|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
206.255.200.146
In Reply to: RE: JA has commented previously, posted by A.Wayne on June 19, 2015 at 12:16:05
John can drag them outside and measure, i even offered to help ..:)
But would an open air test be indicative of a dipole's true response? I remember Dr. West once saying "we don't want to get rid of the rear wave".
Follow Ups:
Well, inside or outside, some of JA's measurements are easier to interpret with forward radiating speakers than with dipoles. Those who just look at the results without reading the text may well be shocked by the big bass rise in the on axis curves. But measurements need interpretation, and the rules are a different for dipoles because of the partial cancellation in the bass because the front and back waves are of opposite polarity. One person who reviews equipment actually used this as an example of measuring poorly but sounding great. But of course, in room, because of dipole cancellation effects, that big hump disappears.
I have linked to the measurements section of a very fine dipole speaker, the Quad 2805, to illustrate this. One can see a big hump in the bass in Fig. 2 which disappears in the room response in Fig. 6. For a casual reader, who does not understand that dipoles react somewhat differently in a room than forward radiating monkey coffins (which is what I have!) JA always explains this.
There are ways around this, but the one I know involves adjusting the data, and I tend to agree with JA to present the actual data, not that my nonexpert opinion means much. But I can also see why Wendell Diller might not agree. I might also point out that JA shows different displays for the horizontal dispersion. In Fig. 3, he shows the actual data, and in Fig. 4, he shows the difference between the on axis and off axis responses.
-----
"A fool and his money are soon parted." --- Thomas Tusser
Then why no counter measurements from magnapan, can i request imp/mag-phase and FR graphs from magnapan....?
look at how vague :Magneplanar 3.7
Freq. Resp. 35Hz- 40 kHz
Rec Power Read Frequently Asked Questions
Sensitivity 86dB/500Hz /2.83v
Impedance 4 Ohm
Dimensions 24 x 71 x 1.625 inches
really ..!!!!
And so it goes ..............
Edits: 06/20/15
I don't think many manufacturers provide much in the way of measurements of consumer loudspeakers, so I don't think Magnepan is unusual in this regard.
-----
"A fool and his money are soon parted." --- Thomas Tusser
> Then why no counter measurements from magnepan, can i request
> imp/mag-phase and FR graphs from magnepan....?Magnepan's Wendell Diller has stated that he doesn't want the company's
speakers measured because they will will reveal proprietary information
regarding the crossovers. This may or not be correct. However, in my
conversations with Wendell, it emerges that he does not like the waterfall
plots that I publish, feeling that they paint Magnepans in a bad light.With panel speakers, cumulative spectral-decay waterfall plots do tend
to look hashy. This may be due to the multiple arrivals at the microphone,
due to the physically large radiator. Or, as I have conjectured in the
magazine, it might indicate Chaotic diaphragm behavior, ie, while the
average position of the diaphragm responds pistonically to the driving
force, individual elements vary in their response, "shimmering," as it
were.I don't currently know how to best investigate either conjecture.
John Atkinson
Editor, Stereophile
Edits: 06/21/15 06/22/15
Like anyone with an LCR meter can't take apart the xo, trace the circuit, and measure the individual components?
John, you've measured a number of dipolar planar speakers over the years, including electrostats, magnetic planar, and "true" ribbons. Have you found any "typical" measured behaviors that characterize these and distinguish them from monopolar box speakers?
> Have you found any "typical" measured behaviors that characterize these
> and distinguish them from monopolar box speakers?
The dipolar radiation pattern of course, and the fact that other than in
larger rooms, not just the measuring but also the listening takes place
with more of a nearfield effect than with a physically small speaker. And
the hashy-looking waterfall plots. These are all common factors with large
panel speakers.
On the plus size, there are no cabinet resonances. However, depending on
the skill and insight of the designers, there will be some effect from the
stretched diaphragm's "drumskin" resonance.
John Atkinson
Editor, Stereophile
Any rectangular membrane (such as those in Maggies and other panel speakers) is going to produce modes that are even less pretty than the classical drumhead resonance, due to the asymmetry of the diaphragm. Circular membranes produce fairly predictable reflections/resonances; rectangular membranes, less so, with some ugly crush-points at corners.
I can imagine that it wouldn't look pretty under most test procedures.
If I were making a measurement for the purpose publishing in a review and I didn't know what the measurement I was taking actually represented, I might be reluctant to publish it until I did.
Or at minimum, attempt to correlate the measurement with something I heard.
But WTFDIK about publishing?
I would agree with your final conjecture , chaotic behavior of the diaphragm......
> I would agree with your final conjecture, chaotic behavior of the > diaphragm...
In a lecture I attended in the early 1990s, the mathematician Manfred
Schroeder (who passed away a few years back) mentioned that one signature
of a chaotic audio system is that it generates subharmonics.
I have investigated this with a Magnepan speaker. Yes, it does indeed
generate a strong subharmonic at exactly half the signal frequency, ie,
feed the panel a 1kHz signal and as well as the usual harmonic distortion
harmonics at 2kHz and 3kHz, you see a tone appear at 500Hz. But this is
not really conclusive evidence for Chaotic behavior.
