Electric World

We live in a world that is totally dependent upon technology. The production and control of electricity is the basis of, or necessary for most other technologies. An automobile is a product of mechanical engineering not electrical engineering, but the ignition of the fuel is done by electricity. The human genome project is biological technology, but it cannot be done without computers which require electricity. So what is electricity really?

To understand electricity you need to understand the basic structure of matter. Benjamin Franklin began experimenting with electricity in the form of lightning more than 200 years ago. Franklin understood there were two charges which he called positive and negative. Franklin designated the positive charge as the one that moved. It was shortly after 1910 that scientists figured out the basic structure of the atom, although the neutron was not discovered until 1932. Almost all of the atom's weight is concentrated in the center called the nucleus. The nucleus is made up of protons and neutrons, the protons have a positive charge and the neutrons have no charge. Protons and neutrons have the same weight. Electrons move in the space around the nucleus. This movement is usually portrayed like the orbit of a planet but an electron does not move like a planet. The electron is so small and so fast it can't be located and is regarded as having a probability of occupying a region of space. That space can have a spherical or a more complex shape.

Every element has a unique type of atom. Hydrogen is a gas and the lightest element. It consists of a single proton in the nucleus and a single electron balancing tha positive charge. Heavier elements have neutrons in addition to protons in the nucleus and either the same number of neutrons as protons or a few more neutrons. Protons and neutrons have about the same mass and the weight of the atom affects the behavior of the element. Different elements have different states at 'normal' temperatures, be it liquid, solid or gas. The elements that serve our purposes for manipulating electricity are mostly solid metals which allow their electrons to move around easily. Mercury is a strange liquid metal which allows for some unique applications. These metals are called conductors. The most commonly used are copper, silver and gold. These metals appear in a column on the periodic table of elements so they have many similar characteristics. One difference that is of great practical importance is rarity and therefore price. Silver is much more expensive than copper so even though it is a better conductor it is only used in special circumstances. Gold is even more expensive than silver but is used more in electrical applications because of its resistance to corrosion. A thin layer of gold is often put on contacts in electrical equipment. So electricity is the flow of electrons, usually through a solid conductor. If the VOLTAGE is high enough electrons will travel through air or vacuum. CRTs fire a beam of electrons through a vacuum to form a picture on most computers. Static discharges which create pops and lightning which creates thunder are the same thing, only slightly different in scale.

The atoms of a conductor must be relatively free with their electrons. The electrons occupy regions known as shells around the nucleus. Each shell can hold some maximum number of electrons and the next larger type of atom starts a new shell. The electrons in the outermost shell are called valence electrons. These electrons are most free to move or bond with another atom. Molecules are groups of atoms linked together at these valence electrons. Copper has one valence electron and silicon and germanium have four. The first transistors were made of germanium but it was later found that silicon worked much better. Silicon and germanium are used to make transistors by deliberately putting impurities in the crystal lattice. Atoms with either three or five valence electrons are inserted. Since each silicon or gremanium atom will bond with four other atoms an impurity with five valence electrons will leave a loose electron in the matrix. While an atom with three valence electrons will leave a gap. These impurities create N-type and P-type semiconductor material respectively. I mention this about transistors because they are so important to how the world works but the workings of transistors will have to wait.

A closely related and equally mystifying phenomenon to electricity is magnetism. It turns out you cannot have one without the other. To get electrons to flow through a conductor you must apply EMF, ElectroMotive Force, commonly known as voltage. This voltage can be produced by spinning a coil of wire in a magnetic field. This is what generators do. It is just a matter of what source of mechanical power is used to turn the generator. Water in a hydroelectric dam or wind pushing the blades of a windmill can provide clean mechanical power. Alternatively coal, oil or nuclear energy can produce steam to spin a turbine attached to the generator implementing a somewhat dirtier power source.

