Answer MET Question 2

 Question: A. Explain the methods used to control the speed of a 3 Phase induction motors. 

B. Draw and Explain a Variable Frequency Drive used for optimization of energy efficiency of auxiliary machineries on board vessels.

Answer:- Speed Control of Induction Motors
A 3-phase induction motor is practically a constant-speed machine, more or less like a d.c. shunt motor. The speed regulation of an induction motor (having low resistance) is usually less than 5% at full-load. However, there is one difference of practical importance between the two. Whereas d.c. shunt motors can be made to run at any speed within wide limits, with good efficiency and speed regulation, merely by manipulating a simple field rheostat, the same is not possible with induction motors. In their case, speed reduction is accompanied by a corresponding loss of efficiency and good speed regulation. That is why it is much easier to build a good adjustable-speed d.c. shunt motor than an adjustable speed induction motor. Different methods by which speed control of induction motors is achieved, may be grouped under two main headings 
1. Control from stator side 
(a) by changing the applied voltage 
(b) by changing the applied frequency 
(c) by changing the number of stator poles 
2. Control from rotor side 
(d) rotor rheostat control 
(e) by operating two motors in concatenation or cascade 
(f) by injecting an e.m.f. in the rotor circuit. 

A brief description of these methods would be given below 
(a) Changing Applied Voltage 
This method, though the cheapest and the easiest, is rarely used because 
(i) a large change in voltage is required for a relatively small change in speed 
(ii) this large change in voltage will result in a large change in the flux density thereby seriously disturbing the magnetic conditions of the motor. 

(b) Changing the Applied Frequency 
This method is also used very rarely. We have seen that the synchronous speed of an induction motor is given by $\displaystyle \small \mathrm{N_s = \frac{120f}{P}}$ . Clearly, the synchronous speed (and hence the running speed) of an induction motor can be changed by changing the supply frequency 'f'. However, this method could only be used in cases where the induction motor happens to be the only load on the generators, in which case, the supply frequency could be controlled by controlling the speed of the prime movers of the generators. But, here again the range over which the motor speed may be varied is limited by the economical speeds of the prime movers. This method have been used to some extent on electrically-driven ships.

(c) Changing the Number of Stator Poles 
This method is easily applicable to squirrel-cage motors because the squirrel-cage rotor adopts itself to any reasonable number of stator poles. From the above equation it is also clear that the synchronous (and hence the running) speed of an induction motor could also be hanged by changing the number of stator poles. This change of number of poles is achieved by having two or more entirely independent stator windings in the same slots. Each winding gives a different number of poles and hence different synchronous speed. For example, a 36-slot stator may have two 3-ф windings, one with 4 poles and the other with 6-poles. With a supply frequency of 50-Hz, 4-pole winding will give $\displaystyle \small \mathrm{N_s = \frac{120\times 50}{4}}$  = 1500 r.p.m. and the 6- pole winding will give $\displaystyle \small \mathrm{N_s = \frac{120\times 50}{6}}$ = 1000 r.p.m. Motors with four independent stator winding are also in use and they give four different synchronous (and hence running) speeds. Of course, one winding is used at a time, the others being entirely disconnected. 
This method has been used for elevator motors, traction motors and also for small motors driving machine tools. 
Speeds in the ratio of 2:1 can be produced by a single winding if wound on the consequent-pole principle. In that case, each of the two stator windings can be connected by a simple switch to give two speeds, each, which means four speeds in all. For example, one stator winding may give 4 or 8-poles and the other 6 or 12-poles. For a supply frequency of 50-Hz, the four speeds will be 1500, 750, 1000 and 500 r.p.m. Another combination, commonly used, is to group 2- and 4-pole winding with a 6- and 12-pole winding, which gives four synchronous speeds of 3000, 1500, 1000 and 500 r.p.m. 

(d) Rotor Rheostat Control 
In this method, which is applicable to slip-ring motors alone, the motor speed is reduced by introducing an external resistance in the rotor circuit. For this purpose, the rotor starter may be used, provided it is continuously rated. This method is, in fact, similar to the armature may be used, provided it is continuously rated. This method is in fact, similar to the armature rheostat control method of d.c. shunt motors. 


It is well known that near synchronous speed (i.e. for very small slip value), $\displaystyle \small \mathrm{T \alpha \frac{s}{R_2}}$ . It is obvious that for a given torque, slip can be increased i . e speed can be decreased by increasing the rotor resistance $\displaystyle \small \mathrm{R_2}$ . 
One serious disadvantage of this method is that with increase in rotor resistance, R losses also increase which decrease the operating efficiency of the motor. In fact, the loss is directly proportional to the reduction in the speed. The second disadvantage is the double dependence of speed, not only on $\displaystyle \small \mathrm{R_2}$  but on load as well. 
Because of the wastefulness of this method, it is used where speed changes are needed for short periods only. 

