Klipschorn bass response

[bodcaw boy]

Also, what were the measurement parameters for the AK-4 LF response plotted earlier (input power, distance, etc)?

jw

in the corner of the chamber door, mic 3 meters away and 2.83 volts.

roy delgado

[DrWho]

Was the graph normalized to read true SPL? So add about 10dB to compensate for the 3 meter measurement to get the sensitivity? At first I was a bit shocked until I noticed how far away you were measuring... []

[D-MAN]

Khorn bass channel splay angles and the height of the high frequency units makes for requiring a complicated measuring methodology.

The distance is required for the LF waveform to propagate to the optimum mouth size within the 1/8th space environment outside the enclosure, for one thing.

Measuring at the "apparent" mouth/channel terminus would not work either, of course, and would also lack the top-end propagation paths.

The 3 meter measurement result is then calculated for an approximation of what it would/should be at 1 meter.

DM

[bodcaw boy]

Roy,

I'm glad you posted the Klipschorn standard LF graph.

Now, how about the midrange and treble?

We have seen the ones from the late '80s that Klipsch used to include on the spec sheets, and the ones PWK published in the 70s and before. I, for one, would be interested to see how the AK-4 models look different, while expecting only minor changes.

We understand that Klipsch's sweeps are not going to be as flat looking as the 1/3 octave noise plots some others run. Wasn't one of Klipsch's Laws that "All speakers measured with noise sources will be flat?" We have seen so many DIYs' plots reproduced here that we won't be shocked! We can take it! We realize that frequency response is not everything, and many of us were attracted to horns by the clean, crisp, effortless sound, rather than FR response. You guys are the ones with the revolving door, corner containing, anechoic chamber, and the great measuring gear, I sure would like to see how a modern Klipschorn measures in there in the midrange and treble!

here you go....

roy delgado

khornhf1.PDF

[bodcaw boy]

Roy,

I'm glad you posted the Klipschorn standard LF graph.

Now, how about the midrange and treble?

We have seen the ones from the late '80s that Klipsch used to include on the spec sheets, and the ones PWK published in the 70s and before. I, for one, would be interested to see how the AK-4 models look different, while expecting only minor changes.

We understand that Klipsch's sweeps are not going to be as flat looking as the 1/3 octave noise plots some others run. Wasn't one of Klipsch's Laws that "All speakers measured with noise sources will be flat?" We have seen so many DIYs' plots reproduced here that we won't be shocked! We can take it! We realize that frequency response is not everything, and many of us were attracted to horns by the clean, crisp, effortless sound, rather than FR response. You guys are the ones with the revolving door, corner containing, anechoic chamber, and the great measuring gear, I sure would like to see how a modern Klipschorn measures in there in the midrange and treble!

here you go....

roy delgado

khornmf1.PDF

[John Warren]

I took the liberty of assembling a composite of the three amplitude responses.

[garyrc]

Thanks a million, Roy!

Thanks for your trust of our ability to handle information.

Interesting -- I think I hear the peaks more than the dips.

Are all three plots done at about 3 Meters, or are the midrange and treble ones done closer up?

How is the word "standard" being used? Are these plots the average of a sample of Khorns?

Just one more question, which may reveal my deep ignorance: what does "bodcaw" really mean? I'm assuming and I Hope, you should pardon the pun, it is the town in Arkansas, but I'm visualizing a fish, bird, food, type of music, or something more graphic.

Thanks again!

[bodcaw boy]

Thanks a million, Roy!

Thanks for your trust of our ability to handle information.

i hope i don't regret. there are some flame throwers on this forum....

Interesting -- I think I hear the peaks more than the dips.

Are all three plots done at about 3 Meters, or are the midrange and treble ones done closer up?

nope all done at three meters....

How is the word "standard" being used? Are these plots the average of a sample of Khorns?

this is one of our standards (we have severals) that we use to make acoustic calls on...

Just one more question, which may reveal my deep ignorance: what does "bodcaw" really mean? I'm assuming and I Hope, you should pardon the pun, it is the town in Arkansas, but I'm visualizing a fish, bird, food, type of music, or something more graphic.

bodcaw is the great metropolis that i live in.......