If the coil of wire in the generator is rectangular and spins around an axis parallel to the sides of the coil and the magnetic field is perpendicular to the axis then the angle at which the sides cut through the magnetic field changes continuously as the sides rotate around the axis. This produces an alternating voltage at the contacts of the generator. When one side of the coil is moving perpendicular to the field in one direction the other side of the coil is moving in the opposite direction. At this point maximum voltage is generated. When the coil rotates 90 degrees the sides are moving parallel to the field and zero voltage is produced because the coil is not crossing the field. When the coil turns another 90 degrees maximum voltage is produced again but since the sides of the coil have reversed direction the voltage is reversed. The voltage follows the path of a sinewave, rising and falling in time with the rotation of the generator. The AC, Alternating Current, in the United States cycles 60 times per minute or 60 Hertz. The voltage should be about 110 volts at an outlet. Some countries use 50 hertz and 220 volts.

Although Thomas Edison tried to promote the use of DC, Direct Current, power distribution Alternating Current won out because of an irrefuatble advantage. Transformers can be used to step up the voltage to very high levels and the current can be sent long distances with little loss and then stepped back down at the destination. A transformer does not give you something for nothing. You cannot get more power out than you put in. When the transformer steps up the voltage the current goes down. Twice as much voltage, half as much current, same amount of power. When electricity flows through wire some loss occurs due to the resistance of the wire. The longer the wire the greater the loss. How much loss depends on the amount of current, so if current can be reduced then losses can be cut. It turns out the loss is the square of the current. Cutting the current in half reduces the losses to one fourth the original value. So power companies use thousands of volts to get current very low and send electricity hundreds of miles. Mr. Edison was trying to sell more generators since the voltage change trick cannot be done with DC.

A transformer works its magic by converting electricity to magnetism and back into electricity. The transformer uses an iron core to channel the magnetic field. In the diagram, power from the generator is sent to the input coil of the transformer. The magnetism in the generator induced electricity in the coils of the generator, so the electricity fed into the input coil of the transformer induces a magnetic field inside the core of the transformer. This magnetic field expands and shrinks and reverses and expands and shrinks and reverses in time with the spinning of the generator. This cycling magnetic field cuts through the output coils of the transformer reproducing the effect in the generator with a changing field on a stationary coil. This process of the same physical principal being manifested in a slightly different manner repeats itself again and again in electricity and electronics.

The transformer in the diagram has a CENTER TAP. There is an extra electrical connection to the output coil exactly in the center. The voltage at the center tap must always be halfway between the ends. If the center is used as the reference point then ends have equal but opposite voltages, sine waves 180 degrees out of phase. Electric power is fed into American homes this way. Two 110 volt lines and a neutral. The 110 lines are out of phase so the difference is 220. When an electrician installs a 220 line for an air conditioner the two 110 lines are run to the outlet. The normal outlets have a 110 line and the neutral. Approximately half of the electrical outlets in a home are on one 110 line and the rest are on the other. The current on the neutral is minimized.

Some commonly used electrical devices can use 110 volt AC straight from the outlet. Simple devices like light bulbs, toasters, vacuum cleaners and hair driers. The more complex things that have been turning up since World War II are more picky about their electrical cuisine. Most require Direct Current and the voltages may have to be at a particular level for that device. These devices either have built in power supplies or a little box that plugs into the outlet, called a transformer, of all things. Power converter would probably be more accurate but power supply is the accepted name. This conversion from AC to DC takes us back to the silicon atom. The device used to accomplish this feat is called a diode and are almost exclusively silicon these days.

As discussed earlier the silicon atom has four valence electrons. So each silicon atom can bond with four other atoms. In a silicon crystal used for manufacturing electronic components there is nothing to bond with but other silicon atoms. The figure shows a two dimensional representation of the silicon crystal. Si is the abbreviation of for silicon and although the atom has 14 electrons only the four valence electrons are displayed. The diagram shows the atoms at 90 degree angles to one another but in a real crystal it would be closer to 140.