(e) Cascade or Concatenation or Tandem Operation
 In this method, two motors are used  and are ordinarily mounted on the same shaft, so that both run at the same speed (or else they may be geared together). The stator winding of the main motor A is connected to the mains in the usual way, while that of the auxiliary motor B is fed from the rotor circuit of motor A . For satisfactory operation the main motor A should be phase-wound i.e. of slip-ring type with stator to rotor winding ratio of 1:1, so that, in addition to concatenation, each motor may be run from the supply mains separately. There are at least three ways (and some-times four ways) in which the combination may be run. 


1. Main motor A may be run separately from the supply. In that case, the synchronous speed is $\displaystyle \small \mathrm{N_{sa} =\ \frac{120f}{P_a}}$  where $\displaystyle \small \mathrm{P_a}$  = Number of stator poles of motor A . 
2. Auxiliary motor B may be run separately from the mains (with motor A being disconnect.). In that case, synchronous speed is $\displaystyle \small \mathrm{N_{sb} =\ \frac{120f}{P_b}}$  where $\displaystyle \small \mathrm{P_a}$  = Number of stator poles of motor B. 
3. The combination may be connected in cumulative cascade i.e. in such a way that the phase rotation of the stator fields of both motors is in the same direction. The synchro-nous speed of the cascaded set, in this case, is N„ = 120 fl(Pa+ 

When the cascaded set is started, the voltage at frequency/ is applied to the stator winding a machine A. An induced e.m.f. of the same frequency is produced in rotor A which is supplied to auxiliary motor B. Both the motors develop a forward torque.A the shaft speed rises, the rotor frequency of motor A falls and so does the synchronous speed of motor B. The set settles down to a stable speed when the shaft speed becomes equal to the speed of rotating field of motor B.
Considering load conditions, we find that the electrical power taken in by stator A is partly used to meet its and core losses and the rest is given to its rotor. The power given to rotor is further divided into two parts one part, proportional to the speed of set i.e. 'N' is converted into mechanic power and the other part proportional to $\displaystyle \small \mathrm{N_{sa}-N}$ is developed as electrical power at the slip frequency, and is passed on to the auxiliary motor B, which uses it for producing mechanical power and losses. Hence, approximately, the mechanical outputs of the two motors are in the ratio $\displaystyle \small \mathrm{N:(N_{sa}-N)}$ . In fact, it comes to that the mechanical outputs are in the ratio of the number of poles of the motors.

(f) Injecting an e.m.f. in the Rotor Circuit In this method, the speed of an induction motor is controlled by injecting a voltage in the rotor circuit, it being of course, necessary for the injected voltage to have the same frequency as the slip frequency. There is, however, no restriction as to the phase of the injected e.m.f. When we insert a voltage which is in phase opposition to the induced rotor e.m.f., it amounts to increasing the rotor resistance, where as inserting a voltage which is in phase with the induced rotor e.m.f., is equivalent to decreasing its resistance. Hence, by changing the phase of the injected e.m.f. and hence the rotor resistance, the speed can be controlled.. 

 


 One such practical method of this type of speed control is Kramer system, as shown in figure above, which is used in the case of large motors of 4000 kW or above. It consists of a rotary converter which converts the low-slip frequency a.c. power into d.c. power which is used to drive a d.c. shunt motor, mechanically coupled to the main motor. The main motor is coupled to the shaft of the d.c. shunt motor. The slip-rings of main motor are connected to those of the rotary converter. The d.c. output of rotary converter is used to drive d.c motor. Both rotary converter and d.c motor are excited from the d.c. bus-bars or from an exciter. There is a field regulator which governs the back e.m.f. , of d.c motor and hence the d.c. potential at the commutator of rotary converter which further controls the slip-ring voltage and therefore, the speed of main motor. One big advantage of this method is that any speed, within the working range, can be obtained instead of only two or three, as with other methods of speed control. Yet another advantage is that if the rotary converter is over-excited, it will take a leading current which compensates for the lagging current drawn by main motor and hence improves the power factor of the system.


 In the figure above is shown another method, known as Scherbius system, for controlling the speed of large induction motors. The slip energy is not converted into d.c. and then fed to a d.c. motor, rather it is fed directly to a special 3-phase (or 6-phase) a.c. commutator motor-called a, Scherbius machine. The polyphase winding of machine C is supplied with the low-frequency output of machine M through a regulating transformer. The commutator motor C is a variable-speed motor and its speed (and hence that of M) is controlled by either varying the tappings on regulating transformer or by adjusting the position of brushes on C.

Comments