Thanks again!

roy delgado

[3dzapper]

Roy, Your charts compare favorably with this $45,000/pair speaker that got raves in "Stereophile":

Thanks.

Rick

[DrWho]

Hey Rick, what speaker might that be?

[3dzapper]

That's a Wilson Maxx2 Mike. John Atkinson said to the effect that the Maxx sounds so great despite it's measured anomalies because of it's "low distortion". Sound familiar?

The response below 300Hz is not an actual measurement. It is a sum of the many drivers and port outputs. IMO it's probably smoothed a tad. Both the reviewer and JA agreed that the bass was outstanding though. http://www.stereophile.com/floorloudspeakers/805wilson/index.html

Rick

[3dzapper]

Compare those speakers, which must be placed far out into the room, with a pair of Klipschorns tucked neatly into their corners for one sixth the cost. Imagine the electronics and playback equipment one could obtain for the "savings".

Klipschorns may be as big as a refrigerator,yes. They fill the house with food for the soul.

Rick

[JohnA]

Roy, Your charts compare favorably with this $45,000/pair speaker that got raves in "Stereophile":

Thanks.

Rick

That looks like what I heard when I got to listen to "Pipedreams".

[pauln]

Those charts are with the AK-4?

[D-MAN]

It is completely understandable that KlipschCO finds it "unrewarding" to publish FR curves. The well-known competition doesn't even bother to publish ANY specs about its products at all, claiming that any specs are virtually useless, confusing, and unimportant (more or less).

As a matter of fact, it just shows the level of discernment that Joe and/or Nancy Blow are capable of mustering when purchasing an audio product.

Getting their money away from them doesn't take specs of any sort any more. It's a WalMart world now.

Not Klipsch's fault!

DM

[garyrc]

For the '90s and the 2000s, What D man said.

 

In Antediluvian times ('60s -- '80s), no matter how many times PWK said it, advertisers, the Blows, and even some of the magazines wouldn't believe that frequency response was less important than low distortion. The public was sold on frequency response.

 

PWK published Klipschorn curves done with a swept sine wave, in an era in which other companies published noise curves (sometimes 1/3 octave averages, sometimes not). Some put a dot at the average for the 1/3 octave (or other) expanse, and connected the dots. The competition's noise curves looked deceptively honest, in that connecting the dots produced some jaggedness, but no fine jiggles and dense zig zags, compared to a swept sine wave and an old chart recorder. Most magazines (with the exception of Audio) followed suit. At least one manufacturer (JBL Professional) let the reader know what was being done (a good thing) by depicting each average as a straight horizontal line for each 1/3 octave tested, the way Stereophile occasionally does in some of their room tests. That was fairer, IMO.

 

In the Audio Papers the Klipsch company published back then, there was an article titled something like "Plain and Fancy Test signals, that concluded that the swept sine was the signal that most closely resembled musical instruments (? or was somehow the "best").

 

PWK's swept sine wave curves of the '60s and '70s looked like they were smoothed a bit for both intelligibility and to avoid suicide in a dishonest marketplace -- or maybe that's the way those pen and ink chart recorders worked -- I don't know. In any case, there were more squiggles than in the curves of other speaker companies. There was one graph at the time of the introduction of the K- 400 Horn for the midrange, with the K-77 tweeter (EV T- 35 selected by Klipsch for quality control, and matched for stereo), and there was a pair of graphs, comparing SPL loss across the spectrum when moving from 2' to 16' to demonstrate that since the loss, in a room, is much less than whatever law (falls of with the square of the distance in an anechoic environment??) would predict, therefore, all speakers use direct / reflected sound, not just those thus advertised (probably had in mind Bose, JBL Aquarius, and perhaps Empire).

 

In the late '80s, curves were reproduced on the spec sheets, with the Klipschorn being measured in the revolving door corner anechoic chamber.

 

Now, the public is overloaded with information available because of the computer revolution, but seems to know less about audio than ever, as D-man says, A Walmart World.

 

And try to find Khorns, and 3 or 4 brands of other Hi End speakers in the same store, for listening purposes! Now I find out that some manufacturers actively discourage dealers from having certain comparisons available, and a product line can be taken away if there is another dealer (bigger) to close by.

