EnABL Processes

[Graham Maynard]

Hi John,

You say of the computer generated plot
>> That is, simulation of the generation and propagation of transverse waves in the cone and surround. The second part is that associated with the transfer of this vibrational energy from the cone to the air in contact with the radiating surfaces. <<

But does it cover other air-side wave excitation following the spatial representation of cone induced radiation.

The biggest problem I had relating to discussing SS NFB amplifier sound was my statements that internal stabilisation components modified reproduction subsequently to lagging/leading loudspeaker themselves reactivity modifying output (esp at class-AB crossover) quite independently (in time) with respect to original input due to amplifier damping (often in quadrature above 1kHz) failing to control output voltage before the input stage can correct output current wrt ongoing input voltage.

Does this model have boxes to insert values for the amplifier output characteristics and driver voice coil impedances which are in series with everything the voice coil drives, and also in series with every *air-side reaction upon the cone* as well as cone/air impedances as seen by the voice coil ?

If it does not cover amp and voice coil impedances and air-side reflections/wave motions then it can give no more idea of how a loudspeaker will sound than an electronic simulator can for an amplifier circuit with all the other amplifier construction induced variables.

In particular I am looking at that 3.5kHz peak. Is the simulation smoothed or lacking in resolution, or is it failing to compute unavoidable series impedance permitted modifications due to air-side wave motions, because that high 'Q' peak with ripple will be so much more noticeable in real life than the simulated peak.

This is why I suggested looking at two sine cycles in order to get an idea of the nature of the amplitude distorting component and its time or 'phase' relationship with original launch. (OK Jacques?)

I hope my considerations and questions are not seen as demands for explanations set in the same manner as which Bud has experienced in his thread, for that simulator too is obviously a developing model.

Cheers .......... Graham.

 

[sreten]

Sreten,

Actually, reading anything I have written is unimportant to the effects of EnABL. The only way to make an informed judgment is to apply the patterns, in the manner shown in the numerous photos. You can also down load patterns from various postings here on this thread to allow you to correctly apply EnABL to a specific driver. I do provide a detailed guide for those particular drivers and if you follow it, step by step you will have a working EnABL'd driver. However, even if all you were to read were the posts of other people who have been successful and obtained what they thought of as worthwhile results, you could still proceed to try this out for yourself.

Until you do, you are unqualified to make any statements, negative or positive about the effects of EnABL.

You are completely qualified to attack me however. I don't have a problem with this, as should obvious. After all dlr, who has been a member of this forum since 2005 and has 300 plus posts in that period, has seen fit to devote his last two months and 170 posts solely to attacking my thoughts and the intelligence of everyone who thinks enough of them to try out EnABL and explore it's effects in areas other than drivers. So, certainly you are entitled to jump on too..

Bud

Hmmm.....

You cannot see the insulting and condescending nature of your reply ?

/sreten.

 

[john k...]

Graham,

I think you are missing the point. It isn't necessarily about how a loud speaker will sound. The question is how does modifying the cone of a loudspeaker alter the wave motion in the cone, surround and dust cover (whizzer)...

It is not a requirement that the simulation be capable of accurately predicting the sound of the driver. All that is required is that the simulation can, with the required degree of accuracy, predict the effects of modifying the cone on the wave motion in the cone. The 3.5k peak you see is a breakup mode of the cone. If application of some treatment significantly damped the mode the result would show a much lower peak.

Any model is only as good or bad as the assumptions which go into the model. A good model is one that includes sufficient detail to allow the process to be investigated to be revealed without introducing unnecessary complexity which would only obfuscate the result. If I want to investigate a linear phenomena I would not start by developing a nonlinear model of the event.

 

[john k...]