John Atkinson
Editor, Stereophile
| But this is not really conclusive evidence for Chaotic behavior.If it were a system running in a truly chaotic regime, then a periodic input would create a non-periodic chaotic "noise" out. It would be really obvious acoustically: think about a self-feedback oscillation 'squeal' but which then turns into a horrible broadband seeming noise.
A nonlinear system can be chaotic with some parameters and non-chaotic with others.
Subharmonic generation requires a nonlinearity, and chaos requires a nonlinearity as well, but they aren't necessarily the same.
What your observation shows is that the mechanisms of nonlinearity in a panel speaker driver may be different (well it's obvious) from those of a conventional driver and amplifiers which tend to make typical higher-frequency harmonics instead of subharmonics.
My guess is that there is resonance/vibration in the panel's supporting structure which naturally doesn't vibrate very well at 1kHz , but with some weak nonlinear coupling can 'keep up' with vibration and resonate and couple to energy every other cycle.
Subharmonics often happen with nonlinearly coupled resonant systems, superharmonics often happen with nonlinearities/static input-output functions which are worse when amplitude is higher {extremes of signals}, e.g. the magnetic field isn't fully homogeneous at the limits of excursion or the transistor's transfer function isn't fully linear.
Edits: 06/22/15
Wow, you certainly said a 'mouthful' there!
'I don't currently know how to best investigate either conjecture.'
Since no one can really confirm either hypothesis it would be premature to jump to any conclusions.
from the review referenced in this post .
"...measuring physically large speakers with in-room quasi-anechoic techniques is in some ways a fruitless task."
Why bother if the information is known by all to be problematic and not representative in any sense of real world performance?
To provide overtly misleading and incorrect data as to how the speaker actually sounds in the real world?
What is incorrect about the data? The results of the measurements are what they are.
What is problematic about the information?
What is misleading about the information?
What on earth is real world performance as opposed to unreal world performance? I am not aware that measurements are taken in some unreal world.
As I pointed out, measurements need to be interpreted as to their significance.
-----
"A fool and his money are soon parted." --- Thomas Tusser
The difference will be especially pronounced with the wonky dispersion pattern of a large dipole, and the consequent reflections and cancellations. Measurements made IN a real room, in real time, from the listening position, will probably tell you more about its actual behavior than QA ever could. And of course this behavior will be different in EVERY room depending on dimensions, treatment, placement, etc.
What is incorrect about the data?
I'll let JA explain it to you:
"The usual assumption, that the measuring microphone is very much farther away than the largest dimension of the speaker being measured, is clearly wrong."
What on earth is real world performance as opposed to unreal world performance?
Do you really need someone to explain to you the differences in the way the measurements are taken and the way people actually listen to the speakers? It would seem so.
What? Are you trying to accuse the measuring instruments of dishonesty? Sorry, but the measured data is what it is. JA's remarks refer to how the measured data is to be interpreted.
Now, as to conditions of measurements, the ideal is to determine what the speaker itself does. Then, one can try to find out what tends to work best in rooms like the rooms people are likely to put speakers in, which is one of the things Dr. Floyd Toole did.
There is nothing to prevent you from measuring the response of your speakers in your listening room---oh yeah, you actually do that.
-----
"A fool and his money are soon parted." --- Thomas Tusser
Ever mindful of David Bentley Hart's wise counsel that absolute subjective liberty is known only in hell, I'm curious as to what you would do if you were shopping for a speaker, found one that impressed you no end, was affordable, but measured poorly. Would you dismiss it out of hand due to the poor specs or make the purchase?
A Poorly measuring speaker would sound bad........
"A Poorly measuring speaker would sound bad........"
Yeah, just because E-stat comes up with some contrary to fact hypothetical situation doesn't mean it's of some practical value.
-----
"A fool and his money are soon parted." --- Thomas Tusser
Not following you .....
A lot of self impregnated fallacy's floating around .....
Regards
Well, this is a hypothetical scenario E-stat proposed:
"I'm curious as to what you would do if you were shopping for a speaker, found one that impressed you no end, was affordable, but measured poorly."
-----
"A fool and his money are soon parted." --- Thomas Tusser
I answered , It would not impress ,well only in the short run and on special rcordings . You will always be aware of it 's short comings,...John referenced the 3.6 waterfall plot , ironically you can hear the smearing of details when listening to it , so I was not surprised when I saw the waterfall plot . It would have been interesting to compare it to a 3.7...
Regards
Edits: 06/24/15
Try again. Who was it who asked that question?
There is nothing to prevent you from measuring the response of your speakers in your listening room---oh yeah, you actually do that.
At the listening position devoid of employing any overtly wrong assumptions. The bonus is that it is relevant data. :)
I tell people in my lab that if they aren't going to believe the results of an experiment (or be able to convince me that the experiment might be convincing), there is no point in doing it. Similarly, if you feel a measurement is not going to be interpretable, one really shouldn't do it.
But JA's measurements of panel speakers can be interpreted, including those in the bass, and JA does so in the text.
-----
"A fool and his money are soon parted." --- Thomas Tusser
The poor measurement results led to a much improved 3.7 ......
!
The Mind has No Firewall~ U.S. Army War College.
He may be measuring 'real world performance', but who cares how it measures if sounds good even if it doesn't measure the same as a 'monkey coffin'?
As von Recklinghausen said:
"If it measures good and sounds bad, -- it is bad. If it sounds good and measures bad, -- you've measured the wrong thing."
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: