Oh Boy, here we go!
12 responses Add your response
I would suggest some background reading.
I would seek out the chapter on amplification in Robert Harley's "The Complete Guide to High End Audio" for a general and relatively easy-to-understand discussion on amplification and some of the differences between solid state and tube amplification. I'd also read the section on speakers (because amplification is simply driving a load).
Then, I would go to the Atma-sphere website and go to the "Papers" section to read the paper titled Debunking Common Power Amplifier Myths and the other titled Competing Paradigms in Amplifier and Loudspeaker Design, Test and Measurement.
Knowing how tube amps create their "power" and how transistor amps create theirs, and understanding how speakers present a load to each particular kind of "power" will make things easier to get through the Atma-sphere papers though they may be able to be tackled on their own.
First of all the Almighty set ALL watts to be equal to the product of Voltage and Current or physically Work per unit of Time.
The power P that is measured in Watts can be consuming and giving and the remaining power dissipates on heat.
Tube amplifiers and Class A SS amplifiers are only (approx)20% efficient since the remaining 80% of power goes onto heat.
The mentioned 'sonic watts' can be described as SPL that is measured in dB(usually guitar dudes know what it realy is). Decibel is a logarighmic relative value that compares no sound to the actual sound. SPL can be measured also at particular freequency and that's where all the Tube/SS confusion burried as the 'Holy Grail' for some folks arround here.
Normally tube output devices have very high output impedance and amplify voltage that suppose to be converted to current using the output transformer to satisfy nowdays speaker voice coils. Transformer is the device that converts variable electric field of an input coil into the variable magnetic field and the output coil has the induced electric field from the variable magnetic field. The input and output coils of the transformer are actually inductors that have variable impedance depending on frequency. As the frequency drops, the reactance is minimal and impedance of the inductor approaches actual resistance of the wire. In this case the voltage accross the input coil terminal becomes MINIMAL as well as magnetic field across the primary coil! Another words by nature there's hardly any lower-end extention and 'visible' spectrum of it is substantially less than corresponding SS amplifier. So placing 'in front of each other' 100W tube amp against 100W SS amp obviously SS will seem to sound more quiet, but if measured SPLs across the 'visible' frequency spectrum, SS will have it larger in both ends of a slope.
As to the speaker voice coils, lower frequencies should be hadled with much higher current than midrange.
In tube amp transformer would act rather as high-pass filter thus distributing a larger portion of an output power onto substantially smaller frequency range giving the 'illusion' of so called 'TUBE WATTS'.
First off, power is only important if you need it. Meaning, that if you are using speakers with a sensitivity of 105db, you could run them with a transisitor table top radio. You would only need about 500 milliwatts, half of a watt. But if you have maggie 3.6, or Sound lab, then you need a few hundred watts, like 500wpc into a 4ohm load. The load presented by the speaker is important, and the sensitivity also is impoprtant. Finally, they need to be synergestic, less concrete, but very real. I like well matched SS and well matched tubes. A friend has the latest legacy whisper speakers rated at about 96db sensitive, and we compared a 70wpc vs a 600wpc SS amp. Which had the best bass??? The tube amp, best mids, tube amp, best treble...you guessed, tubes. This SS amp is said to be a great match for the legacy whisper, but tubes beat it. All tubes won't as my quickies were a bit slow on these speakers. they are great with mine, 3.6's. Factors like damping factor, speaker sensitivity need to be included in the consideration of amplification. I prefer tubes, but cheap good watts are also available with SS.Jallen
Moot or mute...either way they don't make any music without placng them into a system and plugging them in:O) Ditto Unsound on the interesting details. You can get a PHD in electrical engineering with a minor in audiofoolery and still put together a lousy sounding system. It's all in the alchemy. Consulting with an audio wizard like Audiofeil might help as well:O<>
There has been a debate in audio- tube vs transistor- for the last 50 years. The results of the debate are unimportant. What *is* important is understanding that the debate arises from a deeper conflict as described in the links to the Atma-Sphere site in T_bone's post.
As others have already pointed out, matching between amps and speakers is paramount. The equipment matching conversation also arises from the Voltage vs Power conflict.
I personally feel that the important thing here is the rules of human hearing. I think we can all accept that these rules are going to be violated by equipment that is not perfect (IOW all equipment), but that some will come closer to following those rules than others. I contend that the Rules of Human Hearing are in fact the most important thing in audio- without ears, we would have little interest in audio equipment :)
So what is the most important Rule, IOW what is the thing that is the most important to get right? IMO/IME it is the way we perceive loudness. This is done by the ear/brain system by sensing the presence of trace amounts of odd-ordered harmonics in the sound. The louder these harmonics (5th, 7th and 9th of the fundamental) the louder we will think the sound is.
If these harmonics are unnaturally enhanced, the sound will loose its natural quality. Many things can cause this enhancement, or distortion, but the 2 chief culprits are transistors and loop negative feedback. In fact, loop feedback is the bigger offender. It is possible to build zero-feedback transistor amplifiers to avoid this distortion, and those that have done so successfully are at the leading edge of the art of transistors. However, its a lot easier to do it with tubes (which, for the record, can have full power bandwidth as wide as the best transistor amplifiers).
This why I say the results of the old debate are unimportant- because either technology can be successful if we only know what it is that we are supposed to do (which is obey the rules of human hearing rather than meet an arbitrary set of specs that are meaningless to the human ear).
The objectivist vs subjectivist debate is nearly as old as the tube/transistor debate, and for the same reason. It too arises from the conflict of the Power and Voltage paradigms. The Voltage paradigm is responsible for a set of arbitrary specs that I referred to earlier; we know from listening that they are not important. The Power Paradigm attempts to follow the Rules of Human Hearing.
Don't feel bad if you go into a dealership and get a blank stare when you ask about this stuff, but any dealer worth his salt will invite you to audition the equipment. Until someone has developed a set of specs that if followed, will guarantee that the Rules are being followed, audition is about the only game in town.
Atmasphere, there's one important detail missing on your link mentioned by T_bone and which you perfectly might know as engineer is meaning of ideal source vs. non-ideal. It can also be explained in simple language to the public.
Tube or even transistor output devices can't be ideal by default: What device is closer to ideal source tube or transistor? The total impedance including impedance of the source should've been mentioned as part of the integrated source --> load circuit. There figures will come quite different especially in terms of current passing through the load.
Watts is Watts, at least into a resistor, which NO load is, as far as I know.
One major difference is the delivery of power into reactive loads, which to a greater or lesser extent includes ALL speakers. Huge phase angles at low impedance will take the wind out of many amplifiers sails.
Transistors typically do better. I don't remember which, but tubes don't like either Capacitive or Inductive loading, again as a general rule.
The articles ref'd to in an above post are a good read. Thanks! Also, Nelson Pass has at least 1 article on the web concerning Current Source amplification vs Voltage Source along with extensive testing of some single driver systems.
Marakanetz, true enough- but since this **is** the real world, an ideal source exists in dreamland only- we can safely ignore it as a result.
So, what is, is tubes and transistors. Either can have a high or low source impedance, and a price is paid either way, and both (high and low source impedance) have very distinct advantages.
IMO/IME, there is no such thing as an ideal source, and I am not even sure that there is even a good consensus about what that even is. Keep in mind, the reason I used the word Paradigm in the way I did is that those who operate inside one Paradigm will by definition be unable to accept any way of thought that exists outside of that Paradigm. So what is an 'ideal source' for the Voltage Paradigm is very different from that of the Power Paradigm, and if you re-read the paper, you will see the issue clearly addressed.