Impurities are deliberately inserted into the silicon to make a useful electronic component. In the next figure arsenic, As, atoms are displayed in pale blue and have five valence electrons. So by bonding with four silicon atoms each arsenic atom leaves an extra electron loose in the crystal matrix. If the electron wanders away from its atom then this leaves a stationary positive charge held in the crystal which attracts the electron. It would take something more attractive to keep the electron away. On the other side of the crystal Indium, In, atoms have been added and are shown in green. Indium has only three valence electrons and therefore leaves a gap in the matrix. This is displayed with a circle and is called a "hole" in semiconductor parlance. If an electron moves into this hole then the indium atom takes on a negative charge and some other atom must be missing an electron. If it is a silicon atom then it must have a positive charge. An atom with an unbalanced electrical charge is called an "ion."

Interesting things happen at the boundary between these two types of semiconductor. The side "doped" with the arsenic is N-type silicon and has extra electrons. The indium doped side has a shortage of electrons called holes. At the boundary some of the electrons from the N side migrate across and fill the holes on the P side. This leaves arsenic ions on the N side and creates indium ions P side. The arsenic ions are positive and the indium ions are negative. The valence bond holding the electron to the indium atom is strong enough to resist the positive charge of the arsenic ion. But these two layers of ions create an electric field across the P-N junction. Once the electric field forms the migration of electrons from the N side to the P side stops. This electric field has the strength of about half of a volt. The field is portrayed by the yellow arrows.

This single P-N junction forms a diode. If a positive voltage is applied to the N side and a negative voltage to the P side then the charge carriers, holes and electrons are attracted to the electrodes. This pulls them away from the P-N junction, called the depletion zone, increasing its size. More ions form at the junction counteracting the externally applied electric field. The flow of current is blocked. If the polarity is reversed and negative applied to N and positive to P then the charge carriers are propelled to the P-N junction and engage in mutual annihilation. Since holes are nothing but places without electrons then electrons are simply flowing in the opposite direction of the holes and electric current is flowing through the diode. So current is blocked in one direction and passed in the other.

The voltage from the transformer is AC with the polarity constantly changing. A circuit that cures this ailment is called a rectifier. The circuit shown is called a bridge rectifier or a full wave rectifier. A half wave rectifier is cheaper to build but wastes potential power by blocking out the entire negative cycle of the sine wave and will have more "ripple" because of that missing half. Four diodes are used in the bridge while the half wave can use one. During the positive, red, side of the cycle the current flows in the direction of the red arrows through two diodes. No current flows through the other two diodes. During the negative, violet, portion of the cycle the diodes with the violet arrows are used. So coming off the rectifier we have two positive peaks instead of a positive and negative peak. This still isn't good enough for most electronic circuitry. The valley between the peaks is too low and the device will malfunction. This introduces the need for another component. Something that stores and filters electricity, the capacitor.

A capacitor consists of two conductive plates placed close together usually seperated by an insulator called a dielectric. The larger the plates and the closer the spacing the greater the capacitance. In most electronic equipment the largest capacitors are used in the power supply for filtering. I have a 120 watt amplifier that uses 4 10,000 microfarad capacitors for filtering. The peculiar unit farad is named after Faraday who was among the earliest electrical experimenters. The insulator used between the plates affects the capacitance and polar characteristics of the capacitor. Some capacitors must be positive on one end and negative on the other, if installed backwards they can short circuit and explode. A microfarad is a millionth of a farad and 100 10,000 microfarad capacitors would be required to yield one farad. Many capacitors are much smaller than this and measured in picofarads, billionths of a farad.

In the circuit shown the capacitor gets charged up to the peak voltage of the rectifier. When the voltage from the rectifier drops the capacitor begins to discharge, supplying current and maintaining significant voltage. The voltage from the capacitor gradually drops but the rectifier supplies another peak before there is significant loss from the capacitor. The voltage flucuation in this peak-discharge cycle is called ripple. A half wave rectifier would have a gap equal to half the length of the sinewave. A capacitor would have much more time to discharge and would fall to a lower voltage than on a full wave rectifier design. Many circuit designs are based on the lowest cost of manufacture. What is the price of diodes and capacitors? How much power does the device need? How dirty a power supply can it withstand and still function correctly? How long should it last and what failure rate will we tolerate? The customer is not supplied with some of the answers to these questions and would not know to ask them.