John,

Can you explain for me what you presented in post #2851? You've used measured data from drivers to develop a model (predictive) which you've then used to simulate a driver's response, and hope to use that to simulate EnABL? And the graph shows the simulation of an actual driver (specs fed into the model) vs actual measurements? And what kind of predictive accuracies have you found when using the model and simulation on drivers that weren't used in the model development? (R2, F-value and it's significance level, etc, etc)

Thanks,
Carl

No. I did not develop the model of the simulation code. I just posted some preliminary results form a simulation tool being developed from other sources. The tool is not based on the measured data. The model is not based on a specific driver. The tool is base on a physical model of a driver including the surround, cone, dust cover and motivating mechanism. The parameters of the model were then specified to be consistent with those of the driver to be tested. That is the cone material, dimensions and shape were specified, the size, shape and material of the surround were specified, etc... The simulation tool was than executed and the predicted results compared to a measurement. Depending on the point of view and what is expected to be revealed by the simulation the results can be evaluated by how well they agree with the measurement.

As I said, this is not my tool. I posted this only to point to the direction in which things can go to truly investigate the way enable, or any other cone treatment, can alter the behavior of the cone vibration. There are lots of considerations with regard to accuracy that come into play here, but the general approach is more than sufficient to investigate such treatments.

These types of simulation tool are used throughout industry. For example in the automotive industry closely related tool are used to simulate crash worthyness. It’s a lot cheaper to perform simulations than it is to build cars and crash them until the right structure is found.

 

[Graham Maynard]

Hi John L,

I cannot get any point without sufficient explanation !

Re cone shape - I suggested inverting the dust cap - which does have a most significant effect upon real world reproduction.
If this does not show up during simulation, then what is the real world value of simulation ?

So, air side is not addressed at all, and this means it cannot have any relevence in relation to Bud's EnABL, or aperture felt etc., and thus any findings cannot relate directly to end user aperture or surface edge modifications which give rise to reproduction change or improvement.

Cheers .......... Graham.

 

[john k...]

Hi John L,

I cannot get any point without sufficient explanation !

Re cone shape - I suggested inverting the dust cap - which does have a most significant effect upon real world reproduction.
If this does not show up during simulation, then what is the real world value of simulation ?

So, air side is not addressed at all, and this means it cannot have any relevence in relation to Bud's EnABL, or aperture felt etc., and thus any findings cannot relate directly to end user aperture or surface edge modifications which give rise to reproduction change or improvement.

Cheers .......... Graham.

Again, it depends on what you are interested in. All radiated sound is the result of the vibration of the cone, right? Before the radiated sound can be evaluated it must be possible to predict the cone motion accurately, whether in a vacuum or in some atmosphere. If a treatment like Enable is suppose to alter the wave propagation in the cone then what needs to be investigated is how the cone vibrates prior to and after application. It would not necessarily be required to include the effects of the air load on the driver. If it isn't possible to predict how enable alters cone vibration in the absence of an air load then adding the air load is unlikely to improve things. However, if you want to consider how the air mass load on the cone could contribute to that result then the cone/air interface can be considered.

The direct answer to you question is, however, is yes, the air mass load on the cone and the transfer of energy from cone to air is part of this tool. As a mater of fact, as I look back at the initial document I received from the developer, the result presented is supposed to include 1) the effect of the sealed box enclosure, 2) edge diffraction of the enclosure, 3) the effect of motor Le, and 4) the transfer function for the cone breakup. With regards to enable the area of interest is item 4. Until that is exploted ther eis little reason to look further.

 

[BudP]

sreten,

Hmmm.....

You cannot see the insulting and condescending nature of your reply ?

Yes, I can see that tone in those remarks. I apologize to you for the attitude I displayed. On the other hand, I really was prompted to that attitude by these words of yours.

Seems 114 pages is a lot easier to fill with hot air than any technically
rigorous subject.

My observations :

EnABLng any solid edges to "eliminate" diffraction will not work.

EnABLing cones does something, sadly beyond the proponents understanding.

The claims are entirely fanciful, not demonstrated and not measured.

Coating a cheap 2" tweeter improves it ? really ? who'd have thought ....

And FWIW I have no interest in a pseudotechobabble "mental
model" of what is going on, it is simply arrogant and ignorant.

I must ask if you consider any of your remarks to be arrogant or insulting?

Bud

 

[JacquesToo]

 

[sreten]

I must ask if you consider any of your remarks to be arrogant or insulting?

Bud

Hi,

I'm not interested in a slanging match, but you do have a point,
it may read that way.

Published patents / articles are open to fair comment and I'll stand
by my "remarks" in principle, if not the apparent tone of them.

Anything that is not technically consistent is essentially meaningless,
and there is far, far, too much "technical inconsistency" spouted
about what EnABling does, consequently getting nowhere fast.

If what you have written about it is "unimportant" (the irony of
the actual accuracy of the statement ....), then we must beg to
differ, because from a technical perspective this is unacceptable.


/sreten.

 

[Graham Maynard]

Hi John K,

You wrote
>> If a treatment like Enable is suppose to alter the wave propagation in the cone <<

?? Has Bud ever claimed this ?? Is Bud not considering air layers/boundaries and waves ahead of the cone ??

I noted your sketch of wave progression within the material of the cone and wondered what this was about.
Now I wonder if your comments about 'established knowledge' relate to something different to what Bud is considering.

Also
>> what needs to be investigated is how the cone vibrates prior to and after application. <<

I have always understood that Bud is trying to modify air-side wave behaviour with respect to the cone, not the behaviour of the cone itself. But then of course we can all have different interpretations. Investigating and simulating cone behaviour is unlikely to reveal much in this regard.

And
>> It would not necessarily be required to include the effects of the air load on the driver. <<

I am sitting here wondering what you are thinking about then, because 1/4 and 1/2 wavelength delayed (air spring) reactions within cone to aperture space can significantly modify reproduction.

You wrote
>>However, if you want to consider how the air mass load on the cone could contribute to that result then the cone/air interface can be considered.

Yet my earlier suggestions were that the cone air interface must be considered in view of the fact that the cone is not directly controlled by an amplifier, but takes on a characteristic behaviour of its own because it is driven/damped via a complex and reactive electrical network (spring) - the voice coil in series with amplifier plus cable impedances.

And finally
>> 4) the transfer function for the cone breakup. With regards to enable the area of interest is item 4. Until that is exploted there is little reason to look further.

Who says so ?
Cone break-up is a mechanical/air loading phenomenon.
Surely the EnABL pattern has been modifying reproduction whether a cone is breaking up or not ?


Cheers .............. Graham.

 

[BudP]

sreten,

Thank you and please rest assured that I agree with your statements and point of view.

There has been some technical progress. dlr pointed out that the patterns do provide frequency response correction in specific areas, on a specific driver, from a specific set of tests. Later John K spent a fair amount of time testing a specific driver with a number of mass load profiles. EnABL patterns were among that set.

John put the CSD plots up as a roll over blink comparison, bare driver to mass loaded driver. All of the other profiles provided significant advantages in frequency response. Some dealt with the strong systemic ringing that the driver exhibited. I do think John was surprised by what he found amongst the non EnABL mass load profiles. The EnABL patterns also provided some very interesting changes and they can easily be pointed to as what would be expected by a small mass loading.

The prominent storage mechanisms were moved down in frequency a few hundred Hz, one of them was eliminated. The interesting change, to my eyes, was a general "tightening up" of all of the artifacts on the plots. As if they had been shrink wrapped and all extraneous information had been compressed back into a more condensed event.

To my eyes, this CSD plot "looked" like what I would expect, from the subjective experience of listening to the a driver before and after EnABL patterns have been applied.

I do not present any of this as proof of any more than what these two well seasoned individuals have accepted it as. Proof that EnABL is a testable event. Due to this, a number of folks are planning some very extensive objective testing, soon.

I have deliberately refrained from reporting alterations to specific areas of driver performance, to allow those who applied the process to come to their own conclusions about what they found. Just prior to these tests I will post a series of detailed findings. They will be effects that are always available, after a correct treatment has been applied.

I welcome your questions and opinions, i sincerely wish I had more to offer, in support of this odd phenomena. I think it likely that we will see some returns on testing, especially since an audio engineering mantra is that, "if you can hear it I can test for it and find it". These reports will be on the clearly audible effects, that can and should be tested for.

Bud

 

[john k...]

Hi John K,

You wrote
>> If a treatment like Enable is suppose to alter the wave propagation in the cone <<

?? Has Bud ever claimed this ?? Is Bud not considering air layers/boundaries and waves ahead of the cone ??

I noted your sketch of wave progression within the material of the cone and wondered what this was about.
Now I wonder if your comments about 'established knowledge' relate to something different to what Bud is considering.

Also
>> what needs to be investigated is how the cone vibrates prior to and after application. <<

I have always understood that Bud is trying to modify air-side wave behaviour with respect to the cone, not the behaviour of the cone itself. But then of course we can all have different interpretations. Investigating and simulating cone behaviour is unlikely to reveal much in this regard.

And
>> It would not necessarily be required to include the effects of the air load on the driver. <<

I am sitting here wondering what you are thinking about then, because 1/4 and 1/2 wavelength delayed (air spring) reactions within cone to aperture space can significantly modify reproduction.

You wrote
>>However, if you want to consider how the air mass load on the cone could contribute to that result then the cone/air interface can be considered.

Yet my earlier suggestions were that the cone air interface must be considered in view of the fact that the cone is not directly controlled by an amplifier, but takes on a characteristic behaviour of its own because it is driven/damped via a complex and reactive electrical network (spring) - the voice coil in series with amplifier plus cable impedances.

And finally
>> 4) the transfer function for the cone breakup. With regards to enable the area of interest is item 4. Until that is exploted there is little reason to look further.

Who says so ?
Cone break-up is a mechanical/air loading phenomenon.
Surely the EnABL pattern has been modifying reproduction whether a cone is breaking up or not ?


Cheers .............. Graham.

I think we are looking at this from very different points of view. If you believe that enable alters the way an acoustic wave launched from a vibrating surface propagates in space as a result of considerations other than geometrical changes in the surface (shape) or because of differences in the energy transferred to the air as a result of vibrational changes of that surface, then I can't help you. My position always has been and shall remain, until there is hard evidence to contradict it, that it starts and ends with the behavior of the radiating surface. All the hard evidence indicates this. I don't think much will come of further discussion on this. If you believe that the physics of acoustic wave propagation in air has suddenly been changed because little dashes are painted on the cone that's ok with me.

 

[Carlp]

The tool is not based on the measured data. The model is not based on a specific driver. The tool is base on a physical model of a driver including the surround, cone, dust cover and motivating mechanism. The parameters of the model were then specified to be consistent with those of the driver to be tested. That is the cone material, dimensions and shape were specified, the size, shape and material of the surround were specified, etc... The simulation tool was than executed and the predicted results compared to a measurement. Depending on the point of view and what is expected to be revealed by the simulation the results can be evaluated by how well they agree with the measurement.

John,

I asked because I have done modeling and simulation (developed models and simulations, albeit for water quality and not acoustics), have hand-cranked regression analyses, and simply wonder what model and simulation statistics have been run. If the model is intended to be used for evaluating anything, we should know it's predictive capabilities. Starting with a theoretical driver (one source of potential error), developing a model around it (another source of potential error) and entering stock driver data (a third, though not necessarily final, source of error), as it seems this model is doing, is NOT a robust development protocol, and is fraught with EXPLOSION of residuals (errors) due to squaring of each level of potential error in the model development. But maybe I misunderstood the model and simulation framework...

Carl

 

[Alex from Oz]

G'day All,


My drivers (Fostex FE167E) now have somewhere between 500 - 1000 hours on them (I stopped counting after about 300 hours).
I have not altered the cones in any way or applied anything to the cones other than the EnABL foil blocks.

Just for a sanity check, I removed all of the EnABL foil blocks from my drivers this afternoon just to see what difference it would make going back to 'bare' drivers.

Basically by comparison, without EnABL:
- they lack clarity
- detail is blurred
- they sound muffled and lifeless.
The sense of presence and immediacy has gone.
It’s like a blanket has been thrown over the speakers.

The positive effects of EnABL on the baffles and ports are still there.

HELP PLEASE!!
I need advice about cone treatments that I can apply before applying EnABL – and yes I will be using acrylic paint this time.

Anybody?

Cheers,

Alex

 

[Graham Maynard]

Hi John K,

Please don't make incorrect assumptions about what I am thinking, especially the condescention expressed here;-
>> If you believe that the physics of acoustic wave propagation in air has suddenly been changed because little dashes are painted on the cone that's ok with me. <<

There is nothing wrong with our fundamental physics, but there can be mistakes in the way hypotheses based upon it are applied.

What are the tiny foil blocks applied by Alex to a non-coated cone doing for break-up ?
Can you model this tiny additional edge distributed mass with your simulator to find a significant effect ?
I doubt that EnABLing a port affects (heavy/stiff) driver cone behaviour sufficiently to improve bass reproduction as has been reported ?

So ? Does EnABL modify air molecule behaviour in boundary regions of high surface pressure/velocity gradient which can develop dimensionally related differential products wrt either more passive regions (aperture/cabinet) or more active regions (axially towards driver or port centre) ?


Cheers .......... Graham.

 

[Alex from Oz]

I doubt that EnABLing a port affects (heavy/stiff) driver cone behaviour sufficiently to improve bass reproduction as has been reported ?

G'day Graham,

The effect of EnABL on a port is easy to test.

Sadly, so far, all people who have expressed doubts about the efficacy of EnABL in a port have lacked the courage to try it for themselves - and post their results (positive or negative).

If you have a ported speaker (subwoofer or otherwise), I encourage you to read this and try it for yourself.
If you do, please post your results.

Cheers,

Alex

 

[john k...]

Hi John K,
Please don't make incorrect assumptions about what I am thinking, especially the condescention expressed here;-

>> If you believe that the physics of acoustic wave propagation in air has suddenly been changed because little dashes are painted on the cone that's ok with me. <<

I didn't mean to come off that way.

There is nothing wrong with our fundamental physics, but there can be mistakes in the way hypotheses based upon it are applied.

Yes, and that is what is wrong with the hypotheses offered by Bud. It suggests that the energy transfer is some how altered and/or the wave propagation in air somehow obeys different laws.

What are the tiny foil blocks applied by Alex to a non-coated cone doing for break-up ?

Since I haven't seen before and after measurements how would anyone begin to know what they do qualitatively?

Can you model this tiny additional edge distributed mass with your simulator to find a significant effect ?

Again, not my simulation. But there is no reason to expect that this type of simulation would not show the effect of these alterations, if properly done.

I doubt that EnABLing a port affects (heavy/stiff) driver cone behavior sufficiently to improve bass reproduction as has been reported ?

And I doubt that however enable alters the sound radiated form a cone has anything to do how enable patches alter what happens in a port. And I doubt that it does anything to port radiation at all. It didn't in my tests. Something I did try with my HiVi system designed for ESpeakers

An externally hosted image should be here but it was not working when we last tested it.

So ? Does EnABL modify air molecule behavior in boundary regions of high surface pressure/velocity gradient which can develop dimensionally related differential products wrt either more passive regions (aperture/cabinet) or more active regions (axially towards driver or port centre) ?

Let me answer by posing another question. Do any of the following alter the same phenomena? Straight, convex, concave, or exponential sided cone shapes. Aluminum, titanium, paper, Kevlar, fiberglass, Bexterene, Aerogel, carbon fiber, bamboo, polyprop, wood fiber, NRSC coated paper, Nomex, carbob fiber impregnated paper, etc. cone materials. Rubber, foam, cloth surrounds. Concave, convex, accordion surround shapes. Serrated cones, wrinkled or folded cones, embossed cones, dimpled cones... How about the "enable like" embossments at the cone dust cover junction seen on Dynaudio and HiVi cones?

It would seem that only the magical combination of clear dope (model airplane paint) over dashes painted with nail polish have this effect?
If enable alters the way air behaves at the interface it is because the geometry of the surface and/or the vibration of the surface changes. The mechanism of energy transfer doesn't change.

Don't misunderstand Graham. I've never said there isn't a change. I'm just saying I don't accept Bud's hypotheses of how the change arises. The primary result of Enable, according to Bud, is the damping of standing wave in the cone by the pattern.

Quote from patent...

It will be seen below that each pattern of blocks or other features serves as an artificial terminus for waves propagating through the diaphragm, and that the effective size of the diaphragm, free of standing wave interference, is determined by the spacing of the patterns. It can also be seen that if only one pattern is applied to the diaphragm, its placement would determine the area essentially free of standing waves.

So, let's look at what happens in the cone, no?

 

[auplater]

model errors

John,

I asked because I have done modeling and simulation (developed models and simulations, albeit for water quality and not acoustics), have hand-cranked regression analyses, and simply wonder what model and simulation statistics have been run. If the model is intended to be used for evaluating anything, we should know it's predictive capabilities. Starting with a theoretical driver (one source of potential error), developing a model around it (another source of potential error) and entering stock driver data (a third, though not necessarily final, source of error), as it seems this model is doing, is NOT a robust development protocol, and is fraught with EXPLOSION of residuals (errors) due to squaring of each level of potential error in the model development. But maybe I misunderstood the model and simulation framework...

Carl

One could wonder whether such a definitive error analysis of Bud's model of driver behavior based on the subjective positive outcomes reported incessantly on this thread would yield anything revealing wrt predictive capabilities, eh?

John L.

 

[john k...]

John,

I asked because I have done modeling and simulation (developed models and simulations, albeit for water quality and not acoustics), have hand-cranked regression analyses, and simply wonder what model and simulation statistics have been run. If the model is intended to be used for evaluating anything, we should know it's predictive capabilities. Starting with a theoretical driver (one source of potential error), developing a model around it (another source of potential error) and entering stock driver data (a third, though not necessarily final, source of error), as it seems this model is doing, is NOT a robust development protocol, and is fraught with EXPLOSION of residuals (errors) due to squaring of each level of potential error in the model development. But maybe I misunderstood the model and simulation framework...

Carl

I agree. Any simulation tool needs to be evaluated. However, these tools applied to loudspeaker development have been around for some time. The point of posting the pics was not to suggest that this specific tool is ready to make predictions of Enable, but rather that similar tools which are ready and have been rigoriously evaluated are available. I suspect that major driver manufactires all use them to study cone dynamics. There is an entire industry which is based on the development of such tools and marketing them to the industries which apply them to their problems.

 

[Alex from Oz]

And I doubt that however enable alters the sound radiated form a cone has anything to do how enable patches alter what happens in a port. And I doubt that it does anything to port radiation at all. It didn't in my tests. Something I did try with my HiVi system designed for ESpeakers

G'day john k,

I have applied EnABL to the ports of Voight pipes, different subwoofers (BP and BR) and PA speakers.
In each case, there has been an audible difference heard not only by me, but by others with very diverse audio experience/interest.

I am interested in understanding situations where EnABLing the port does not create an audible difference just as much as those where it does.

If I understand your post correctly, you EnABL'd the port of your HiVi system with no audible difference?

Can you give me some more details about your HiVi system?
- What type of system? eg. BR, MLTL etc
- What type of port/s and what size? Flared?
- Where does the port exit the box?
- How many drivers are loading the port, and what size are the drivers?
- Did you apply EnABL inside the port? Where?
- What material did you use for the EnABL pattern?
- What block size did you use?
- How did you test?

Cheers,

Alex