Components Specifications use and the Truth

[A.wayne]

At 1 ohm , most may smoke like one .....


One has to be careful when comparing double down specs, amplifier distortion rises for each halfing of impedance, because of such most will give the distortion spec into the lowest load when quoting power ratings and distortion.

for eg.

.1%thd

8 ohm 200watts
4 ohm 400 watts
2 ohm 800 watts

while others will state

.0001thd @ 8 ohm 200watts
.0001 @4 ohm 300 watts
.0001 @2 ohm 500 watts



but if they gave their distortion factor the other way , at .1% the amplifier would appear to double down in ever load. What is best is to compare distortion numbers and to make sure power is increasing as load changes and not the other way , which would suggest a weak PSU for low impedance operation ..


If you look back on the measurements previously posted for distortion vs power output, you will see a knee as distortion increases, technically this is the max power for best sonics (at knee) not at the reported 1%(SS AMPS) but at the apex of the knee , for this reason TUBES can sound as loud as their SS counterpart with half the power, because their distortion is less objectionable at max output as an SS.

So yes a 200 watt tube amp can appear to play as loud as an 400 watt SS amp ...

Regards,

 

[FrantzM]

Hi

One of the most interesting thing I got from this discussion is the fact that few, very few amps are Class A... Most are Class AB. I am also reinforced in my view toward Class A designs.. It has always been: It depends!! Some Class A amps are great sounding , some are not IMO of course. I don't see Class A as a requisite for an amp.. So far my favorites are all Class AB

Another point and to me that was the main aspect is to look at what marketing can do to specs ... for example last post by A. Wayne .. The game played when touting Power figures not that the audible differences between 0.01 and 0.1 THD would be an issue truly ..
Another point I would like to see clearly is the impedance plot of speakers as it related to their FR... Supposing that an amp can double-down and that the speaker dips to 2 Ohms on an arbitrary range say 60 to 120 Hz... while staying t 8 ohms everywhere else .. What is the impact of a doubling down amplifier on the FR/reproduction the speaker seeing more power in this region wouldn't that equate a bump in the level of this region? Wouldn't in that that case an amplifier with a constant output regardless of frequency be more appropriate for this speaker?
I am starting to see that is is an easy thing to say that all amplifiers sound the same but the Science doesn't seem to support it ... Simply speculating here .. More inputs please ... It seems that correctly interpreting specs can help one in making correct decision in optimizing a system.. thereby more enjoyment of music.

 

[FrantzM]

Hi

I am mightily interested in Speakers specs. My interest about Specs is that they may tel us more than we thought. In general data often exist but the use of it may not be easy or straightforward. I do think for example that speaker directivity curves are important. I favor limited dispersion speakers. i believe they sound more "neutral", in most rooms at the cost of a smaller listening window. I don't mind anyway when i listen to music seriously , I sit in the sweet spot .. A speaker that has very wide dispersion in an untreated room may sound too bright ... Speaker efficiency (apologies to microstrip ) is another pet peeve of mine. They are all over the place .. Speakers are usually much less sensitive than they really are and I m nto sure why a manufacturer plays that game , since it translates in dissatisfaction of the user .. then again audiophiles are a curious bunch and may well try to address their dissatisfaction with tweaks, speaker and/or Power cables change. Reporting, of course, "night and day" differences ... End of cheap shot to my brothers and sisters . I don't call that . "synergy" by the way, Component matching will suffice. Specs help when they are accurate.

 

[A.wayne]

Speaker sensitivity goes a bit deeper , for eg you can't compare directly point source sensitivity to line source , because point source drops off at 6 db for every doubling of distance while line source is 3db and they energizes the whole room in away that point source speakers do not ..

Regards ,

 

[JackD201]

Hi

I am mightily interested in Speakers specs. My interest about Specs is that they may tel us more than we thought. In general data often exist but the use of it may not be easy or straightforward. I do think for example that speaker directivity curves are important. I favor limited dispersion speakers. i believe they sound more "neutral", in most rooms at the cost of a smaller listening window. I don't mind anyway when i listen to music seriously , I sit in the sweet spot .. A speaker that has very wide dispersion in an untreated room may sound too bright ... Speaker efficiency (apologies to microstrip ) is another pet peeve of mine. They are all over the place .. Speakers are usually much less sensitive than they really are and I m nto sure why a manufacturer plays that game , since it translates in dissatisfaction of the user .. then again audiophiles are a curious bunch and may well try to address their dissatisfaction with tweaks, speaker and/or Power cables change. Reporting, of course, "night and day" differences ... End of cheap shot to my brothers and sisters . I don't call that . "synergy" by the way, Component matching will suffice. Specs help when they are accurate.

On my wish list of specs is polar response patterns like those provided for microphones.

 

[microstrip]

Speaker sensitivity goes a bit deeper , for eg you can't compare directly point source sensitivity to line source , because point source drops off at 6 db for every doubling of distance while line source is 3db and they energizes the whole room in away that point source speakers do not ..

Regards ,

Although this is true in an anechoic chamber, things become much more complicated in the listening room, as you say.

IMHO there are two ways of choosing the power of an amplifier - overpowering by a large margin, and than telling that as much power as possible is the best because transients get better, or trying the amplifier with the speakers at home - my usual choice.

 

[microstrip]

On my wish list of specs is polar response patterns like those provided for microphones.

Jack,

What would you do with them?

 

[Atmasphere]

Atma-sphere, excellent post. But this begs a question: how are, we as consumers, to know which amps will pair well with what speakers. For example, it is a well known fact that your amps paired with notoriously tough-to-drive Soundlabs make SUBLIME MUSIC. Your 150 wpc or 200 wpc, or whatever the specs of your amps are, seem to be more true than claims of other amp manufacturers for some reason.

How can we know if your amp can drive other notoriously difficult to drive speakers such as MBL 101 or Magico Q5? Is it only through trial and error?

Not really. The first thing you have to understand is that a lot of transistor amps can double power as impedance is halved, but that no tube amps do that. Then read this article:
http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php

It has everything to do with the intention of the designer of the amp or speaker. For example if you are contemplating a speaker- what sort of amp does the designer use? That will tell you a lot.
BTW the term 'output impedance' is one that has some controversy IMO. This conflict of the Voltage and Power paradigms is why.

There was the question how do you get low distortion wtihout feedback? The answer is, use every trick in the book. Avoid the things that are known to cause distortion and stay with those that are known to reduce it. For example transistors are a source of distortion- stay away from them. Use Triodes instead. Use fully differential circuits, so even orders are cancelled at each stage, rather than compounded thoughout the design. Going for super low distortion (using feedback) is likely going to exact a price on the amplifier! You might get low distortion but with increased odd-ordered content, the thing to which the ear is the most sensitive.

There is this sort of goal to create a perfect voltage source out of an amplifier so it can drive any load (this is one of the arguments for feedback). But the idea has problems- not all speakers are best driven in this manner. A great example are ESLs, which don't work all that well with a voltage source, as their impedance curve is not based on box resonance but a capacitor. A typical ESL will have a high impedance at low frequencies and a low impedance at high frequencies. A transistor amp thus has troubles making bass while at the same time has trouble making too much treble energy. But if the amplifier had a power source response rather than a voltage source response this problem would be eliminated. The same holds true for most horns, nearly all single-driver speakers (Lowther, PHY) and a good number of box speakers (the Acoustic Researech AR-1, the world's first acoustic suspension speaker, is a good example).

It is my contention that the Voltage Paradigm is made mostly to sell product from paper specs. It is the Power paradigm that is based on human hearing rules. Becuase of these two fundamentally different approaches, the tubes/transistor debate will rage on, as the objectivist/subjectivist debate. It will continue that way until we find a way to resolve what we hear against what we measure.

When you mix Power Paradigm technology with Voltage Paradigm technology, you get colorations. An example is out MA-2 driving a set of mbl101s. There is plenty of power, but the speaker is 4 ohms except for an 8 ohm peak in the upper midrange. The amplifier had better run feedback or some means of cutting its power in half into the 8 ohm region, else it will be bright in that range and that is how the MA-2 behaves on it. Neither the amp or the speaker are *inherently* bright, but that is what you get when you mix paradigms.

The thing is, yes, this should be a cut and dried process, but its not. Tubes did not go obsolete 50 years ago because of the transistor. They are still around for a reason. But I would think anyone contemplating an expensive purchase would check ith the manufacturer....

 

[JackD201]

Jack,

What would you do with them?

It would be a nice quick start to determine speaker placement options in a room (long or short wall), decide which room design type (LEDE, RFZ, etc.) mainly.