We have discussed coils in transformers and capacitors but there is one other basic component which cannot be ignored and that is the resistor. The trouble with this component is that it takes so many forms and may not be regarded as a resistor. The filament in a light bulb is a resistor. Its purpose is to give off light not to resist the flow of electricity but it does that in the process of performing its function. The resistance of the filament changes with temperature, low when cold and rising as it gets hot. The same applies to the heating elements in a toaster but the light given off by them is the unnecessary byproduct. Coils and capacitors do something to electricity that resistors do not. They change the phase of the voltage and the current and this phase will be frequency dependent. A coil behaves as if it has electrical inertia. Applying DC voltage across a coil causes the current to build up as the magnetic field is created and the voltage drops as the current goes up. A capacitor acts like a short circuit to the DC voltage but the current decreases and the voltage increases to maximum when the current stops completely. So voltage leads current in a coil and current leads voltage in a capacitor. In a resistor voltage and current are in phase. RLC circuits can exhibit very complex behavior and the math can get hairy. R is for resistance, C is for capacitance and L is for the inductance of coils and I don't remember why they use L unless it is to avoid using I which is used to represent current.

I have avoided using mathematics so far because it is possible to understand important electrical concepts without mathematics and many people are turned off by mathematics. Most of the time electronics technicians don't need to use math. They need to understand how circuits work to track down bad components and replace them. The engineers had to know math to select the components to get the desired results.

This is some of the most basic math involved with electricity. First there is Ohm's Law:

I = E/R

E = I * R

R = E/I

These three equations are merely algebraic variations of the same thing. E is voltage, I is current in amperes or amps and R is resistance measured in ohms. Which equation is used depends on what is known and what is sought. Voltage is usually easy to measure while a circuit is operating. Resistance can be measured but the circuit would have to be turned off and if there are parallel paths in the circuit the measurement could be wrong. Measuring current would require modifying the circuit to insert the meter in series.

The power equation is:

P = E * I

since

E = I * R

then

P = I * I * R

or

P = I ** 2 * R

This form of the power equation explains why reducing current accounts for the advantage of using transformers to step up voltages for power transmission. It also accounts for why those losses are referred to as "I squared R losses." Since power lines get warmer during the summer and resistance goes up it also explains why power companies raise bills during the summer due to significant energy loss.

Electric circuits can be extremely complicated but mostly they are combinations of series and parallel subcircuits. Resistors are measured in ohms which are represented by the letter omega from the Greek alphabet. The upper line from the power supply has two resistors in series, a 100 ohm and a 200 ohm. All of the current flowing through the circuit must go through both of these resistors. The circuit then splits and the 300 ohm and 150 ohm resistors are in parallel. So how do you figure out how much current is flowing through the circuit? Ohm's law is I = E/R. E is 120 volts so if we can compute R then we can compute I. The hard part is figuring out what to do with the parallel resistors. The voltage across the the 300 and 150 resistors must be the same so I = E/R means that twice as much current must be flowing through the 150 ohm resistor as is flowing through the 300 ohm resistor. If we can compute the total resistance of the two in parallel then the rest would be a snap. The formula for parallel resistors is:

1/Rt = 1/R1 + 1/R2 + 1/R3 + ...

The reciprocal of the total resistance is equal to the sum of the reciprocals of all of the parallel resistors. In this case:

1/Rt = 1/300 + 1/150

therefore

1/Rt = 1/300 + 2/300

1/Rt = 3/300

1/Rt = 1/100

So the two parallel resistors are equivalent to a 100 ohm resistor. Since these numbers were selected to be easy, a quick and dirty method would be a 150 ohm resistor is equal to two 300 ohm resistors in parallel. Three 300 ohm resistors are equal to one 100 ohm.

So this circuit is equivalent to a 100, a 200 and a 100 ohm series of resistors. Resistors in series just add to the total resistance so this is like a 400 ohm resistor. Ohm's Law indicates that the current will be 120/400 which is 0.3 amps. Standard electrical jargon would call that 300 milliamps. 300 thousandths of an amp. The voltage drops as current flows through each component around the circuit. The voltage drop at each component is E = I * R, another arrangement of Ohm's law. So the voltage drop from A to B is 100 ohms times 300 milliamps which is 30 volts. The next resistor will be twice as much, 60 volts, and a final 30 volt drop across the parallel resistors.

Resistors can be connected in very complex networks which can be difficult to analyze but these usually only exist in books supplying students with idiotic busywork. Things get complicated and useful when you put capacitors and coils into the circuits. Capacitors and inductors react to electricity rather than passively and uniformly resisting the flow, like resistors. This is called capacitive reactance and inductive reactance. Many circuits mix inductors and capacitors and the total opposition to the flow of current is called impedance and is represented by a Z, although measured in ohms just like resistance. The complication sets in when you must deal with the fact that this impedance varies with frequency. The Impedance of capacitors decreases as frequency increases while that of inductors increases with frequency, these are desirable characteristics in many applications.

The circuit shown has a capacitor and inductor in series and the adjacent graph represents the impedence. The reactances of the capacitor and the coil will cross at some point as frequency increases and this point is the RESONANT FREQUENCY. This is comparable to a tuning fork that vibrates at a particular frequency. Capicitors were used in the old style radios before digital circuitry took over. Turning the tuning knob rotated the plates in a variable capacitor changing the capacitance and the resonant frequency. That frequency is going to be the broadcast frequency of the station being received. In old tuners there are adjustable coils which have to be tuned but only electronic technicians are supposed to adjust these since special equipment is needed. But what happens after the station is tuned in?

The radio signal has to be amplified and sent to a speaker. What happens in the speaker is the province of inductors and capacitors also. A speaker shows us another use for electromagnetism. The solid blue portion of the speaker crossection is the magnet. The magnetic field has to be concentrated to maximize efficiency. The magnetic field guides concentrate the field into a narrow circular gap where the voice coil is suspended. As AC current is sent through the voice coil it produces an alternating magnetic field that pushes and pulls the the cone so the vibrations reproduce the original sound. The cone is held suspended in a metal frame by a surround at the outer edge and a flexible spider near the voice coil. This arrangement needs considerable precision since the gap for the voice coil is less than an eighth of an inch and the voice coil must not scrape the sides or noise and distortion will result.

The pictured speaker is most like a midrange or the woofer in a small two-way speaker. A true woofer would have a larger magnet and a bigger, heavier cone. The surround would have curves or ridges to allow for a great degree of movement. A tweeter would have a completely different shape. It might just be a dome over the voice coil. These different shapes are necessary to cover the full audio spectrum. A woofer needs to move large volumes of air to produce low frequencies at reasonable volumes. The lower the frequency the more difficult this becomes. The human audible spectrum is commonly regarded as 20 cycles per second, or Hertz, to 20,000 Hertz. Small speakers commonly drop off by 50 Hz. Getting to 35 Hz or lower requires significant size and cost. A good cabinet must be rigid to prevent resonances so the cabinet should be 3/4ths of an inch thick and the labor to build it will cost. A good 12" woofer can cost $100 alone. If a pair of 3-way speakers retails for less than $500 I'm inclined to wonder how "good" they are.

Even though woofers, midranges and tweeters are optimized to produce certain frequencies They can still produce sound outside that range. A woofer designed to work best below 400 Hz can still produce 600 Hz. If that frequency is supposed to be handled by the midrange then the woofer must be prevented from producing it. This frequency control can be done by inductors and capacitors. The circuit diagram of the 3-way speaker system shows an LC crossover network distributing the proper frequencies to the 3 speakers. A real speaker system would probably include resistors to make the midrange and tweeter play at the proper levels. It is not unusual for the high frequency speakers to be user adjustable in relation to the woofer. The values of the inductors and capacitors are not shown since they would vary with the speaker design and this diagram is only intended to communicate the concepts.

