Speakers, Models and Power!

[kkantor]

>Perhaps I am the one missing something in the model. How can

>these passive elements simulate the back emf? Shouldn't there

>be a voltage sourse here which generates a time varying

>signal, probably an impulse which is dependent on the applied

>signal?

Here's one way to think about it: the speaker isn't producing any net power, as a voltage source would. Instead, the takes electrical energy provided from the amp, and does three main things with it:

1- The vast majority of the energy input winds up as heat. In turn most of this heat leaves via the voice coil. A small amount may be generated directly in the box and in the suspension, too.

2- A fairly small portion of the energy is turned into sound, vibration of the air. Typically, this is well under 1%.

3- Finally, a small portion of the electrical energy is "stored" within the speaker. Storage takes the form of potential energy (box pressure) and kinetic energy (cone momentum). The energy is, of course, not stored for very long. Most of it also winds up as heat, some as sound, and the remaining amount returns to the amp in the form of "Back EMF."

In fact, capacitors and inductors also store energy. Resistors turn energy into heat. So, the electrical circuit can be made to do the same thing the speaker does: radiate some of the energy, store some of the energy and return some of the energy. (For example, you can think of accelerating the mass of the cone as similar to filling the cap with electrons.)

>Is the impedence of a loudspeaker constant with voice coil

>displacement?

Yes, to the extent that the loudspeaker is "linear." In practice, you only see level-dependent change right at resonance, mostly due to surround and spider effects. You can see small effects across the whole frequency band, if you look very carefully.

>Wouldn't moving the coil with respect to the magnet change its inductance dynamically?

Well, the whole point of good motor design is to keep the relationship of the magnet and coil as constant as possible, over the longest distance. Once you exceed this range, nominally "Xmax," various things start to change, and inductance is but one of them.

> How about voice coil heating? Surely the change in temperature

> effects its DC resistance.

Of course, but this is a longer term process. Heating the parts in the equivalent dummy load changes their resistance, too.

>These non-linearities suggest to me advantages of

>multiple amplifiers with active crossovers where frequency

>response and relative loudness can be adjusted precisely and

>independently of driver non linearities and where far fewer

>compromises are necessary over passive crossover multiway

>systems driven by a single amplifier. This would have been

>outrageously expensive and extravagent in 1966. In 2006, it

>seems to me to be a viable option for a well designed high

>accuracy speaker/amplifier. I'm surprised it isn't used more

>often so that the speaker designer has far more control over

>the performance of his product. The notion of buying speakers

>and amplifiers separately strikes me as an example of

>"the Chinese Wall syndrome" mentality where the

>consumer has to supply experimentation as a substitute for

>manufacturer expertise. Who would ever even have considered

>way back when, that what was proported as a high fidelity

>loudspeaker system would need a user supplied outboard

>subwoofer (except for full range electrostatic systems)?

Easily 90% of the speakers sold today contain built-in or dedicated amplification. That tiny niche of the market that we hobbyists pay most attention to earns much of its income from the mix and match game, and we seem to love it. The average consumer is not expected to do this anymore.

[DaveD]

To provide reference to the origin of this discussion, I post below my message of today to the AR Forum string which I started and titled "SS amp and vintage speakers-- how to emulate tube amp?"

------------

TECHNICAL RATIONALE FOR THE RESISTOR TECHNIQUE--

The fellow who gave me the advice to try the resistors sent me his technical rationale the other day. It is as follows:

My question--

"I have tried the Rat Shack 1 ohm, ten watt wirewound resistors on the 17's, with my NAD C740 receiver. The midrange did open up and soundstage became more dimensional and separated and detailed. Maybe the highs became more transparent and extended, or maybe I only thought so because there was more detail and less congestion. However, the bass became too bloated and uncontrolled. The Shack had to order the .47 ohm, 5 watt resistors that you also recommended trying, and I hope that they will give the midrange improvement without hurting the bass control. Does the damping factor of the particular direct coupled SS amplifier being used determine which resistor will work best? Is there a better choice than the .47 ohm, 5 watt Rat Shack wirewound resistor that I will be trying?"

His answer--

"0.47 is what I wound up with. I think you'll like those. The amplifier DF is really a measure of the amp's internal impedance, but expressed in a back-handed way. The idea of the series R is to emulate the series impedance in the tube amps the 17's were designed to work with. DF = 8 divided by the internal impedance. Tube amps of the 60's ranged from 6 to 18 in DF, which works out to .45 ohms to 1.3 ohms. So that's the range of total series impedance you're looking for. The SS amp you use contributes a small part of that, but typical SS amps have DF from as low as 25 up to 1000. As a rule of thumb, I'd say you would need to worry about the amp's series impedance if it's more than 0.2 ohms or a DF = 40 or less. In any case, there are other things that affect the sound, so you need to try it and see. The math only gets you close for the first try."