 

[KeithR]

Frantz

Power amps that "start in Class A but slide into Class B" are called "Class A/B" - anything else would simply be a lie. It is not unreasonable to spec a Class A/B amp as "being Class A for the first 30 watts" if it is biased such that both sides of the output stage are carrying current up to the 30 watt level, but that needs to be made clear. The benefits of Class A operation are dubious to begin with, so it's probably more a matter of hyped marketing than a deliberate attempt to mislead in most cases... but who can say. Anything with a single-ended output (say a SET tube amp or a headphone amp with a single power FET on each channel, HAS to run Class A in order to deliver a listenable output. Unfortunately, "Class A" has pretty much devolved into an audiophile buzzword anyway. There are many products running in Class A/B that far outclass others running in Class A - co Class A operation is NOT a good indicator of sound quality. The only thing Class A operation guarantees you in a power amp is a lot of heat, a lot of weight, and a high price tag (or, in the ca

Can you explain your claim that Class A amps have dubious benefits? Thanks for joining conversation btw--very informative posts.

 

[Atmasphere]

Can you explain your claim that Class A amps have dubious benefits? Thanks for joining conversation btw--very informative posts.

Class A amps will have lower distortion over the same thing set up as AB. If you are trying to create an amplifier that is low distortion and has no feedback, class A is an important tool. It does not have to create an excessive amount of heat BTW. If the amplifier has a sliding bias system its heat generatation can be nearly the same as an AB amplifier.

 

[DACMan]

Sure......

A well designed Class A/B solid state amp has very low distortion, while the distortion on a well designed Class A amp can be even lower.
If you look at the numbers on a good example of each, the distortion numbers go from "ridiculously low" to "even more ridiculously low".
However, a well designed Class A/B solid state amp will also certainly have lower distortion and better sound than a less good Class A design.
I've only ever heard a few Class A solid state amps, but they did NOT sound better than the best Class A/B amps I've heard....

The payback is that a Class A SS amp consumes a LOT more power, and runs a lot hotter because of it, which leads to:
* bigger amp, bigger heat sinks, higher cost
* hotter running makes it more prone to thermal bias drift and other similar problems (short term)
* higher current and higher temperature in the internal components shortens their life A LOT (long term)
* higher current may actually increase the distortion and noise of some components
* gotta run that noisy air conditioner in the summer with all that extra heat

Now, with tube amps, things are a bit different....
In particular, SET tube amps, because they are single ended AND Class A, run their transformers in about the most demanding
mode possible (in a push-pull design, the DC currents in the two halves of the output transformer cancel out, and so
tend not to saturate the core; in a SET design, you are running higher current, and only one way, which is pretty much the
worst possible thing you can do in terms of saturating the core, which gets you a much bigger transformer - and much
worse performance - than a similar powered push-pull design.)
Oh, yes, being single ended, they DON'T cancel out the power supply noise in the output stage either, so you either
need an incredibly good power supply or, to flip it around, if you use the same power supply, a SET will be about 20 dB
or more noisier than a push-pull unit with the same output power and degree of power supply filtering.
(More complicate power supply= higher price, lower efficiency, lots more parts, lots more complexity, and lots more likelihood of failure.)

In addition, and not that it matters much, but most Class A/B solid state amps run in Class A for the first few watts anyway
(for my 350 watt Aragon 8008, it is in Class A for the first 20 watts or so... not that it matters much).

There are indeed some very good sounding Class A amps... but, unless they sound better than the best sounding Class A/B ones
(which I don't think they do), then the benefits are, to me, dubious - meaning that they aren't worth the down-side.

If I have a choice between a 350 watt Class A/B amp, which runs Class A for the first 20 watts anyway,
or a 20 watt Class A amp - probably for similar prices because of the added mystique and legitimately higher cost to build
the Class A unit, where is the real benefit?

Can you explain your claim that Class A amps have dubious benefits? Thanks for joining conversation btw--very informative posts.

 

[Phelonious Ponk]

Sure......

A well designed Class A/B solid state amp has very low distortion, while the distortion on a well designed Class A amp can be even lower.
If you look at the numbers on a good example of each, the distortion numbers go from "ridiculously low" to "even more ridiculously low".
However, a well designed Class A/B solid state amp will also certainly have lower distortion and better sound than a less good Class A design.
I've only ever heard a few Class A solid state amps, but they did NOT sound better than the best Class A/B amps I've heard....

