When I first started to look at this thread, the first thing that popped into my head was, "is this a way to try and get SL to join the diyAudio community"...
If it was then perhaps a slightly softer tone would have been better
especially considering what kevinahcc20 just posted (specifically the quote from SL saying he hoped it sparked interest that lead to more research Tony.
Well - I'm not exactly after a "gentlemen agreement" - as you already might have realized
My argumentation - and reason to put up this thread - are actually two fold.
First - I address the mere technical / physical topic regarding "energy storage" with respect to speakers - and - pin point where many people (SL included) have difficulties to separate "their view" from simple laws of physics.
(This I was not aware of being a severe problem before the "Babylonian discussion" in the ZDL thread).
Second - I address the impact of the superficial wording/ view with respect to the topic - in other words the fogging resulting from what I see as SL's sales speak.
(in a more philosophic light its actually what I call "the American self selling habitus")
in short
Tell me : what's actually the reason to bring up / suggest a rather complex method of measurement *if* all to reveal is well known filter patterns ???
To avoid such patterns anybody else is looking at a simple FR - no?
If FR is rugged in the frequency band of interest - well that tells you all - no ??
Michael
PS
I didn't "study" SL's page - just picked the plots that most clearly show what SL was after - but I think I have read somewhere that he "invented" that specific measurement to find a midband speaker for the Orion?
PPS
I certainly do not agree that the difference between my point of view and the general one (SL's included) are "mainly semantic" - this would be pretty pointless for me
PPPS
"is this a way to try and get SL to join the diyAudio community"...
SL already is present at DIYAUDIO now and then, no need for an extra call....
My argumentation - and reason to put up this thread - are actually two fold.
First - I address the mere technical / physical topic regarding "energy storage" with respect to speakers - and - pin point where many people (SL included) have difficulties to separate "their view" from simple laws of physics.
(This I was not aware of being a severe problem before the "Babylonian discussion" in the ZDL thread).
Second - I address the impact of the superficial wording/ view with respect to the topic - in other words the fogging resulting from what I see as SL's sales speak.
(in a more philosophic light its actually what I call "the American self selling habitus")
in short
Tell me : what's actually the reason to bring up / suggest a rather complex method of measurement *if* all to reveal is well known filter patterns ???
To avoid such patterns anybody else is looking at a simple FR - no?
If FR is rugged in the frequency band of interest - well that tells you all - no ??
Michael
PS
I didn't "study" SL's page - just picked the plots that most clearly show what SL was after - but I think I have read somewhere that he "invented" that specific measurement to find a midband speaker for the Orion?
PPS
I certainly do not agree that the difference between my point of view and the general one (SL's included) are "mainly semantic" - this would be pretty pointless for me
PPPS
"is this a way to try and get SL to join the diyAudio community"...
SL already is present at DIYAUDIO now and then, no need for an extra call....
Last edited:
Since Dr. Linkwitz is German by birth and upbringing (educated at Darmstadt), this personal shot is more than a little ironic.
If you can't be bothered to contact him and nail down just exactly what your disagreement is (or even if there is one), especially considering that he and you can communicate in German if preferred, it seems pointless to argue here to a blank wall. Just my opinion.
Well the "blank wall" is all the people that *might* pick up what I was outlining *and* are able or willing to look at what I'm actually saying here.
I'm not after convincing SL - I'm after bringing back simple physical laws into some audio guys head and strip off the "magical appearance" of SL's "stored energy" term
- should be no big deal - I mean - the terms "energy" and "storage" are technically / physically well defined (except in audio speak as it seems)
Michael
Last edited:
Well the "blank wall" is all the people that *might* pick up what I was outlining *and* are able or willing to look at what I'm actually saying here.
I'm not after convincing SL - I'm after bringing back simple physical laws into some audio guys head and strip off the "magical appearance" of SL's "stored energy" term
- should be no big deal - I mean - the terms "energy" and "storage" are technically / physically well defined (except in audio speak as it seems)
Michael
OK but since you didn't bother to review SL's material perhaps you would learn (as I gleaned) that your differences are small. Might even prompt you to read his stuff before you explained to him which parts were wrong.
The good thing about laws of physics is that they are widely accepted - so - usually no need for "personal" interpretation (except in "audio speak" as it seems).
with respect to the technical / physical meaning (not the magic meaning in audio speak) of "stored energy" this is simply all energy storage mechanisms involved
- as outlined - it also includes the storage in mass' movement and spring's displacement for example.
To store energy hence is the most natural (and in no way ill !) behaviour of any speaker on world.
This is mere physics.
- on the other hand - to tag some "ill behaviour" to the magical coining of "stored energy" (as SL obviously did and many believe in) - more philosophically seen - is just "creating evil to establish holiness"
So - there is really no need to select speakers after low "stored energy" (as SL obviously did and many believe in)!
