In my previous post I mentioned the impact of motor slip on the performance of a centrifugal pump. When I bring this topic up in classes, there are often questions. So,to better understand what is going on, a quick review of how a multi-phase "squirrel cage" induction motor like the ones we typically use for our HVAC equipment works may be in order. A very cool discussion of this including some really great animated graphics can be found on the University of New South Wales website, as illustrated below.

Their is a link from that page to a related page that is also very cool and shows the guts of a lot of different motor types and discusses their operating principle as illustrated below.

What follows is my somewhat simplistic understanding of how a squirrel cage motor works. It seems to work for me so maybe it will be helpful to you. The discussion assumes we are talking about a 3 phase motor.
A typical squirrel cage motor contains stationary windings which we'll call stators, that are connected to the three phase power supply. Suspended inside the stators are rotating windings which we'll call the rotor, that drives the output shaft.
As you will recall from your physics class, when current flows through a wire, it generates a magnetic field. As a result, the current flowing through the stator windings causes them to act like magnets with the strength and frequency of the magnetic field varying as a function of the alternating current supplied by the power company. In addition, since the power is three phase - we'll call the phases A, B and C - and since the current in the phases peaks consecutively in time (versus concurrently), the magnetic field in the windings connected to the A phase will be rising toward its peak when the field associated with the windings connected to the B phase is at its peak. Meanwhile, the field associated with the windings connected to the C phase will be declining.
Of course, since the current is continuously alternating, the fields in the windings will be continuously varying, one after the other. If you consider that for a minute, you can probably see how if one arrange the windings properly, you could create a rotating magnetic field whose frequency was a direct multiple of the utility company frequency and the number of windings. For instance the rotating magnetic field in a two pole motor operating on a 60 hz utility system will spin at 3,600 rpm, which is termed its synchronous speed. For a four pole motor on a 60 hz system, the synchronous speed is 1,800 rpm; for a 6 pole motor its 1,200 rpm.
Now lets talk about the rotor for a minute. There is no physical connection between the rotor and the power supply. But because the rotors' windings are located inside the field created by the stators and because that field expands and collapses due to the alternating current supplied by our utility companies, currents are induced in the windings of the rotor. And since current flowing through wire creates a magnetic field, the rotor itself becomes a magnet.
Now recall from physics that if you place two magnets in close proximity to each other, they try to align themselves opposite pole to opposite pole. So, if you visualize what might happen when you put a magnet inside the rotating field we just discussed, you can probably see that the magnet would start to spin as it tried to line itself up with the rotating poles created by the rotating magnetic field. This is exactly what the rotor does and it is why the output shaft of the motor spins.
But, there is a catch. Remember that a wire must be moving relative to the lines of flux associated with a magnetic field to generate a current. That means that if the rotor ever caught up with the rotating magnetic field created by the stators;
Its windings would no longer be moving relative to the stator's lines of flux, which would mean that
No current would be generated in the windings of the rotor, which would mean that
The rotor would cease to be a magnet, which would mean that
It would stop trying to rotate.
Of course, if it stopped rotating, then the stator's lines of flux would again be moving relative to the windings in the rotor and it would start to become a magnet again. The bottom line on all of this is that the output speed of a squirrel cage induction motor will approach but never equal the synchronous speed of the rotating field in the stators associated with it. The difference between the two speeds is termed slip, and it varies with the load on the output shaft of the motor.
The reason it varies with load is that as you apply a load to the output shaft, it will tend to slow down relative to the rotating field. As a result, the rotor's winding are cutting more lines of flux in the stator's field. If more lines of flux are cut, then the current in the field in the rotor increases, increasing its magnetic strength and allowing more power to be transfered.
With nothing attached to the shaft of the motor, the only resistance is the bearing resistance, windage (aerodynamic losses associated with rotor spinning in air), the work being done by the cooling fan to move air through the motor, and similar losses, all of which are part of the losses associated with the motor's rated efficiency. Under this condition, the rotor spins at nearly the synchronous speed.
As load is applied, the slip increases until at the rated load, the difference between the synchronous speed of the rotating stator field and the slip equals the rated motor speed. If load continues to be applied, the rotor speed will continue to drop relative to the synchronous speed and the motor will actually produce more than its rated horsepower.
But, if you keep trying to take more power out of the shaft, at some point, things will break down and the strength of the field induced in the rotor will not be enough to overcome the resistance to rotation associated with the load, no matter how many lines of flux are cut, and the rotor stops rotating. This point is generally termed the break down torque for the motor.