).
Most industrial three phase motors have power factor of around 0.85 at nominal loads, and the above mentioned table is only valid for nominal load values. When the load of an AC motor is less than nominal, the power factor (along with the efficiency) drops (schematic relation is shown here), which means that the phase current of a lightly loaded motor can not be
used as a reference to determine the output torque simply because it
indicates mostly the apparent power and not the real one. I don't want
to go into detailed AC power theory here, but those of you who are
interested in learning more about AC motors and what a power factor is
can start here.
Due to the fact that the test bench in question was
built to be able to test larger pumps (the AC motor of the rig was
rated to 70 HP), the motor load was well below its nominal value during
the test of the "malfunctioning" pump, and therefore the phase current
the mechanic was reading didn't represent the real load current.
In this case not knowing what a AC motor power
factor is caused an extensive downtime and unnecessary expenses. This
example shows that knowing a thing or two about what is at "the other
end of the shaft" can't hurt a pump mechanic. Understanding at least
basic theory of modern industrial motors and engines operation has
become an essential part of a "knowledge package" any hydraulic
technician should carry behind his back.