-----------------

I will send my correspondent (from the Audio Asylum forums) links to this discussion and he may decide to join in.

[Guest GWCrum]

Ken,

I usually refer to this effect as "the voltage divider effect" as you do. The trick of adding a series resistor to emulate the source impedance of a vacuum tube amplifier (or any other low source impedance amp) was suggested by Ed Vilchur in an article that I wish I had saved. Might have been in Electronics World or one of those, but it was not in a "hi-fi" magazine. In it Vilchur suggested adding a series resistor in response to people who had AR-3's (or 3a's?) and were using solid state amps and were not satisfied with the bass response. That was sometime in the later 60's.

The earliest reference to the effect that I have found was in an article by Johm Tomcik of EV who mentioned it in passing back in 1954.

Oddly enough Bose Corp mentioned it at some length in a beautiful 16 page color brochure for the 1801 power amp from around 1974. They talked about Damping Factor and Wire Factor and showed an expression, that related them and frequency response errors resulting from them. I turned the "expression" into some real math and realized they were talking about a voltage divider. That was a bit of an epiphany for me.

Since then I have been trying to get folks to understand the effect and that it is full range and not limited to the bass resonant frequency. I have a few converts, but precious few. The idea of "damping" of the bass resonance is so ingrained it's hard to break through. I fervently wish the term Damping Factor, had never been invented as it leads one away from the voltage divider effect being full range.

I assume you read Stereophile? At least once in a while. John Atkinson has been publishinghis measurements of amplifier frequency response and ther reaction to a simulated speaker load for years. Yet it never seems to sink in as one of the big factors in why tube amps sound different. In some of his tests of SET's, the source impedance is on the order of 2 ohms and the frequency response variation due to the voltage divider effect was 4 dB. I calculated 3.9, so that pretty well confirmed my math was right.

Bottom line, I'm really glad to see you chime in on this. It has really needed the legitimization of a major name in the industry for a long time. I have some downloads of your comments and John Dunlavy on cables. Since I use 12 gauge "zip" cord, I was pleased to see those.

Oh, by the way, I'm a retired engineer, and a long term audio hobbyist with a fair amount of professional acoustical background, but not in the high fidelity industry. Mostly industrial or in aerospace.

Best regards,

Gerald Crum (Jerry)

[kkantor]

Thanks, and I agree.

Hmmmm, yes, I believe I've heard something about that Stereophile testing thing....

http://stereophile.com/reference/60/

-k

[Guest GWCrum]

Ken,

Thanks for the link! I wasn't a subscriber back then and wasn't aware of that article, so I didn't know of your involvement. A little egg on my face, I guess.

However, this raises another question. Atkinson modified your circuit, ostensibly to make it more like "real" speakers he had measured. Claiming "the rising impedance above 5 kHz, coupled with a positive phase angle...is almost never found in a real speaker". I just looked at 16 test reports by JA and found 15 have a positive phase angle above 5 kHz and 14 have a rising magnitude of impedance. Although not nearly to the degree of your unmodified simulator. This also agrees with the few measurments I have made on my own speakers. Only one speaker I came across, the Linn Akurate 242 has data that looks like that from JA's simulator.

Would you care to comment further on this issue? I'm not interested in starting a controversy, but I am interested in knowing more about the technology. Specifically, I'm interested in which is more likely to show amplifier differences in the upper frequncies.

Best Regards,

Jerry

[kkantor]

Egg?? No egg! Just pointing you to the reference.

As you know, there is no "right or wrong" about this stuff. Each individual speaker requires a different model, and deciding what to use for generalized amp testing is a matter of experience and opinion. Certainly, JA has a very broad database to base his decision on. I think it is great that he is using the load and reporting the results, and I am happy he personalized it towards his ideal. The load he uses gives the reader a consistent and meaningful comparison between amps, at any rate.

My own feeling is that a highly reactive top end (increasing Z(f)) is probably the harder on most amps than a lower magnitude, but more resistive network. There is never an issue of current supply above 20 KHz, but a capacitive load can flip the phase of the feedback and cause oscillation. Thus, the (so-called) "Zobel" network that Stereophile uses might occasionally make things easier on the amp.