The inductor L1 blocks high frequencies from access to the subwoofer. The correct term is "attenuates" since some high frequencies still get through, just less and less as the frequency goes up. How much attenuation at which frequency is determined by the value of the inductor, which is calculated by the speaker designer. To further attenuate the high frequencies reaching the woofer, capacitor C1 acts as a short circuit for high frequency current. Similar principles with opposite selection applies to the tweeter circuit. C4 blocks low frequencies and L4 shorts them out. Both of these processes apply to the midrange system. L2 and C2 work like L1 and C1 but operate at different frequencies. C3 and L3 behave like C4 and L4. The midrange is isolated from the low frequencies intended for the woofer and the high frequencies intended for the tweeter.

A major part of the performance of a good piece of audio equipment can depicted in the frequency response. The most accurate method of specifying the frequency response of a device is with a graph. This will show all of the bumps and wiggles. The response is often expressed like 35-18,000 + or - 3 decibels. This specifies a lower limit of 35 Hz and an upper limit of 18,000 with the proviso that no frequency within that range goes outside of a 6 decibel volume limit. The trouble with expressing frequency response this way is that there are an infinite number of curves that could satisfy this criterion. Some curves would sound very different from others. Some people can hear variations as small as 1 decibel so a graph is best. Human hearing works logrithmically. The graph shows the distance from 1,000 to 10,000 Hz is the same as the distance from 100 to 1,000 Hz, ten times as many frequencies in the same space. Music is in the same form, it must match human hearing after all. An octave is a doubling of frequencies with 8 notes within each octave. The low keys on a piano have about a 3 Hz difference between them. In the middle of the piano the difference is 20 Hz and at the top it is 230 Hz. So the compression in the graph makes sense compared to instruments and human hearing.

The vertical scale is in decibels. A decibel is a tenth of a bel, named after Alexander Graham Bell. 3 decibels is doubling of sound energy and therefore volume. A 10 decibel increase is a 10 times gain in volume. The graph of the 3-way speaker shows the responses of the crossover network, the power delivered to each speaker over the frequency range. The woofer is flat at +1 decibel from 15 to about 180 Hz, from there it drops to -2 at 300 Hz. The midrange is also at -2 db at 300 Hz, this is no accident. Since both the midrange and woofer are 3 db below their normal output, the two sound sources combine in acoustic space and fill the gap, maintaining the +1 db level. Getting these crossover points right is critical to good speaker design. The same thing happens between the midrange and the tweeter at 5,000 Hz. It is after the speaker turns the electrical power into sound that things really get complicated.

The electronic reproduction of sound can be traced to the invention of the audion tube by De Forest in 1907. The audion made radio possible and record players that didn't just use a horn to amplify mechanical vibration. The first serious attempts to reproduce lifelike music in the home began in the 50's. The development of tape recording during WWII made it possible to record reasonably good sound which could then be pressed onto records. Rock and roll of the 50's and stereo LP's and American Bandstand kicked off the music and media driven electronics world we have today. The amazing thing is that most people could get higher quality sound with a little knowledge and a little effort and it might even cost less in the long run. 8-tracks were among the most egregious offenses to acoustic excellence in known space-time. I believe these devices were designed to destroy their tapes while torturing their listeners with wow and flutter. Wow and flutter is sound distrotion due to speed variation in the reproduction device, wow is slow speed variation and flutter is high speed variation. 8-tracks had an infinite loop design which often had uneven tension and the pinch rollers in some cartridges were plastic instead of rubber so those tapes were worse than usual. The infinite eventually ran out and the tape broke or bound up or came out of the cartridge. I assume manufacturers loved this because of the limited cartridge life. Cassettes were vastly superior some still sound decent after 20 years.

Cassettes are analog technology. This means that what is recorded on the tape has a similar, or analogous, form to the original sound. Sound consists of alternating pressure waves moving through the air. Continuously changing pressure waves at multiple frequencies cause the eardrum to vibrate resulting in the detection of the sound. Thomas Edison made the first sound recordings by using his voice to vibrate a needle which scratched an analogous track into metal film. By having the needle follow the track and vibrate a diaphragm, a similar sound could be reproduced. Peaks and valleys in the track would correspond with high and low peaks in the sound wave. A paper cone with a straight pin from the tip can be used to play back an LP duplicating Edison's technique. A cassette recorder creates a magnetic analog of sound on tape. The record head uses a coil to produce a fluctuating magnetic field as the tape rubs against the head. This leaves a varying magnetic track along the tape. On playback the tape creates a voltage in the coil of the playback head, this voltage is amplified and reproduces sound. These analog recording techniques have characteristic flaws which need to be compensated for but cannot be totally overcome.

Turntables and cassettes have problems with wow and flutter just like 8-tracks but since turntables have only one rotating component determining the smoothness of the sound it is rarely a problem and easy to locate when defective. Cassettes can be much more problematic. There are four factors that can affect the playback speed of a cassette. The capstan is the rotating metal pin pulling the tape across the heads and can be bent, dirty or have some propulsion problem. The pinch roller that squeezes the tape against the capstan can be dirty, dented or hard from age. After the tape comes from the capstan it must be picked up on the take up reel. If the take up tension is too high this can cause speed problems and if too low the tape can wrap around the capstan or pinch roller. Finally the supply reel tension could be too high causing excess drag. This is more complex than an 8-track but not inevitably destructive because the tape doesn't twist and slide over itself. The transport mechanism has more reliable control in a cassette and the playback head doesn't change tracks, so head allignment is more reliable. The head in an 8-track could move to four different positions to play the two tracks of stereo. This added vibration reduced the certainty that the head would be in the perfectly vertical position for proper high frequency response.

Analog audio equipment has numerous problems to contend with. LPs get scratched and are laborious to keep clean. The stylus causes wear on the disk simply by playing it. Good turntables only put about a gram of force on the record but this is concentrated on a very small point so the pressure is considerable. The stylus must then vibrate at 18,000 cycles or more so there is significant stress on the plastic. It is amazing that it works as well as it does. The record players most people used didn't deliver great performance and caused much more wear. Cassettes could give good sound but many people bought low quality tapes and the oxide would build up on the heads and reduce the sound quality unless removed. Tape recording uses a high frequency bias that needs to be optimized for the tape formulation being used to give best performance and most users know nothing about this so cassettes didn't live up to their potential. The huge market for prerecorded cassettes is another source of underachievement. These tapes are recorded at high speed and don't use the best tape so the sound quality isn't as good as can be achieved by a home recording with little effort. A comparison of a prerecorded tape to a recording from an LP indicated that the performance of the factory tape didn't come close. The home made tape had vastly superior high frequency response. This test was performed 20 years ago. A tape made from a CD today would be better than one from an LP.

Escape from analog confinement began in the late 70's with the introduction of CDs. This became possible because of the increasing ability to put more transistors on a single integrated circuit reducing the cost and increasing the power of the "chips." Transistors can operate in either analog or digital mode. A transistor has an input, an output and a control. On a bipolar transistor these are the emitter, the collector and the base. A bipolar transistor can be either N-type semiconductor material sandwiched between two P-types or P-type between N-types. These are called PNP or NPN bipolar transistors. An entirely different style of transistor, the Field Effect Transistor or FET, operates on somewhat different principals but performs the same input, output, control tricks. The electrical connections are called source, drain and gate on the FET.

In digital mode a transistor is either all of the way ON, or all of the way OFF. When ON the maximum possible current is going through the transistor, this is called "saturation." When off almost no current is flowing. The control signal goes in the base or gate which controls the current going to the collector or drain. The time required to switch from ON to OFF or vice versa is the "switching speed" and can be billionths of a second. Transistors usually use less power in digital operation than in analog. Since power is voltage times current then if either voltage or current is very low then power dissipation will be low. In saturation, current is high but voltage is low, in cut off, voltage is high but current is low so both states are low power. Things get more complicated in analog mode.