The payback is that a Class A SS amp consumes a LOT more power, and runs a lot hotter because of it, which leads to:
* bigger amp, bigger heat sinks, higher cost
* hotter running makes it more prone to thermal bias drift and other similar problems (short term)
* higher current and higher temperature in the internal components shortens their life A LOT (long term)
* higher current may actually increase the distortion and noise of some components
* gotta run that noisy air conditioner in the summer with all that extra heat

Now, with tube amps, things are a bit different....
In particular, SET tube amps, because they are single ended AND Class A, run their transformers in about the most demanding
mode possible (in a push-pull design, the DC currents in the two halves of the output transformer cancel out, and so
tend not to saturate the core; in a SET design, you are running higher current, and only one way, which is pretty much the
worst possible thing you can do in terms of saturating the core, which gets you a much bigger transformer - and much
worse performance - than a similar powered push-pull design.)
Oh, yes, being single ended, they DON'T cancel out the power supply noise in the output stage either, so you either
need an incredibly good power supply or, to flip it around, if you use the same power supply, a SET will be about 20 dB
or more noisier than a push-pull unit with the same output power and degree of power supply filtering.
(More complicate power supply= higher price, lower efficiency, lots more parts, lots more complexity, and lots more likelihood of failure.)

In addition, and not that it matters much, but most Class A/B solid state amps run in Class A for the first few watts anyway
(for my 350 watt Aragon 8008, it is in Class A for the first 20 watts or so... not that it matters much).

There are indeed some very good sounding Class A amps... but, unless they sound better than the best sounding Class A/B ones
(which I don't think they do), then the benefits are, to me, dubious - meaning that they aren't worth the down-side.

If I have a choice between a 350 watt Class A/B amp, which runs Class A for the first 20 watts anyway,
or a 20 watt Class A amp - probably for similar prices because of the added mystique and legitimately higher cost to build
the Class A unit, where is the real benefit?

Silly man. You listed the real benefits in your own post:

Class A SS amp consumes a LOT more power
bigger amp, bigger heat sinks, higher cost
a much bigger transformer
need an incredibly good power supply
More complicate power supply= higher price, lower efficiency, lots more parts, lots more complexity

Cue Tim the Toolman: "Ugh ugh ugh ugh...."

I'm only half-joking, of course. Most of these things you've listed as negatives have become core features and benefits of many high-end amps, to the point where many seem to believe a small, efficient, cool-running amp can't be good, and big, heavy and hot are evidence of quality.

Tim

 

[A.wayne]

Wait , what, its not , now i'm confused .....

 

[Tonepub]

So why do you all waste so much time arguing about specs anyway?

There is no real spec for tonality...

Every amplifier designer I've ever met (and I've met most of the majors) all say measurements are a place to START not a place to END.

Just seems like a big excuse for either justifying your purchase or NOT listening to music.

Gotta go, big pile of records just arrived and I know what I'd rather be doing!

 

[Bill Hart]

So why do you all waste so much time arguing about specs anyway?

There is no real spec for tonality...

Every amplifier designer I've ever met (and I've met most of the majors) all say measurements are a place to START not a place to END.

Just seems like a big excuse for either justifying your purchase or NOT listening to music.

Gotta go, big pile of records just arrived and I know what I'd rather be doing!

You like cleaning records?

 

[mep]

You like cleaning records?

I don't mind cleaning records. Less than two minutes from the RCM to the table for me.

 

[rockitman]

So why do you all waste so much time arguing about specs anyway?

There is no real spec for tonality...

Every amplifier designer I've ever met (and I've met most of the majors) all say measurements are a place to START not a place to END.

Just seems like a big excuse for either justifying your purchase or NOT listening to music.

Truth...following this notion would save alot of forum bandwidth too....

 

[JackD201]

I'm only half-joking, of course. Most of these things you've listed as negatives have become core features and benefits of many high-end amps, to the point where many seem to believe a small, efficient, cool-running amp can't be good, and big, heavy and hot are evidence of quality.

Tim

Unfortunately, in my case it is based more on experience than text books or googlin'. IME linear is still generally better sounding than switching and the better high bias' distortion artifacts sound less artificial than anything but the best A/Bs. I still think it is a matter of tastes.

I alternate between a pure class A and an A/AB when I am not using them together. They have their strengths and weaknesses (which is why I Bi-amp) and I can see why one might choose one or the other. As for me, I have 5 class A amps, only two A/ABs (not counting those in my 2 pairs of actives) and also only 2 Class D integrateds. I guess that shows where I lean.

 

[microstrip]

So why do you all waste so much time arguing about specs anyway?

There is no real spec for tonality...

Every amplifier designer I've ever met (and I've met most of the majors) all say measurements are a place to START not a place to END.
(...)

Jeff,

Because we all love to have feedback on these matters from people who are in close contact with the high-end industry!