I continue to revise my own loads and simulations, and have several different ones that I use for different purposes. For example:

http://www.aural.org/audio/dummy_load/.

I appreciate your comments.

[soundminded]

"The earliest reference to the effect that I have found was in an article by Johm Tomcik of EV who mentioned it in passing back in 1954."

As luck would have it, there is a comparable thread running on another site and the Tomcik article is referenced.

http://www.paulspeltz.com/tomcik/index.html

I've got a lot of problems with this article and these concepts. The article takes the notion of controlling speaker bass frequency response by controlling amplifier damping factor to another level by discussing adjustment to the negative voltage and current feedback loop levels in the amplifier circuit to accomplish this end. What amplifier engineer would incorporate this into a real amplifier for sale on the market? Who has ever even seen one offered commercially? It's all they can do to get it right when the level is fixed.

For me the problem stems from the fact that there are two separate factors operating simultaneously, mechanical and electrical resonance and the need to damp them out precisely. IMO they should be dealt with separately, not as a single problem. This is where the AS design facilitates the best possible answer as I see it. The mechanical resonance problem is dealt with by tuning the mass, spring constant, and mechanical damping due to velocity related frictional loss to achieve the desired F3 and critical damping of 0.707. This is the approach of AR1 and its descendants and barring any repeal of Newton's second law of motion as applied to forced oscillation, is as true today as it was when Newton discovered it and Villchur applied it. This leaves the electrical resonance problem where it belongs, to be solved with an electrical circuit solution. Compensating for the FR characteristics of the Woofer's in box impedence characteristics (as AR9 did) and/or by active equalization at the preamplifier signal level is merely a matter of applying complimentary equalization. The best amplifier therefore is the one with the highest damping factor without exception and passive or better yet active equalization is applied to obtain the deepest flattest response possible. The best performing subwoofers on the market today use this approach.

[Guest GWCrum]

There were a number of ampluifiers in the 50's and 60's that had adjustable DF, including some of the Marantz tube amps. I can recall others as well (one of the few benefits of being older). Go to the hifilit web site and look at the Marantz literature. I didn't take the time to check, but I seem to recall the Eico HF-50 and HF-60 having a knob to adjust DF, as well; and I'm certain there were others.

Your comments that highest DF is best, begs an argument. The very high numbers for some of the 1970's amps came from very large amounts of NFB. Nice for advertising, but it can cause a host of other problems and a DF of 1000 isn't needed for anything but advertising. In any case we should be talking about source impedance of the amp, since DF is a somewhat arbitrary, calculated "figure of merit"; and source impedance is a real, measureable parameter that has effects over the entire frequency range because of the "voltage divider effect". DF tends to focus one's attention solely on the issue of damping of the bass resonance. In that respect, a DF higher that 25 has no effect; while the "voltage divider effect" can have effects at fairly low source impedances (higher DF's). Also, if the speakers were designed for use with an amp with a high source impedance, than a low source impedance (high DF) can significantly alter the sound of the speaker, moving it away from the designer's intentions.

I know you were speaking from a standpoint of an ideal situation, but I'm not sure everyone will pick up on that. In most real cases, it seems we need to look for an optimum match of properties rather than chase an ideal.

Jerry

[Guest GWCrum]

Ken,

The idea of "a higher capacitive load can flip the feedback" made me think. Can this occur with capacitive loading on the outputs of line level devices such as a CD player or preamp output? And particularly those with linear IC outputs.

I ask, since I have found that with certain CD players, going to lower capacitance interconnects has noticeably smoothed an overly bright or edgy top end and I have wondered about the possible causes.

Regards,

Jerry

[soundminded]

>There were a number of ampluifiers in the 50's and 60's that

>had adjustable DF, including some of the Marantz tube amps. I

>can recall others as well (one of the few benefits of being

>older). Go to the hifilit web site and look at the Marantz

>literature. I didn't take the time to check, but I seem to

>recall the Eico HF-50 and HF-60 having a knob to adjust DF, as

>well; and I'm certain there were others.

>

>Your comments that highest DF is best, begs an argument. The

>very high numbers for some of the 1970's amps came from very

>large amounts of NFB. Nice for advertising, but it can cause

>a host of other problems and a DF of 1000 isn't needed for

>anything but advertising. In any case we should be talking

>about source impedance of the amp, since DF is a somewhat

>arbitrary, calculated "figure of merit"; and source

>impedance is a real, measureable parameter that has effects

>over the entire frequency range because of the "voltage

>divider effect". DF tends to focus one's attention

>solely on the issue of damping of the bass resonance. In that

>respect, a DF higher that 25 has no effect; while the

>"voltage divider effect" can have effects at fairly

>low source impedances (higher DF's). Also, if the speakers

>were designed for use with an amp with a high source

>impedance, than a low source impedance (high DF) can

>significantly alter the sound of the speaker, moving it away

>from the designer's intentions.

