EnABL - Technical discussion

Hi All.

Just want to start things off correctly here, and post my idiot thoughts.

I am and have been in agreement with John K's description of a drivers fundamental activity, as described by classical physics. Have been since I thought my way through Lincoln Walsh's patents and Beranek's short sector on diaphragms in "Acoustics"(page 199 through 201) sometime in the late 70's.

What has always given me grief, is that for the various parts of an EnABL'd driver to enforce the rather drastic sounding changes they do, even that model seemed inadequate. I mean really, how can a pattern in the first few tenths of an inch provide enough of a control over what the ongoing transverse wave emits, into the air purely by displacement, to enforce actions beyond the pattern? It makes no sense.

And then, how do the two, and now three, sets of pattern blocks accomplish this EnABL effect, upon different portions of the time train of a musical note, from a discrete instrument, like a piano. All of a piano note is certainly not on the driver surface all at once, none of it is on for very long and yet the typical corruptions, that do occur from a purely displacement event driver, are erased to a degree substantial enough to "trick" the ear / correlators/ mind, into locking onto only the corrected signal.

This seeming impossibility is why I have been so tenacious in putting forth some sort of additional effect, in my usual techno babble, head ache inducing form.

Bud
 
it's not due to boundary layer effects

My knowledge of boundary layers implies there is a mass transport phenomena occurring. That is, eg., at a solid to liquid interface, where a change is occurring, the properties by definition transition from those describing the bulk solid to those describing the bulk liquid (or gas-to-liquid, etc.). The area in between has a gradient from the one form to the other, a flux, of the material in transition, be it airflow over an airfoil, sulfur dioxide to sulfur trioxide on a lead dioxide catalyst to make sulfuric acid, iron to iron oxide on a car bumper, etc, all governed by Fick's laws of diffusion, Reynold's #'s etc.

I can't envision how a vibrating speaker cone producing an acoustic output (via trasverse waves, surface waves, planar waves etc.) involves any net mass transport across the energy interface, hence the conditions to develop a boundary layer don't exist. Furthermore, if somehow some sort of application of boundary layer kinetics were to be attempted, theory would seem to predict that any projection of significant height within the area of such application would make a simple situation more complex, not the reverse.

John L.
 
Re: it's not due to boundary layer effects

auplater said:
Furthermore, if somehow some sort of application of boundary layer kinetics were to be attempted, theory would seem to predict that any projection of significant height within the area of such application would make a simple situation more complex, not the reverse.

That is indeed what I think is the core here. That extra distortion is added by the process but that that distortion to some people is more "natural" or "melodic". That would indeed also explain the perceived higher output of treated drivers.

Very similar to tube amps vs. other amps discussions, few can argue that empirically tube amps are inferior but yet many people insist on using them for their "natural" or "warm" sound.
 
auplater,

Furthermore, if somehow some sort of application of boundary layer kinetics were to be attempted, theory would seem to predict that any projection of significant height within the area of such application would make a simple situation more complex, not the reverse.

That is certainly in agreement with what John K has been providing. I would point to the strange and unscientific wave tank results for an additional confusing element. Not, as was assumed, so much because a Soliton wave is rolling across the cone surface, as some sort of enforced circular energy transform mechanism, but as a pointer to what sort of organization the patterns do appear to enforce upon passing wave structures. This sort of activity would allow for an after the fact control of emissions, but seemingly to me, only in a form of boundary layer gradient.

Certainly does not mean either are actually in play, but some equivalent activity, that does control that pressure transfer gradient, between deep mesh structure, passing transverse wave and the adjacent air seems waranted.

I do like dlr's comments upon trapped transverse waves ringing to extinction between final pattern and surround or cone edge, but I have real difficulty in understanding how that pattern, with it's locally huge mass differential, but overall mass to mesh mass ratio, can accomplish this event. Unless there is some hidden factor in the importance of the local mass ratio and pattern structure.

You know, the secret sauce thingy.

Bud
 
Saturnus ,

I agree, it could be something of the sort. Certainly the difference in odd to even distortion ratio's found in the two different amplifier types has a large bearing upon what a person likes and dislikes.

The EnABL patterns may be doing something similar, but the only distortion tests done so far, by me, don't show that sort of help. To the contrary in fact, just as John K pointed to.

Also it does not seem to be quite as closely tied to personal experience and preferences, at least in the ad hoc A/B testing done to date. SY has indicated he is closing in on control of the tools needed to perform some objective quantifications, of drivers he is interested in. He also plans to run a double blind test with those same drivers. Perhaps even one that conforms to John l's trusted ANOVO procedures.

