Induction Generation: an exciting possibility Paul Cunningham Why does it make a difference what type of generator we use to produce power? Let's take a look at the standard types and see what the features are. Two broad categories include most types. Either the output coils can rotate or they can be stationary. Almost all of the older designs used output coils of wire that rotated. These designs used a stationary "field" which provided a magnetic flux for the moving output coils to pass through which in turn generated an electrical flow in the coils. This design is represented by direct current (DC) motors and most older alternating current (AC) generators (alternators). The major disadvantage of this type of machine is that the full output must pass through carbon brushes. Many generators of this type are used in alternative technology applications but they require more maintenance. Also, because of the rotor design, the wire is more difficult to retain at higher speeds as it tries to fly outward from the rotor. It is for these reasons that automotive generators (DC) were replaced by alternators. The other major category of generators include those designs in which the output coils are stationary and the field rotates. This includes automotive alternators. All machines of this type produce alternating current output. If DC output is required, then RECTIFIERS are used to convert AC to DC. These are solid state electrical one way "valves" usually using silicon diodes. Thus far, all of these designs mentioned could use permanent magnets for the field. This means several things. The field requires no electricity to operate, so efficiency is higher. It can operate at very low speeds since the power of the field is not taken from the output of the machine. On the negative side, there is no easy way to control the output of such a machine. With a wire wound field the output can easily be varied by alternating field current. A rheostat is a simple way to do this, and in this way output is easily optimized. EXCITATION IS WHAT AN INDUCTION GENERATOR IS ABOUT You can use most motors as generators to produce electric power. A standard induction motor can also be used this way. These motors consist of stationary coils of wire that carry the current to operate them wound through slots in steel laminations. The rotor consists of steel laminations with aluminum conductors (usually) cast into slots in the steel. These are called squirrel cage rotors. When alternating current is applied to the stator coils, a rapidly changing magnetic field is produced. Once such a machine is running, there is always a speed difference between the rotor and changing field in the stationary coils. This difference is called "slip". This difference in speed INDUCES an electric current an electric current in the rotor and as a result a magnetic field. It is this field in the rotor that now causes it to "follow" the direction of the field in the stator. For quite some time it has been recognized that if shaft power were applied to an induction motor already running, it would operate as a generator and push electricity back into the source used to operate it. For this to occur, our motor must now be running slightly faster than the "synchronous" speed instead of slightly slower. This technique is widely used on a large scale in commercial power generation systems. The electrical power already present provides the necessary "excitation" to correctly operate the machine. In this context, the system is fail safe.... if the grid power fails, generator output ceases also. How is all of this going to help us with our stand alone remote system? There is the possibility of using a standard electric motor to efficiently generate electricity. One technique is to generate an "exciting" current for the motor/generator to "follow". Induction seduction, sort of. I have not been successful with this. Anyone who has should contact me with their findings. What DOES work with excellent results is to simply apply capacitance in parallel with the output lines. I ignored this tantalizing possibility until I met Bill Thomson and Fred Howe (of Thomson and Howe, Kimberly, B.C., makers of electronic controllers for hydro systems) at a small hydro conference in March '87. It was their encouragement and information that enabled me to progress. The simplicity, low cost, and high efficiency of such a system were all self evident, once work was begun in this direction. In the first issue of Home Power, I wrote about the conversion of a standard three phase induction motor to a permanent magnet alternator. With my new information, I removed the P.M. rotor and replaced it with the original. Then I added the 15 microfarad capacitors across each line (parallel). When the machine was started again, I found that not only did it start generating by itself (yes, "self excitation" an interesting term for a dry subject) but the output was identical to the P.M. rotor machine. This was a revelation to me.... how easily it could be done. It should be instructive to note what makes up a complete battery charging system. The water driven turbine in turn drives a 1/3 H.P. three phase 230 VAC motor that has the three capacitors connected across the output lines. In this case power is generated at 120 VAC and can thus be transmitted very long distances with minimal losses. Then at the point of use three transformers step the 120 volts down to battery voltage and with a bridge rectifier, produce direct current. It looks like this: INSERT ILLUSTRATION You are probably wondering how induction generation works and why it isn't more widely used. In a stand alone system, the key to operation is the presence of capacitance. This gives electricity somewhere to "go" without the capacitors acting as a load. Thus enabling current to flow in the motor and get it all excited. Most motors I have tested as generators will start producing power on their own with the use of capacitors. This is due to the small residual magnetism in the rotor. It is also necessary that the generator not "see" a load until it is up to proper voltage. If a load is present at the start, the voltage will be unable to rise at all. In a battery charging system this is more or less inherently provided for, as the generator only "sees" transformers as a small load until proper voltage is reached. Induction generation is more limited than a P.M. alternator in the type of situation in which it can be used. The induction machine should be operated at or near its rated speed. This can be as low as 800 rpm depending on the motor specs. A P.M. machine can be operated at very low speeds and still work well. However, if a site can use an induction generator, then it can be implemented at low cost since the motors are not expensive and the capacitors are only a few dollars each. Motors are also available in different speed ranges. You might wonder why I am using three phase systems when a single phase one might do. It is possible to use single phase motors for this. However, they require more capacitance, operate at lower efficiency, and are not easily excited. Three phase alternating current is also more efficiently converted to DC for battery systems. For those of you wishing to experiment, some further information may prove useful. The size of capacitor will largely control output voltage. Smaller capacitors are needed as voltage rises. Use only AC motor run capacitors. Not all electric motors are created equal and may produce results differing from what I found. Also keep in mind that if the system is to operate at a fairly fixed speed (like most hydro systems) that no adjustments are required from minimum to maximum output. As a starting point, a 1/3 HP 3 phase 230 VAC 4 pole (1800 rpm nominal) Westinghouse motor needs 15 µf. per line to generate 120 VAC at 1500 rpm. A 1 1/2 HP Leeson 3 phase 230 VAC 4 pole motor requires 40 µf. per line at 1500 rpm, 230 VAC. If any readers have trouble getting things to excite, the most effective technique is to apply 12 VDC to one phase (two output wires) of the motor while stopped. After a few minutes remove the DC and try starting again. This "imprints" the rotor with magnetic poles and should get things going. Try no load at first just to see if it works. There are some further points of interest that will probably be discussed in a future update. Presently there is still much work to be done before a more complete understanding is possible. Readers are encouraged to both try experiments and report their results.