>

>I know you were speaking from a standpoint of an ideal

>situation, but I'm not sure everyone will pick up on that. In

>most real cases, it seems we need to look for an optimum match

>of properties rather than chase an ideal.

>

>Jerry

IMO, it is a very poor strategy to rely on the mechanical and electrical resonance of speakers to produce bass. Speaker drivers are inherently resonant devices to begin with and when a woofer is put into an enclosure, that only makes matters worse. The goal of a high fidelity loudspeaker system is to act like it is a single non resonant entity in its passband. This is best achieved by damping out resonances electrically and mechanically and then equalizing for any remaining FR nonlinearities.

I said I hadn't seen any amplifier with an adjustable feedback loop. Negative feedback when properly applied does much more than just reduce the damping factor, it extends and flattens frequency response and reduces noise and distortion all at the expense of gain. It also adds stability to performance. Non feedback amplifiers generally are poor performers and are not stable with changes in voltage or ambient temperature. This is especially true for vacuum tube amplifiers where small changes in filament current can make a big difference in thermionic emission. Sometimes audiophiles wait for hours for them to stabalize, sometimes they never settle down. Even with feedback, vacuum tube amplifiers rarely if ever achieve less than 1% non linear distortion at any power level, often a lot higher. And you can see how sensitive they are to different speaker loads from the referenced article. The best solid state amplifiers achieve virtually unmeasurable noise and non linear distortion, ultra flat and extended frequency response, are completely stable over a very wide range of operating conditions, and shrug off even the most difficult elecrical loads. This last aspect is in part due to their very high damping factors which are indicative of massively oversized bulletproof power supplies. When feedback is poorly applied, that is another story and amplfiers can be marginally stable exhibiting all types of transient distortion and sometimes even being driven into oscillation which amplifier designers of yore called "motorboating." The absolute criteria for stability is that the loop gain must be less than unity when the phase angle is minus 180 degrees but there are degress of stability and an improperly designed feedback circuit may be worse than none at all. When I went to school, those who taught feedback theory called it the greatest advance in amplifier design since the invention of the triode. That's probably still true. Every modern control system relies on it completely.

[kkantor]

You might be interested in the first listed paper, given your background and interests.

Just FYI. No editorial.

http://www.aes.org/journal/toc/AES-JanFeb2006TOC.cfm

[kkantor]

Generally not, though there will probably always be pathological circuit designs...

Why? Because a line level driver should, (and all I know of do), include a few tens or hundreds of Ohms in series with the output to isolate the load reactance, protect from ground differentials, etc. This simple approach is not possible to apply to a power amp, and so it is common to insert a small inductance in series with the output, instead. I learned this building my first Dynaco amp. It used a few turns of wire wrapped around the output coupling caps for this purpose.

Harsh CD player? Hmmmm... should never be. I would suspect an improper de-emphasis flag or filter. Trade it in!

[kkantor]

I'm afraid I don't understand what you mean by "electrical" resonance in a speaker, that is not "mechanical" in origin. Both Qes and Qms originate in the mass/compliance system, as far as I can tell. They are differentiated only for the purposes of modelling the physical system, BL coupling, V-I transduction, etc. There is no meaningful energy storage in the form of charged particles or time-varying magnetic fields, at least anywhere near "resonance."

[kkantor]

I concur with your overall take on negative feedback. (I guess that sentence would be positive feedback...)

Just for the record, there are some patents granted and pending that alter the feedback characteristics in such a way that the amp behaves more like a current source below a certain frequency, and a voltage source above it. This is an attempt to achieve the electrodynamic linearity benefits that come with a current source drive, while maintaining better stability. This can really only be practically applied in an "active" speaker, where the designer can apply known frequency equalization ahead of the power stage.

[soundminded]

>I'm afraid I don't understand what you mean by

>"electrical" resonance in a speaker, that is not

>"mechanical" in origin. Both Qes and Qms originate

>in the mass/compliance system, as far as I can tell. They are

>differentiated only for the purposes of modelling the physical

>system, BL coupling, V-I transduction, etc. There is no

>meaningful energy storage in the form of charged particles or

>time-varying magnetic fields, at least anywhere near

>"resonance."

As I see the problem, I choose to break it down into two separate sub problems. If the speaker were a non electrical device having exactly the same mass and configuration, it would still be a mechanically resonant system in the classic spring, mass, dashpot model. This can be tuned in the same way as an automobile suspension is tuned to achieve a desired response to excitation (the automotive engineer has the additional problem of a variable mass.) By substituting the magnet and voice coil for the non electrical and magnetic components, a new element is introduced which is also frequency selective. This as I see it is its electrical resonance. The tendency to not segregate the problem or treat it as one which can be solved mosty by electrical solutions is IMO a disadvantage. But when you are trained as an electrical engineer, that may be the instinctive thing to do. When you are a hammer, does every problem become a nail?

[Guest GWCrum]

I think you misread the intent of my comments. I am not in favor of non-NFB amps, or amps with higher source impedances. I was simply pointing out the effects and the ranges of values over which they occur. Surely you don't believe all power amps should have DF's of 1000. (for what it's worth, the amps I mostly use are SS and have DF's of 87 and 180) I'm also quite aware of the range of effects of NFB on amplifier performance.

I'm also trying to point out that the effects of amplifier source impedance are full range (20 Hz to 20 kHz) and are primarily frequency response effects, NOT "damping" effects.

By the way, I am a mechanical engineer by education, but have a decent understanding of electrical and electronic principles.

Since the electrical and mechanical parameters of a speaker are coupled so tightly, it makes litle difference whether they are handled in the electrical or mechanical domain. You point out a car suspension, but increasingly, many vehicles have electro-mechanical elements in the suspension. Formula 1 cars pioneered in this with their active suspensions (now banned), but some passenger cars now employ the technology. So even there, "mechanical" problems are being handled on the electrical side. Perhaps a better example was the "ACE Bass" circuitry in the Audio-Pro B2-50 subwoofer. Unlike car suspensions or motional feedback woofers, which are closed loop; the ACE Bass circuit was an "inverse plant controller" where parameters like cone mass were added in the equivalent circuit without feedback from the mechanical system. This resulted in performance not attainable in conventional woofers, and at least as good as I have heard from motional feedback woofers. One of the two or three best bass performaces I have ever heard.

Jerry

[soundminded]

>I think you misread the intent of my comments. I am not in

>favor of non-NFB amps, or amps with higher source impedances.

>I was simply pointing out the effects and the ranges of values

>over which they occur. Surely you don't believe all power

>amps should have DF's of 1000. (for what it's worth, the amps

>I mostly use are SS and have DF's of 87 and 180) I'm also

>quite aware of the range of effects of NFB on amplifier

>performance.

>

>I'm also trying to point out that the effects of amplifier

>source impedance are full range (20 Hz to 20 kHz) and are

>primarily frequency response effects, NOT "damping"

>effects.

>

>By the way, I am a mechanical engineer by education, but have

>a decent understanding of electrical and electronic

>principles.

>

>Since the electrical and mechanical parameters of a speaker

>are coupled so tightly, it makes litle difference whether they

>are handled in the electrical or mechanical domain. You point

>out a car suspension, but increasingly, many vehicles have

>electro-mechanical elements in the suspension. Formula 1 cars

>pioneered in this with their active suspensions (now banned),

>but some passenger cars now employ the technology. So even

>there, "mechanical" problems are being handled on

>the electrical side. Perhaps a better example was the

>"ACE Bass" circuitry in the Audio-Pro B2-50

>subwoofer. Unlike car suspensions or motional feedback

>woofers, which are closed loop; the ACE Bass circuit was an

>"inverse plant controller" where parameters like

>cone mass were added in the equivalent circuit without

>feedback from the mechanical system. This resulted in

>performance not attainable in conventional woofers, and at

>least as good as I have heard from motional feedback woofers.

>One of the two or three best bass performaces I have ever

>heard.

>

>Jerry

Unlike the results of wine tasting, I don't consider amplifier performance as a matter of opinion. I think there is an idealized model of what an amplifier is supposed to do, objective design criteria, and performance measurements to judge how real amplifiers stand up to those ideals. It is true that in many cases one amplifier may be better in one aspect of its performance and another amplifier better in some other aspect. For instance, amplifier A may have flatter FR than amplifier B but B may have less harmonic distortion than amplifier A. Which is more valuable? depending on the particular circumstances, that could be debatable. I don't think adjusting the output impedence either by adding resistors or by adjusting internal amplifier feedback is a good way to compensate for shortcomings in a speaker's FR. A graphic and/or parametric equalizer is the best tool for that as I see it. That is its sole purpose and when properly used, even modest units can do surprisingly well.

[kkantor]

I >am< trained as an electrical engineer. That is why I know the term "electrical resonance" has a specific and broadly accepted meaning. You cannot simply say that any frequency dependent current flow is a "resonance." There is no meaningful electrical resonance in a woofer/box system, in the sense of energy being cyclically stored as charge and magnetic field.

Anyway, this thread is well past its -3 dB point, and you may have the final word. I only urge you to choose your technical terms with some rigor, to avoid a situation in which future readers with less advanced scientific background begin to adopt incorrect terminology. Amateur speaker building is somewhat notorious for this problem.

[Guest Terry Again]

I >am< trained as an electrical engineer. That is why I know the term "electrical resonance" has a specific and broadly accepted meaning. You cannot simply say that any frequency dependent current flow is a "resonance." There is no meaningful electrical resonance in a woofer/box system, in the sense of energy being cyclically stored as charge and magnetic field.

Anyway, this thread is well past its -3 dB point, and you may have the final word. I only urge you to choose your technical terms with some rigor, to avoid a situation in which future readers with less advanced scientific background begin to adopt incorrect terminology. Amateur speaker building is somewhat notorious for this problem.

Any Chance you guys can post what this all means to a regular Joe that likes good sound and is willing to modify speakers to get it? It might help me to know when I'm being BS'd by sales people of speakers? Or in buying a new amp/receiver? I can sense this post has something in it that I need to know or should do but I'm lost and wondering I've heard that there's a difference between a tube amp and Solid State amp which I assume is the SS mentioned? But I never have heard one myself to see if I can tell. I was born in 54 so missed I think the time of Tubes?

Thanks for a reply and education of the brain damaged,

Terry

I'm brain Damaged,what's your excuse?

[dynaco_dan]

Any Chance you guys can post what this all means to a regular Joe that likes good sound and is willing to modify speakers to get it? It might help me to know when I'm being BS'd by sales people of speakers? Or in buying a new amp/receiver? I can sense this post has something in it that I need to know or should do but I'm lost and wondering I've heard that there's a difference between a tube amp and Solid State amp which I assume is the SS mentioned? But I never have heard one myself to see if I can tell. I was born in 54 so missed I think the time of Tubes?

Thanks for a reply and education of the brain damaged,

Terry

I'm brain Damaged,what's your excuse?

Hi Terry;

First, I will ask that you please do not continue to use the, "brain damage", comment.

You may be a happy, jolly fellow but today this is an insensitive comment.

Second, there is no quick lesson on your request.

There is a lot of members here and everyone is still learning.

Some have tube equipment.

Some have solid state equipment.

Some have both.

Some may argue that theirs is the best, whatever it may be.

Rather a fun and diverse group of people here.

All will voice their opinion in what they like about one or the other or both even.

Reading all the topics here and at other websites will enrich your view.

What I dislike, you may love.

It is better if you post a specific question or two than ask such a vast question.

You should also allow at least two weeks for everyone to see your questions and respond, rather than just over night.

I have only been interested in hifi since about 1965 and I was out of the picture for about 20 years.

A few opportunities came my way that not everyone has had, I was very lucky.

I do not know very much but I will try to help you, as you ask, as will other members.

By the way, Terry, Ken Kantor, is by far the most capable member here to comment, with extreme credibility, on anything technical.

[Guest Terry Again]

Hi Terry;

First, I will ask that you please do not continue to use the, "brain damage", comment.

You may be a happy, jolly fellow but today this is an insensitive comment.

Second, there is no quick lesson on your request.

There is a lot of members here and everyone is still learning.

Some have tube equipment.

Some have solid state equipment.

Some have both.

Some may argue that theirs is the best, whatever it may be.

Rather a fun and diverse group of people here.

All will voice their opinion in what they like about one or the other or both even.

Reading all the topics here and at other websites will enrich your view.

What I dislike, you may love.

It is better if you post a specific question or two than ask such a vast question.

You should also allow at least two weeks for everyone to see your questions and respond, rather than just over night.

I have only been interested in hifi since about 1965 and I was out of the picture for about 20 years.

A few opportunities came my way that not everyone has had, I was very lucky.

I do not know very much but I will try to help you, as you ask, as will other members.

By the way, Terry, Ken Kantor, is by far the most capable member here to comment, with extreme credibility, on anything technical.

Cool that you stand up for the disabled!! I am Brain Damaged from a car accident dec.21,1999 in which my 7 year old son and his 5 year old sister recieved minor brain damage. I had 3 brain surgerys one to remove the right side perriatol(sp) lobe and two more for bleeding deep hemotomas.Being my age at the time which was in my 40's I wasn't given much chance to live. But here I am today!! I may have to take a large fist full of expensive medication to control seizures and depression Heavy on the depression side because of grief of loss of a child and loss of what I used to be able to do!! The one thing that gets and got through all of this is a big sense of humour!! I think one thing this great site could use is a forum where one has to post what and who they are before they can make any other posts?! and please no apoligese needed IF you feel one is needed?

I've heard before that tube amps had a warmer sound from them and I think these posts are about getting that from SS amps by a mod to speakers? I would like to update the drivers in my AR 2ax speakers for digital cds and what not yet explore wether I could hear a diff or not or anyone else in my family might hear? Or if there's a placebo effect here? I know from Ken's other posts and his work experience is very knowing about sound!! It's just the in depth tech speak loses me is all. I have to learn how to use this site's features let alone the new to me tech speak of speakers! I'm glad I found this great site though I can't remember how? my short term memory is almost nonexistant as well as my spelling and puncuation. Though a lot of people on the net seem to have that problem!!:-) So now I wait to see IF Ken joins in to help?

Thanks,

Terry

[dynaco_dan]

Cool that you stand up for the disabled!! I am Brain Damaged from a car accident dec.21,1999 in which my 7 year old son and his 5 year old sister recieved minor brain damage. I had 3 brain surgerys one to remove the right side perriatol(sp) lobe and two more for bleeding deep hemotomas.Being my age at the time which was in my 40's I wasn't given much chance to live. But here I am today!! I may have to take a large fist full of expensive medication to control seizures and depression Heavy on the depression side because of grief of loss of a child and loss of what I used to be able to do!! The one thing that gets and got through all of this is a big sense of humour!! I think one thing this great site could use is a forum where one has to post what and who they are before they can make any other posts?! and please no apoligese needed IF you feel one is needed?

I've heard before that tube amps had a warmer sound from them and I think these posts are about getting that from SS amps by a mod to speakers? I would like to update the drivers in my AR 2ax speakers for digital cds and what not yet explore wether I could hear a diff or not or anyone else in my family might hear? Or if there's a placebo effect here? I know from Ken's other posts and his work experience is very knowing about sound!! It's just the in depth tech speak loses me is all. I have to learn how to use this site's features let alone the new to me tech speak of speakers! I'm glad I found this great site though I can't remember how? my short term memory is almost nonexistant as well as my spelling and puncuation. Though a lot of people on the net seem to have that problem!!:-) So now I wait to see IF Ken joins in to help?

Thanks,

Terry

Hi Terry;

Thank you for clearing up what I thought was an insensitive comment.

Sorry for all of your losses.

At least you are here to enjoy and contribute to this wonderful site.

I am disabled also.

I have been given a second chance at living for the past 3+ years.

Every single day is a bonus for me.

A visit from James, the drafting monkey, and his charming wife at Christmas time 2006 was a super surprise and most welcome.

I can only wish that something like that happens to everyone here.

After all of my experiences, I still do not understand everything written here.

The layout of this site is very fairly good and only of recent design.

I get lost real easy so don't worry about your questions, we will try to answer them.

[soundminded]

Hi Terry;

Thank you for clearing up what I thought was an insensitive comment.

Sorry for all of your losses.

At least you are here to enjoy and contribute to this wonderful site.

I am disabled also.

I have been given a second chance at living for the past 3+ years.

Every single day is a bonus for me.

A visit from James, the drafting monkey, and his charming wife at Christmas time 2006 was a super surprise and most welcome.

I can only wish that something like that happens to everyone here.

After all of my experiences, I still do not understand everything written here.

The layout of this site is very fairly good and only of recent design.

I get lost real easy so don't worry about your questions, we will try to answer them.

Funny how this thread seemed to have come back from the clouds after nearly two years of absence.

I haven't researched it but I have given some thought to the problem of the model from time to time and I am still convinced that the model which prompted the original thread is wrong. This is demonstrated easily by the fact that the reverse EMF will be out of phase with the applied EMF which the model makes no account for. It also depends on a whole host of factors which includes circuit elements describing analogies to the mechanical paramaters in both the storage and release of potential energy and the momentum (kinetic energy) of the moving mass (cone and voice coil.) It also has to take into account the reconversion efficiency of the speaker as an electrical generator. In an accurate model, there will likely be a voltage controlled voltage source in shunt across the voice coil, and undoubtedly with a complex impedence in series with it. Some of the elements in this source will be non linear since the amount of energy stored as potential energy in the compressed air will increase as a percentage of total volume in a small sealed box as the magnitude of compression increases. This means that the greater the displacement of the cone at any given time, the harder the motor has to work to obtain an equal displacement since the percentage of air displaced with increasing compression increases.

There are ways to get an approximate idea of what this would be in real terms. One way would be to energize only one voice coil of a dual voice coil subwoofer and measure the open circuit voltage at the terminals of the second voice coil. You would then plot amplifier voltage applied to the energized coil versus output voltage of the non energized coil. It wouldn't surprise me a bit if the relationship between the two is both frequency and amplitude dependent. Another way is to put a high value resistor in series with the speaker and measure the voltage across this resistor. This would give the difference between the amplifier output voltage and the speaker input voltage which is in part accountable to the back EMF, the voltage due to the resistive drop having to be subtracted. The high value of the resistor does not let the amplifer damp out the back EMF as readily as a low impedence wire alone which we normally use. This is why even with a 1/2 or 1 ohm resistor, the difference is audible, the back EMF experessing itself as the result of stored energy slowly being reconverted to heat in the amplifier output circuit, the wire, and the series resistor as the cone continues to oscillate from the stored energy. It probably favors oscillating at the speaker's mechanical resonance frequency regardless of the driving frequency. This would be another highly non-linear element.

As coincidence would have it, I am researching energy storage motor/flywheel/generator substitutes for chemical storage batteries for UPSs which supply power to computers and other loads when power from the utility fails. These are in many ways almost exactly analagous to this problem although the generator will be kept in phase with the same voltage driving the system. If I come across a useful model, I'll reference it.

As usual I disagree with just about eveything John Atkinson wrote that I have ever read and the referenced article is no exception.

"As mentioned by two readers in this month's "Letters," amplifiers are used to drive loudspeakers but are almost exclusively measured into resistive loads. The reasons for this are twofold: 1) real loudspeakers both produce neighbor-annoying sound levels and tend to break when driven with typical amplifier test signals; and 2) the question as to which "standard" loudspeaker should be used is impossible to answer---at least the conventional resistive loads are consistent and repeatable. "

This is about as dumb an answer as anything I could think of. Loudspeakers are tested in laboratories often deep in the interiors of industrial buildings and in rooms which are acoustically isolated even from those in adjacent areas. If Atkinson is tesing speakers for his magazine in his basement or a spare bedroom, perhaps that explains why I don't read his magazine. As for damaging loudspeakers by testing amplifiers with them, how does he explain that he doesn't damage them when he listens to them? And which loudspeaker do you use to test an amplifier with? How about some representative generic types, especially the more popular or difficult ones. You certainly wouldn't want to use a small 2 way system because the amount of energy stored resulting from the woofer would be relatively small compared to other speakers....like AR3. When you test equipment, you not only want to show what is typical under likely operating conditions, you want to see what equipment does at its worst, not just its best and you want tests which will show meaningful difference between competing designs, not make them all look the same...unless you are in the business of helping advertisers sell amplifiers.

The real reasons that amplifiers aren't tested with real loads is that historically testing with a resistor at 1 watt showed us much about what we needed to know about amplifiers, their FR so poor and so different from one another that was all that was needed. Power bandwidth didn't matter much either in the days when recordings had little or no deep bass and speakers couldn't produce it if they did. So we live with these test methods of the 1930s and 1940s in an entirely different era. If FR were measured at full power and below, the notion of slew rate as applied to amplifies for example would never have come up as an issue, the slew rate being a deliberate obfuscation of the simple notion that an amplifier's high end FR can change for the worse as power output demanded of it increases. The terminology was stolen from op amp designers and users where it had some useful meaning. In the consumer audio amplifier industry, it just became a sales slogan. The real reason they don't test with speakers? 1. They are lazy. 2. It doesn't make amplifiers look good, it makes them look bad. Put a 100 hz square wave into an amplifier whose output is connected across a resistor and it looks pretty good. Now connect a small woofer and watch it tilt, the more the tilt, the worse the amplifier usually due to increased power supply impedence as the result of a puny transformer. Besides, it takes a lot more than a single number or even a few numbers to really understand what an amplifier's performance capabilities are and aren't. The good news is that most solid state amplifiers really do sound pretty much alike (as Ken said) because building a good one has become very cheap and easy to do.

BTW, I'm also trained as and have a degree in Electrical Engineering.