Jump to content

New hardware


John Warren

Recommended Posts

Deano wrote: I posted the old link because John asked a question. The question was in response to a statement by Al. The statement was this:

"The very best way to transfer power is to match impedance."

I tried for a decade to understand this, and I used to tell Al, quite often actually - that John is not a dumb guy, that a guy who spends his life studying motors, might be trying to teach you something important.


John Warren to Al:

'"You must explain to me. To obtain a linear spl [sic] as a function of frequency a loudspeaker(s) cone must have a constant acceleration, independent of frequency, over the entire bandwidth of operation. That criteria is fact and is a topic that has been discussed for many years in the literature (Small, Locanthi as examples).

 

Small, in fact, has a paper that explains how to preserve a constant voltage transfer in a passive network, the necessary requirement for constant cone acceleration. He states, "consideration of the electroacoustic behavior of the common loudspeaker drivers leads to the general crossover network design requirement of constant total voltage transfer." (Proc. IREE Australia 31, 1970)

That said, how can a constant power transfer satisfy this criteria?

That force that drives the cone is F = (Bl) E / (Zc + ((Bl)2/Zm))


where: 
B = flux density in gap
l = length of wire in gap
Zc = dc resistance of coil and coil inductance
Zm = back EMF
E = driving voltage

What Small is saying is that above resonance (i.e. in the operating range), the force driving the cone is "essentially" a linear function of the driving voltage E.

Since the sound pressure is F/Sd (where Sd is the radiation area), the sound pressure then becomes a linear function of the driving voltage.

And like I said before, for a fixed sound pressure (i.e. flat fequency [sic] response) the only thing that is needed is a constant acceleration.

You fail to see that the integration of filter theory REQUIRES A FUNDAMENTAL UNDERSTANDING of the interelationship [sic] between the acoustic mass, the electrotransduction [sic] factor BL and how it is reflected back to the amplifier. You deal with everything from the "microwave" world which, last I looked, has NOTHING whatsoever [sic] to do with sound reproduction.

I'll assume that you will NEVER be able to explain why the "constant impedance" network is an advantage from a SPL vs. frequency perspective and not a "microwave" perspective."' 

 

John,

First, let’s examine your premise that in order to have a “linear SPL as a function of frequency a loudspeaker’s cone must have a constant acceleration, independent of frequency, over the entire bandwidth of operation.”

 

After reviewing my technical library of books and papers, I could find no example where the term “acceleration” was used to describe the motion of a loudspeaker cone. Acceleration is not even in the indexes of Harry Olson’s Applied Acoustics or Leo Beranek’s Acoustics . and these guys literally “wrote the book” on acoustics, loudspeakers, measurement, etc.

 

I believe that in the context of our discussion here, the proper term to express the motion of a loudspeaker cone is velocity, not acceleration. I’ll leave it to the Gentle Reader to look up the difference between the two related terms.

 

Olson writes, “The acoustic power output is the product of the resistive component of the air load and the square of the cone velocity.” Beranek writes, “The performance of a direct-radiator loudspeaker is directly related to the diaphragm velocity. Having solved for it, we may compute the acoustic power radiated and the sound pressure produced at any given distance from the loudspeaker in the far field.” (emphases mine)

 

A loudspeaker cone can only experience an acceleration if the velocity vector changes amplitude or direction. As you can see from the attached nomograms, there is no way a loudspeaker cone can experience “a constant acceleration”. It doesn’t matter whether the loudspeaker is reproducing a steady tone or a movement from Carmina Burana, there’s no “constant acceleration”.

 

With a 2 kHz tone, our displacement limited loudspeaker cone is experiencing a velocity of 10 m/sec away from the magnet for the first half of our 2 kHz sinewave, stops and undergoes a 10 m/sec velocity towards the magnet and repeats. The acceleration is constantly changing with either a positive or negative vector—not unipolar as your wording would imply. Or am I missing something here?

 

Second, you wrote: Small, in fact, has a paper that explains how to preserve a constant voltage transfer in a passive network, the necessary requirement for constant cone acceleration.

Please explain why a constant voltage transfer in a passive network is a requirement for a constant cone acceleration. I just explained why a loudspeaker cone cannot experience a constant acceleration. Or am I missing something here?

 

Third, Small’s paper is only valid for a two-way loudspeaker system and only if the crossover network is a first-order (6 dB/octave slope). Knowing that, it imposes unrealistic constraints on someone wanting to connect two loudspeakers using his “constant voltage” crossover network scheme. Small writes, The derivation of (equation 1) assumes two conditions: that the drivers are mounted closely together and that they are identical. Oh, and don’t forget, “…the drivers used must have a useful frequency ranges which overlap by about four octaves. Huh??

 

John wrote, I'll assume that you will NEVER be able to explain why the "constant impedance" network is an advantage from a SPL vs. frequency perspective and not a "microwave" perspective.” 

 

In the Conclusion of his paper, Small writes, For systems having unavoidably large driver spacings, there is no perfect crossover design. Intuition suggests, however, particularly when room reverberation is considered, that in this case constant power transfer (constant-resistance) networks would provide the best results on average.

 

Looks like to me Small has just vindicated Al’s use of constant power transfer networks. Or am I missing something here?

 

Finally, in the words of Al K., The requirements for filters to work at microwave are so stringent that a microwave design will operate easily at audio.  The designs that can get by at audio will fail miserably at microwave. It shouldn't take a genius to see why microwave standards applied to audio is an improvement.

 

Lee

 

post-2553-0-64300000-1446927787_thumb.jppost-2553-0-77980000-1446927799_thumb.jpConstant-Voltage Crossover Network Design.PDF

Link to comment
Share on other sites

The first part of my post was taken from Dean's re-posting of an older thread in which John Warren asked Al about his networks. I guess Dean thought he'd embarrass Al by bringing up a ten-year-old thread.

 

The balance of my post was simply an attempt to refute the assertions John made back then. They are still relevant today.

 

Lee   

Link to comment
Share on other sites

SP:  After reviewing my technical library of books and papers, I could find no example where the term “acceleration” was used to describe the motion of a loudspeaker cone. Acceleration is not even in the indexes of Harry Olson’s Applied Acoustics or Leo Beranek’s Acoustics . and these guys literally “wrote the book” on acoustics, loudspeakers, measurement, etc.

 

Perhaps you need a better library.  To start…

 

For the “gentle” reader (your words, not mine for damn sure) I provide a link that is of interest.  Please read the paragraph associate with Rice and Kellog, two GE engineers working on loudspeakers in the 1920s.

http://acousticsresearchcentre.no/a-brief-history-of-electroacoustics-pt-5-moving-coil-loudspeakers-of-lasting-impact-2/

 

 “..They soon realized that the loudspeaker would produce a flat frequency response if operated above its fundamental resonance. In the low frequency range, the radiation resistance is very low, and the diaphragm mass is the dominating factor. The loudspeaker operates in a constant acceleration mode, and since the radiation resistance at the same time increases, it will produce a flat frequency response....

 

Perhaps you’ve heard of servo-controlled loudspeakers?  Guess how they work, accelerometers mounted to the cone providing feedback to the system to maintain constant acceleration which yields flat frequency response:

 

“...The Genesis servo bass system reduces this distortion to below one percent at almost any output level. The system also drives the woofer to constant acceleration, which makes the frequency response of the woofer anechoically flat to the lowest frequencies...”.

 

http://www.genesisloudspeakers.com/tech_servobass.html

 

Audioholics has a nice treatment

 

“...Between fl and fh is the mass-controlled reference region of the driver. In this portion of its output spectrum, the cone is mass-limited and the driver functions as a constant acceleration device, producing constant volume velocity...”.

 

http://www.audioholics.com/editorials/five-5-audio-myths-dispelled/five-5-audio-myths-dispelled-page-3

 

Do you believe what you read in the Handbook for Sound Engineers?.

 

p.s. I've only just started...and yes, Beranek does mention just not in Acoustics

Edited by John Warren
Link to comment
Share on other sites

As you can see from the attached nomograms, there is no way a loudspeaker cone can experience “a constant acceleration”. It doesn’t matter whether the loudspeaker is reproducing a steady tone or a movement from Carmina Burana, there’s no “constant acceleration”.

 

 

 

Yet, for many systems it is, in fact, a reality.

 

Please extend your reading material to beyond two publications.

Link to comment
Share on other sites

In the Conclusion of his paper, Small writes, For systems having unavoidably large driver spacings, there is no perfect crossover design. Intuition suggests, however, particularly when room reverberation is considered, that in this case constant power transfer (constant-resistance) networks would provide the best results on average.

 

Looks like to me Small has just vindicated Al’s use of constant power transfer networks. Or am I missing something here?

 

Lee

 

The paper was written in 1968 for a conference he attended in 1969 so tubes were still the dominant.

 

That said, good net design attempts to optimize frequency response letting the impedance "fly".

 

 

 

 

 

 

 

post-864-0-53340000-1446951353_thumb.png

  • Like 1
Link to comment
Share on other sites

The old thread reveals the primary technical point of contention between John and Al, containing the question which was never resolved. I was also trying to help those following the thread by providing some history and relevant information -- I wasn't trying to embarrass Al.

Link to comment
Share on other sites

A new addition to Al's site/

"Since I have learned that another vendor that I foolishly helped has claimed authorship of one of my designs calling it the "Super X", I am willing to share it with the DIY community. It was initially called the "Super AA" and was intended to be an economical rework of the Klipsch "AA". It changes the part values in the tweeter filter and adds an inductor and a resistor. The original T2A transformer of the AA was retained but it could be replaced with the Crites 3636 to give you 1 dB attenuation steps."

I thought the Universal was a "rework of the AA"?

This is a new low for Al. I told the story of how this network was developed on the forum many times, and never once did Al dispute it. I suggested the design change in a phone conversation, and upon request, sent a schematic. Al did an analysis on it and tweaked the tweeter filter. Calling it his design is nothing less than a lie, and anyone here with half a memory knows it.

I feel sorry for you Al.

Link to comment
Share on other sites

I know what happened, he knows what happened. He used the program to adjust the values in the tweeter circuit after I sent him my schematic. You know, I really don't care, it's not important. What is important is that the both of you are running around loose without medication.

I wonder if Al remembers why he doesn't have to solder all of those leads to the autoformer anymore?

hi Dean... do you mean that it was your idea to go from this :

post-14752-0-57940000-1446982362_thumb.j

 

to this :

post-14752-0-25300000-1446982410_thumb.j

 

excellent !....

 

also, remember, with regard to the Super X, "imitation is the greatest form of flattery"

 

also, also.. I have been using your wax impregnated 3670... anyone using a 3619 or 3619-ET should try it... immediate, noticeable improvement....

Link to comment
Share on other sites

Btw, constant voltage creates flat on-axis response for a piston, but this results in a total system power that decays as frequency increases. This is because the high frequencies beam. We see this effect in the rising electrical impedance due to the mass corner and voice coil impedance.

This means an ideal system doesn't use a flat voltage profile. This is also why a flat on-axis response isn't ideal for a system with beaming polars. The most common place to implement this is in the xover. You can't do it with the speaker design, and you don't want to add an additional line level stage to do it either (unless running active xovers).

John is the only guy here focusing on the acoustic response.

You other three couldn't voice a system to save your life.

Link to comment
Share on other sites

Mike,

 

You wrote, John is the only guy here focusing on the acoustic response. You other three couldn't voice a system to save your life. 

 

If the attached plot from John's website is an example of his voicing skills, then, yeah, we three could probably do as well. That response droop is where our hearing is most sensitive and is somehow caused by the horn+driver+network interaction.

 

Here's what John wrote about the response "anomaly", Are the responses perfect?  No.  Can I eliminate that response anomaly? Yes but it will add to the package and that adder is about $200 to the overall cost.  Maybe I'll add it as another network option. FWIW, I've measured a lot of horns including ones made by folks that post on this site.  I've also measured the response on networks made by others.  From where I'm at, this setup can compete with any of them.

 

Lee

 

post-2553-0-95980000-1446999138_thumb.jp

Link to comment
Share on other sites

No. Most of that post was from the older thread I had cited earlier, that included the heated exchange between John and Al. Lee just just lifted most of the thread and dropped it into his post. There was nothing relevant from Lee until the end, which was actually just another Lee posting for Al response.

Link to comment
Share on other sites

  • Moderators

Posting by "proxy" is not allowed on this forum. I'm hoping members are not engaging in this type of activity. Posting of messages between parties, private emails etc are also not allowed.

There will be no more verbal/written warnings in this thread. Next step is nine warning points or being banned.

Edited by dtel's wife
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...