Audio Power Amps: Channels & Output Device operation

By Published by
Audio Power Amps: Channels & Output Device operation
. Views . Comments Comment . 66 Votes
The design of audio amplifiers is a hugely complex subject.  This guide is intended to help you sort out some of the more obvious and important differences among audio power amplifiers, because every different design has it's own set of compromises along with it's own set of advantages.  Because of the nature of an eBay Guide, I must be brief, and I will leave out many details.

There are many different ways to build a power amplifier, and then there are many different ways to package the electronics into a metalwork "box".  The most important differences are:  the number of channels inside the box, and the choices the designer makes to provide a balance among energy usage (from the wall socket) , power output to the speakers , and overall cost, all while maximizing sound quality.  And quite a balancing act it is!

Number of channels of amplification

At it's core, an audio "amplifier" is a single channel device. (It is monophonic.)  A single channel of amplification in it's own box is a " monoblock" amplifier.  There are stereophonic amplifiers ( stereo, or two channel audio with a "left" and "right" channel) but that's just two individual amplifiers sharing a single housing.   Now, with the popularity of multichannel home theater, and multichannel audio via SACD and DVD-Audio, what's sold as an "amplifier" may have many individual channels (generally, 3-, 4-, 5-, or 7-channels) of amplification built into a single box.

Monoblock Amplifer Advantages:
  • Having only one channel of amplification in a chassis is regarded as the best--but most costly--solution; it is intended to provide the utmost sound quality, all other considerations are secondary.  Since everything in the box is devoted to a single channel, a monoblock amp is generally (not always!) more powerful than a stereo or multichanel amp.  There is no electrical connection between the channels except at the wall socket, and so signal seperation is maximized.
  • The amplifier can be placed close to the speaker (as in, within inches) minimizing the amount of speaker cable needed.
Monoblock Amplifier Disadvantages:
  • Cost:   You buy a seperate amp for each channel in your system.  It's not uncommon for a single monoblock to cost the same as a stereo amp.
  • There are those who argue that it's better to have short interconnect cables than to have short speaker cables.  One thing is guaranteed:  short cables do less damage to the signal than long cables, whether we're talking speaker cables or interconnect cables.
  • Some speakers (Polk SDA series, for example) are designed to work with amplifier channels that share a ground path for the two channels.  (These speakers are not common.)  Seperate amplifiers don't have that common ground connection.
  • You need to be able to "plug in" each amplifier seperately.  Your home wiring needs to have outlets near where each amp is located.  You may need to hire an electrician to "beef up" the wiring in your house so that the monoblocks have a source of enough electrical power.
Stereo Amplifier Advanatages
  • Two-channel amps are more common than monoblocks or multichanel amps, because high performance audio was all about "stereo" for thirty years.
  • Cost:  Compared to monoblock amps, you're buying half the amount of chassis metalwork.  One "box" instead of two.  Make no mistake:  The metal box that houses the electronics is a huge part of the total cost of an amplifier.
  • "Dual Mono" is a variation of stereo, in that while both are two-channels of amplification in one box, a Dual Mono amp is designed in such a way that it has nearly the electrical seperation as a pair of mono amps.  The two channels of a dual-mono amp share a chassis, (and usually a chassis ground,) a power cord and power switch, and--perhaps--the primary winding in a shared transformer.  Some dual-mono designs use totally seperate transformers, and some (generally the lower-powered or less-expensive) use a transformer with three windings--a shared primary winding, and a secondary winding for each channel.
Stereo Amplifier Disadvantages
  • They are the compromise between monoblocks and multi-channel.  So, you're buying less metalwork (chassis) than if you bought a pair of monoblocks, but if you buy several stereo amps, you're buying more metalwork than if you bought a single multichannel amp.
Multi-Channel Advantages
  • If you actually need several channels of amplification for a multichannel music or home theater application, you'll generally spend less money if you buy all the amplifiers built into one box.
  • Some multi-channel amps are designed so that pairs of channels can be combined into a single channel but with higher output power.  As an example:  You may buy a 100-watt per channel, 4-channel amplifier (100 X 4) that allows you to combine a pair of channels so that instead of 100 X 4, you could have 100 X 2, plus a third channel of 200+ watts; or even combine the second pair so that you end up with 200+ watts X 2 channels.  Not all multichannel amplifiers have this capability--be sure to check before you buy!
Multi-Channel Disadvantages
  • Since size and weight becomes a concern, and because all those channels of amplification are drawing electrical power from one power cord off of a single wall socket, maximum amplifier output power can be compromised.  There are "clever" designs that can reduce the price, size and weight of a multichannel amp, and still provide tremendous "rated" output.  The problem is, those "clever" designs often compromise the sound quality in the process of providing a more consumer-friendly package.  Some of those "clever" designs are described below.