When operating in analog configuration the signal at the base is amplified at the collector. Idealy this amplification should be perfectly linear, the percentage of increase at the collector should be exactly the same as the percentage of increase at the base. A transistor has a relatively linear region in analog mode but becomes more nonlinear as saturation or cutoff are approached. The circuit designer must anticipate the signal strength and make circuit handle the input. The same applies to an output like a speaker load. This becomes an exercise in probability since some users will push the equipment closer to the edge than others. Of course some will go over the edge. Amplifiers have been known to blow and speakers have been known to catch afire. How well equipment performs is the province of test reports, but for the reports to be useful potential buyers need to understand what they mean.

Nonlinearities in electronic equipment are referred to as DISTORTION. In a modern amplifier the distortion is quite low until it begins to exceed its rated output. A 50 watt per channel amplifier might have a distortion figure of less than 0.03% up to 55 watts and then rise to 2% by 58 watts. This is the result of a phenomenon called CLIPPING. The power supply is designed to deliver some maximum voltage. As the amplifier approaches its limit the power supply sags. The amplifier cannot deliver the voltage required and chops off the top of the signal. Amplifiers are tested with sine waves and the top of the wave flattens when clipping is reached and distortion goes straight up in relation to power output. Twenty years ago an article in RADIO-ELECTRONICS said, "All high quality amplifiers sound the same." There are probably people who will argue with that today but the variation has certainly gotten quite small. Sometimes the audible differences are not enough to care about.

In order to have a digital revolution in audio the analog had to be converted to digital. This is the job of the Analog to Digital Converter or ADC. The original sound is analog and must be converted to an electrical signal to be stored or manipulated. This is the function of the MICROPHONE. The microphone produces a low voltage signal in the 5 millivolt range. Five thousandths of a volt. This needs to be amplified for the ADC to process so some analog circuitry is in the audio chain. The signal is raised to about 1/2 of a volt and the ADC can begin converting the sound to bits and bytes.

sine wave

0,37,65,81,95,102,110,117,124,127,130,132,132,131,128,125,118,111,103,96,82
66,34,10,-14,-53,-75,-89,-99,-110,-117,-122,-126,-128,-130,-131,-131,-131,-130
-127,-123,-117,-112,-104,-94,-79,-61,-32,0

sound wave

-1,17,34,52,32,28,34,26,0,-77,-97,-78,-51,-48,-47,-59,-96,-109,-86,-17,82,119
98,102,105,101,62,46,59,61,51,40,23,19,17,12,0,-48,-45,-41,-18,45,68,56,43,40
37,38,41,44,43,38,34,33,31,19,11,4

The example sound wave is a sine wave to the left of the vertical axis and a more typical sound signal to the right of the axis. Each vertical line within the wave represnts a sample by the ADC. The distance between each line is the time between each sample and remains constant. This is called the SAMPLING RATE. The sampling rate for CDs is 44,100 samples per second. The height of each line is the numerical equivalent of the height or AMPLITUDE of the audio signal at that time. CDs take 16 bit samples in stereo at 44,100 per second so that produces 176,400 bytes per second or 10,584,000 bytes per minute or 630 megabytes per hour. This large data requirement is the reason for the LOSSY COMPRESSION used in MP3s. Data is thrown away which presumably cannot be heard.

The sine wave has a total of 49 samples which at the normal CD sampling rate means this wave would be 900 Hertz. The wave rises quickly then levels off at 132 then drops to complete the negative half of the wave. The sound wave to the right is much more complex. The hightest positive peak occurs between two samples and is not detected and therefore could not be reproduced.

Cassettes are still widely available though competition is on the scene. CDs have been around for 20 years and some people are now predicting their demise. CD ripping makes it possibe to overcome the major flaw of the medium. The manufacturers don't put enough good tracks on their CDs so making custom CDs with 80 minutes of music is lightyears ahead of LPs. The threat to the future is MP3s. MP3s don't sound as good as cassettes recorded from CDs but people get them free from the internet. Portable MP3 players are superior to portabele cassette players since they have no moving parts. If they don't sound as good then what is the point though. We live in a world where people bought 8-TRACKS and MP3s are cool and cassettes are passe.

Copyright © 2009, Netbookation.com All Rights Reserved.