Michael
Last edited:
Well I suppose it would only matter if mige0 we're actually barking way up the wrong tree on this one. It's like the old FR vs cumulative spectral decay thread we had some time ago.
I have also seen other reputable people talking about energy storage and always wondered what exactly they were talking about. Surely this would be represented in a CSD plot and then in turn in the FR graph (or the other way around if you like). Zaph showed reasonably clearly that if you take two quite different drivers, then hammer their end acoustic responses to the same target, that the CSDs look pretty much identical.
I personally find the way in which data is represented in a CSD a worthwhile thing to look at, is it strictly telling me anything else then what the FR does? No, but often it's easier to notice certain things when data is presented to you in a different way. Maybe this is why SL did what he did, because it fitted in best with his own internal picture of how things work.
If anything this thread has helped to clear up the issue of what energy storage actually is, because I had been confused by its mention and had always thought, well surely that should show up in a FR or CSD plot? It is then surely not something you need a special measurement technique to show, nor is it a seemingly 'invisible' parameter that you more have to guess at then anything else.
From mige0's last couple of posts it seems that this kind of realisation was his original intent and as I do not posses a background education in engineering, I am thankful.
I do however think that the title
of the thread was perhaps poorly chosen. It appears that this was done to grab peoples attention, and that it surely did, but I do think that this was perhaps a little misguided. It almost seems like you're issuing a personal attack and maybe even accusing Linkwitz of trying to confuse people (yes it did confuse me, but that's due to my own ignorance, rather then anything Linkwitz did). You'd have done far better if you had presented your results purely as a way of discussing/understanding the measurement and data presentation technique that Linkwitz uses.
I have also seen other reputable people talking about energy storage and always wondered what exactly they were talking about. Surely this would be represented in a CSD plot and then in turn in the FR graph (or the other way around if you like). Zaph showed reasonably clearly that if you take two quite different drivers, then hammer their end acoustic responses to the same target, that the CSDs look pretty much identical.
I personally find the way in which data is represented in a CSD a worthwhile thing to look at, is it strictly telling me anything else then what the FR does? No, but often it's easier to notice certain things when data is presented to you in a different way. Maybe this is why SL did what he did, because it fitted in best with his own internal picture of how things work.
If anything this thread has helped to clear up the issue of what energy storage actually is, because I had been confused by its mention and had always thought, well surely that should show up in a FR or CSD plot? It is then surely not something you need a special measurement technique to show, nor is it a seemingly 'invisible' parameter that you more have to guess at then anything else.
From mige0's last couple of posts it seems that this kind of realisation was his original intent and as I do not posses a background education in engineering, I am thankful.
I do however think that the title
of the thread was perhaps poorly chosen. It appears that this was done to grab peoples attention, and that it surely did, but I do think that this was perhaps a little misguided. It almost seems like you're issuing a personal attack and maybe even accusing Linkwitz of trying to confuse people (yes it did confuse me, but that's due to my own ignorance, rather then anything Linkwitz did). You'd have done far better if you had presented your results purely as a way of discussing/understanding the measurement and data presentation technique that Linkwitz uses.
I do have to ask of this though. Why does the driver having operated over 100x cycles or just 1 cycle make any difference?
If the excitation signal applied, happens to contain 1 cycle of the test tone, or 10, surely the end state in which the cone/voice coil/etc is left in, after the signal is turned off, is exactly the same(ignoring compression)? Or are you saying that the resonances within the driver take 'time' to get going?
It's not that the number of cycles by itself is particularly relevant...rather it's a different context in which to consider the energy dynamics. Over each cycle energy is exchanged between the mass and suspension and Michael is indeed correct in his assertion that this aspect of energy storage is fundamental to and not an anomaly in driver dynamics. The longer term energy storge domain of SL's concern is the context where the driver is excited by a relatively brief shaped impulse and responds with a much longer decaying series as a function of resonance. So the prolonged tail represents the delayed return of energy stored.
On your previous post I think SL would agree with your position as he noted in the discussion of the energy storage issue, "The impulse, ETC, frequency, CSD and tone burst response give different representations of the same reality. Some are easier to interpret than others. Some are more related to certain aspects of linear distortion. There may be preferences depending upon one's experience with relating a visual feature to an auditory sensation."
Yes and no
The title is IMO just exactly right - as - like shown - the monumental misunderstanding among audio guys regarding "stored energy" has its roots in SL "ignorance" - actually exactly the same "ignorance" you were facing.
Its just that usually no one would expect an audio icon to be in such massive error nor didn't give a second thought about what he's putting up.
Actually its the same "limited understanding of big guys" situation we had (and still have) as with the "least diffraction waveguide" bogus.
Someone who is after low diffraction (LOL - or even after zero diffraction / ZDL - as is the stated design target in a current thread ) in speaker design only documents most clearly that this person does not know anything about the simple nature of "diffraction" - as anyone who is after "low stored energy" most clearly does not know enough about the simple nature of "energy storage".
Well – sometimes even the most obvious has to be stated !
No big deal.
Hence I sad my intent in this thread is two fold.
Michael
Last edited:
Hi Michael,
IMHO what you have shown is not a "Debunking SL's „Stored Energy“" rather your own misunderstanding on the subject.
The Linkwitz JAES article, JAES papers, linked from his webpage starts with the paragraph:
" The test technique described here developed out of the need for evaluating and characterizing loudspeaker drivers..."
The use of windowed sine to test loudspeaker is known since long time and I never seen SL claim he invented it. It has been used by several others. Since I had the JAES loudspeaker Anthology 1 nearby, a random chosen example is:
The Development of a Sandwich-Construction Loudspeaker System, DA Barlow. pp 159-171. Here the use is the same a by SL.
That a filter change the measurement is discussed in the JAES article too in explaining why it's a good idea to use a shaped tone burst. Nothing new here. Since resonances are a kind of filter it's what we are looking for.
Then in section 3 SL states whats he uses this test for:
Loudspeaker drivers often show high Q resonances involving cone and surround interaction. These are easily identified by the envelope distortion of the shaped burst response....localized high Q resonances are difficult to pinpoint in swept frequency magnitude response curves and often require careful analysis of the associated phase response."
I couldn't copy and paste from the pdf file so I might have made some typing error...please read the linked aticle itself.
As an example of intended use of this test look at fig 7.
/örjan
IMHO what you have shown is not a "Debunking SL's „Stored Energy“" rather your own misunderstanding on the subject.
The Linkwitz JAES article, JAES papers, linked from his webpage starts with the paragraph:
" The test technique described here developed out of the need for evaluating and characterizing loudspeaker drivers..."
The use of windowed sine to test loudspeaker is known since long time and I never seen SL claim he invented it. It has been used by several others. Since I had the JAES loudspeaker Anthology 1 nearby, a random chosen example is:
The Development of a Sandwich-Construction Loudspeaker System, DA Barlow. pp 159-171. Here the use is the same a by SL.
That a filter change the measurement is discussed in the JAES article too in explaining why it's a good idea to use a shaped tone burst. Nothing new here. Since resonances are a kind of filter it's what we are looking for.
Then in section 3 SL states whats he uses this test for:
Loudspeaker drivers often show high Q resonances involving cone and surround interaction. These are easily identified by the envelope distortion of the shaped burst response....localized high Q resonances are difficult to pinpoint in swept frequency magnitude response curves and often require careful analysis of the associated phase response."
I couldn't copy and paste from the pdf file so I might have made some typing error...please read the linked aticle itself.
As an example of intended use of this test look at fig 7.
/örjan
You got me !
I confess that I have not read what was written in AES Bible book # xyz in chapter # yxz on sentence # xyz.
Mea culpa, mea maxima culpa...
On the other hand you possibly should not lock too much into that measurement „inventing“ thing – I'm certainly no profund audio measurement historian – but, to my knowledge TEF was already „invented“ and the measurement method SL tried to popularize is less then a quantum leap form that IMO.
So – it certainly does not come as a surprise for me that there is/ was someone befor – nothing new under the sun ...
Just keep fokused on what I was saying with respect to SL's and common lack of „understanding“
the issue of energy storage in loudspeakers - plus - how he (and many) interpertate the plots.
The myth around „stored energy“ in audio speak is better to be stripped off and the term brought back to its technical / physical meaning.
That will allow us to realize that „stored energy“ in loudspeakers is the most natural thing on earth and no one actually has to seek after avoiding „ill stored energy“ (not being possible anyway)
The good news : to trash that myth will free you from just another „mind blocker“.
Michael
Possibly I should also address that I think there is much too much focusing on "the tail" in audio speak and with those fancy "time domain analysis" invented in the past.
*If* one wants to get a feeling about time domain behavior it may be best to look at the *total* outcome of a sine bust.
Here we go:
Below is a 10ms bust of a 1kHz sine on a ruler flat FR :
below shows the behaviour of a boost of 6dB at 1000Hz with a Q of 2 :
below shows the behaviour of a boost of 6dB at 1000Hz with a Q of 10 :
below shows the behaviour of a boost of 12dB at 1000Hz with a Q of 10 :
To make comparison easier I added the appropriate plots shown at the beginning of the thread
Michael
Last edited:
Basically the shape would be as is seen in the "SL styls" plots. - just imagine the rectifying not happening (in blue trace below)
Rest of the SL circuit is only a log amp that makes the log-decay in the time domain plot look like a straight-decay in the SL style plot (after the red marker in both plots):
The SL stimulus is - more or less - kind a "soft fade in > soft fade out" sine burst - just have a look at the stimulus pix of ARTE or SPICE (green trace in first pix).
Michael
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.