Operation of the output devices:  Single ended vs. push/pull; Bias and Rail voltage
There are many, many ways to build an audio amplifier, but every amplifier has at least one "output device" for each channel.  This is the component within the amplifier that actually powers the speakers.  Everything else in the amplifier is the "life support system" for the output device.  These output devices can be vacuum tubes, or they can be transistors.  The transistors may be packaged within an integrated circuit--a "chip", or they may be individually mounted to the circuit board.  Very few amplifiers use a single output device per channel, those that do are called "single ended".  In a single ended amplifier, the whole signal is produced by that single output device.  The earliest amplifiers (1906) were single ended, it's a very simple circuit.  Single ended amplifiers are making a modest comeback as of the last few years, supposedly because of the "purity" of sound they produce.  They tend to be very low powered--on the order of one to ten watts, although there are exceptions--and very expensive because of the low volume of production, and because they are "exotic".  Most amplifiers will use two or more output tubes, or two or more output transistors, working as a team.  One or more output devices will be responsible for the positive voltage part of the waveform, and one or more devices will provide the negative part.  This is called a "push/pull" type of amplifier, and the first use of "push/pull" dates from a 1913 patent.  If a push-pull amplifier is advertised as having six output tubes; or six output transistors per channel, it will use three to power the positive half of the waveform and three to power the negative half.  How push/pull devices "hand off" the waveform to each other at the zero voltage level is very important, as you will see.  How they hand off to each other depends on the way they are biased.  Bias is an electrical signal that is not related to the musical signal the amp is reproducing.  If the output devices were an office worker in a cubicle, the bias would be the coffee the worker drinks to keep him "perked up".  It's something that keeps the output devices awake and ready to amplify their half of the signal.

Once upon a time, there were a few simple classifications for an amplifier circuit, based on the amount of bias used.  Of course, nothing stays simple for long.  Now there are many classifications which differentiate both differences in bias; and rail voltage.  Rail voltage is the "pure power" that passes through and is processed by the output device(s) and eventually powers the speakers.  Or, put another way:  the output devices control the flow of the rail voltage in such a way as to cause the speakers to make music.  I'll list a few classifications here:

Some Standard Bias Classifications
  1. Class A:  Output devices are "on" all (100%) of the time. (whenever the amp is turned on)  They draw a lot of current, and produce a lot of heat--low efficiency.  Very linear, so they produce little distortion and therefore the "best" sound.  ALL single-ended audio amplifiers are Class A, it would be somewhat unusual for a push/pull audio amp to be Class A--but there are a few, and they tend to be expensive.
  2. Class B:  Output devices are "on" exactly half (50%) of the time.  There is a "burst" of distortion as each device turns on and off at the 0 voltage point of the musical signal.  The signal crosses from negative to positive, or from positive to negative at the 0 voltage point--but before the signal can have it's "other half" amplified, the transistor has to turn on, hense the burst of distortion.  Power draw is much less than in Class A, and they tend to run cooler.  Better efficiency at the expense of distortion and therefore sound quality.
  3. Class A/B:  The devices are on more than half (50%) the time, but less than "all the time" (100%).  Better efficiency than Class A, with less distortion than Class B.  The transistor has already turned on when the music signal gets to the 0 voltage crossover from negative to positive, or from positive to negative.  This means the Class B burst of distortion is eliminated or greatly reduced.  Class A/B was the most-common method used for audio amplifiers for many years and is still very popular.  If the designer has the bias set at 51%, the amp will have more of the Class B attributes; if the bias is set at 99%, it will have more of the Class A attributes.  Finding the best compromise for bias level is part of the challenge for the engineer.
  4. Class C:  Output devices are "on" less than half the time.  There is a great deal of distortion, this design is just fine for some applications, but audio is not one of them.  Don't expect to see Class C audio amplifiers.  Very efficient, very low heat, it would sound very terrible.
  5. Class D:  Output devices are switched on and off faster than the signal that they are amplifying.  If the signal is positive, the positive device spend more time "on" than the negative device.  If the signal is negative, the negative device is turned on longer than the positive device.  If the devices can't switch on and off fast enough, the audio signal gets mangled.  Because of this, Class D has been popular for subwoofer amplifiers where they don't have to deal with high frequency signals.  Very efficient, very low heat, good sound if used within the frequency range of the devices.  Class D is becoming much more popular!
If you then start to play games with the rail voltages--the sources of the pure + and - power that eventually drives the speakers, you can also make gains in the efficiency of the amp.  That is, more output power with less input power, and less wasted energy given off as heat.  A normal amp has fixed rail voltages.  We'll say for this example that there is a + rail voltage of 70 volts, and a matching - rail voltage of 70 volts.  At low volume, that + and - 70 volts is mostly wasted as heat output, it does very little that is useful.  The heat has to be dissapated, or the amp cooks itself.  So, you have to have a fan, or lots of heat sink area so the waste heat can escape.  But what if the rail voltage was variable?  You could have a + and - rail voltage of, for example, 20 volts.  That would supply enough power for low volume listening, but not enough for loud music passages.  During loud passages, the amp would have to be smart enough to temporarily increase the rail voltage to perhaps 50 volts--and do it just before the music signal actually needed that boost--in order to properly power the speakers.  And when the cannons go off in the 1812 Overture, the amp would instantly increase the voltage to 70 volts, and you'd be able to have full power from the amp, as much power as the first amp that has 70 volts all the time, only without most of the waste heat.