 

In the commercial area, it is no longer the World of Paul Klipsch, Saul Marantz, Rudy Bozak, or any of those guys! Thank goodness we have Roy.

[John Warren]

WIth all this talk about frequency response, I dug out (literally) my Nov. 1986 issue of Audio mag. In it, Heyser reviews the Klipschorn ("The Classic That Stands The Test Of Time"). Heyser publishes a 3m amplitude response of the Klipschorn (Fig. 4) that, I can only assume, shows a "fabricated" bass response. In fact, from 40-100Hz, it's "ruler flat" (?) and from 100 to 400Hz it is within a 2.5dB window(?). Above 1kHz the response looks "real" (+/- 5dB) but who really knows.

[mas]

I guess that I should preface this response with an admission that I should not be responding to this!<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

I tire of the seemingly never-ending need for some to continue jousting windmills in their continuing effort to rationalize and justify a speaker they claim to love. One wonders even if they were to succeed if their victory would not be Pyrrhic. It seems an endeavor more akin to the classic Napoleon complex and I fear illustrates more about the person than the speaker itself!

But, be that as it may, I am very curious about Johns comment.

There are a number of very interesting and valid claims in the review that many go to great lengths to avoid or to gloss over. One prominent issue is the time offset mentioned on page 68 that many will dismiss where the crossover is smack in the upper middle of vocals and is a classic example of the intelligibility problems involving the Haas effect. (although the even more critical range of 800 Hz- 2500 Hz for vocals is avoided)

But Johns comment regarding Heyser and his measured response alluding to its being fabricated is rather amazing to me. (Read: can we say UNDERSTATEMENT!?)

One may choose to debate Heyser in any manner you choose, but I have NEVER heard even his staunchest critics (and I will mention Lipchitz and Vanderkooy by name as being prominent among them!) make the slightest allusion to even the smallest issue of integrity!

As far as a relatively flat low end response, there are many simple and legitimate possible reasons for this, among them the fact that the response is a free air response without the effects of any superposition due to reflections as well as 1/3 octave smoothing a technique that is valid and focuses the result on measurements that correlate more closely with what we are capable of hearing. And that is not to even mention the amazing resolution of the printing that renders the response curve to be almost 1 dB thick in itself!

It seems that this rationale might be satisfactory had Heyser himself not explained the nature of the measurement on page 65 of the review. But I guess that this is unimportant.

But fabrication!? I would love to hear any possible rationale for such behavior anyway especially as Dick was very sick at the time of the review. Over the years I was lucky to witness several debates where individuals attempted to debate Dick, and I have listened to several with personal politically motivated vested interests try to denigrate him which ironically only served to subsequently validate Heysers position!), but to have someone intimate that he fabricated and hence falsified a response?...

As I said, I probably should not have responded and I doubt that no one really wants to hear my real feeling about the issue of Heyser fabricating a response! Lets just suffice it to say that I have now read a most fascinating(sic) statement on this forum that I never suspected that I would hear! It is a statement that anyone even remotely familiar with Heyser would be loath to make!

I'm done, as I have paid this 'allegation' more due than it deserves.

[John Warren]

Compare Fig 4. in the Audio issue to the JAES plot I published on the first page.

The bass response shown in the Audio magazine is not real, the amplitude response is not flat from 40-100Hz (a straight line!!) and within 2.5dB from 100-400Hz.

jw

edit-

Look at Roy's plot, he's got the bass response within an 8dB window, quite good IMO and still nothing like the Heyser plot from 1986.

[mas]

Frequency anomalies narrower than 1/3 octave are not audible. The notch at 75 Hz is narrower than 1/3 octave. 1/3 octave smoothing results in a display that is flatter as small audibly inconsequential anomalies are smoothed. With a TEF or other instrument I can make the best response look horrendous or the worst response look wonderful depending upon the parameters selected. Heyser simply chose to focus on the consequential charactristics and choose parameters accordingly! The ability to display infinately small anomalies at incredible resolutions does not necessarily aid in understanding the response of a speaker. And the failure to select a resolution capable of displaying such minute variations in this case does not constitute the "fabrication" of anything. May I suggest that there is a big difference between asking what parameters may have been selected and why, as opposed to suggesting that something is fabricated and hence falsified!