A three-voltage rail like I've described above, in combination with an ordinary Class A/B bias system, is called Class G.  Class G could use two rail voltages, or four, five, or a hundred.  The point is, you have two or more discrete rail voltages.  Hitachi gets most of the credit for inventing Class G in 1977.  In fact they merely popularized it, it was written about at least as far back as 1965.  Other amp manufacturers besides Hitachi have used--or are using--multiple, switched rail voltage systems.

What if, instead of two or more discrete rail voltages, you increased the rail voltage smoothly, so that it exactly matched the signal voltage (except that it was always a little bit higher?)  A smoothly-increasing/decreasing rail voltage is part of Class H.  Class H is credited to Soundcraftsmen in '77, but again the true origin of the modulated power supply goes back many years before, to 1964.  Soundcraftsmen used the modulated rail voltage along with some other technological twists as part of a "package" they marketed as "Class H".  Other amp manufacturers besides Soundcraftsmen have or are using the modulated (also called "Tracking") power supplies, although they may not use the rest of the "Class H" circuitry.

The benefit of the multiple-voltage; or modulated-voltage rails, is that most of the time, the amp would run much cooler, which means that you could cut back on the amount of expensive heat sinking needed to cool the output devices. Since there's less weight and bulk from using less heat sink, the amp can be packaged in a smaller and lighter metal enclosure.  The company spends less to ship the lighter, smaller amp from the factory to the dealers.  The designer makes a more complex circuit to control the rail voltage, but overall the unit may cost the consumer less money.  It is said that the complexity of switching or modulating the rail voltage can interfere with the sound quality of the amp.  That may--or may not--be true.  Some folks claim to be able to hear a degradation of sound quality from changing the voltage on the rails.  What's important is whether YOU can hear a difference.

Where Class G and Class H are variations of class A/B, Class T is an offshoot of Class D.  Class T is an invention of Tripath Technology Inc., used for their proprietary "Digital Power Processing" which modulates the input signal with a high-frequency switching pattern.  Tripath claims 80-to-90 percent power conversion efficiency for this technology.

The most expensive parts in an amplifier tend to be the metal housing, the metal heat sinks, and the power supply--the transformer, rectifier, and storage capacitors that supply all the electricity to the amp circuits.  Anything that can be done to reduce the size of the "box"; and/or to reduce the size of the power supply will have a major effect on the price of the unit.  The engineer's goal is to maintain the sound quality while figuring out ways to reduce the cost.  Believe me, not all attempts at that are successful!  Listen before you buy!  When it was my own money I was spending, I chose a Class A/B design that had a huge, heavy power supply.  Others will make different choices.

If this guide has been helpful, please give me a "Yes" vote by clicking the button below.  If you have suggestions for improvement, please contact me through the "My Messages" feature of eBay by clicking on my user name above, and then click on "Contact Member".

Entire content copyright (C) 2007, 2008 Camino3X2    Feel free to LINK to this Guide in your auctions.
Explore More
Choose a template