RonC is poking at some A/B drivers with a laser interferometry tool and RAW plans an extensive set of tests on a six piece spread of a specific driver. All to be posted and discussed here. So, data is on the way.

Bud
 
And it shall indeed be interesting to see.

What physically happens is hard to make educated guess about without. There has been many suggestions in the original thread, and many of them could be the reason. Either on their own or in a complex combination of different phenomenii.

My best guess is that it adds a form of controlled distortion while at the same time dampening random distortion. Much like dither and jitter in digital circuits where you try to reduce random distortion, jitter, as much as possible, and in many highly acclaimed circuits add controlled distortion in the form of dither.
 
Re: Re: it's not due to boundary layer effects

Saturnus said:


That is indeed what I think is the core here. That extra distortion is added by the process but that that distortion to some people is more "natural" or "melodic". That would indeed also explain the perceived higher output of treated drivers.

Very similar to tube amps vs. other amps discussions, few can argue that empirically tube amps are inferior but yet many people insist on using them for their "natural" or "warm" sound.

This hints at something that occurred to me and why I keep mentioning confounding and statistical methods. There could be codependent variables acting; if one were to postulate, for instance, a perception or expectation variable, that confounds with a distortion variable, that yields the overall response of "I like that better" or "that sucks".

One efficient way to find that out is a statistical design with known distortion components at different levels that are expected to influence the sound (maybe planet10's -40dB signal), set up with enough trials and treatment levels to be significant across the second cofounded variable of perception and expectation (find volunteers who are believers and others who are skeptics, have them listen and score, for example) then do the analysis to decide the null or alternative is correct.

John L.
 
Quite interesting how this thread keeps expanding.


HAHAHAHA! So very true.
Lets see i have been critized for :
1. Being human and saying i am sorry.
2. My evaluation methods were incorrect.
3. Stating the obvious.
4. Declining to be involved in technobable.
5. Actually investigating realistic answers.

If this was an industry question:
1. Does it actually improve performance.
2. How much does it cost.
3. Does the cost/performance ratio justify the added expense.
4. Can it be implemented with controls and what is the additional cost.
5. Will it sell, and can the customer be satisfied with the results and can it be delivered on time.

do a fishbone analysis.
6. Will we profit from this.

ron
 
Properly applied patterns will improve performance. And quite significantly. The catch is if there are other problems, they may stand out and become annoying if not solved altogether.

If you want low recurring cost, then there is an up front investment. Otherwise the recurring cost is higher. I think companies can profit if the quality is stable enough. The key is finding the right balance between recurring and non-recurring cost.
 
The greatest question i see is if there is a consistancy in the application of the medium. A felt pen or other(manually applied) is not an adequate answer.

My initial tests indicate that if there is a change then it has to be a consistant change, I keep seeing a variable. The effect has to be over the entire surface of the cone and equal over any degree change.

ron

I think companies can profit if the quality is stable enough. The key is finding the right balance between recurring and non-recurring cost.


Thats called SPC.
 
The greatest question i see is if there is a consistancy in the application of the medium. A felt pen or other(manually applied) is not an adequate answer.

Luckily we poor pen twiddler's, and our legions of devoted fans and groupies, don't even have chart recorders in our feeble brains, so we have no reference except the lies we remember telling.

But I have little doubt that you will find even greater clarity through optimization of the process. What in the world you would do with something that scarily realistic sounding, escapes me just now, but then, I am one of the feeble minded ones.

Bud
 
ronc said:
The greatest question i see is if there is a consistancy in the application of the medium. A felt pen or other(manually applied) is not an adequate answer.

My initial tests indicate that if there is a change then it has to be a consistant change, I keep seeing a variable. The effect has to be over the entire surface of the cone and equal over any degree change.

ron

I think companies can profit if the quality is stable enough. The key is finding the right balance between recurring and non-recurring cost.


Thats called SPC.
Just from experiments, 1mm location difference was when difference becomes measurably significant.

What's SPC?
 
As I see things, the fastest and most profitable way for a company to introduce EnABL lies in application to speaker ports.

This would involve some re-tooling to get the EnABL pattern into the ports:
- either by incorporating it into the injection mould, or
- appliying it using some other material.

- It is easier to maintain consistent quality in manufacture of ports (eg. no moving parts, no assembly of discrete components).
- There is no visible change to the speaker.
- The drivers will not have to be re-engineered or undergo additional processing.
- The existing speaker designs will not have to be modified in any way.
- The EnABL ports can be easily introduced across the full range of ported speakers from sub woofers to multi-way.
- The audible improvement is significant enough to justify a premium price over an identical model without EnABL'd ports = more profit for minimal extra cost.

A very powerful demonstration to a prospective company would be to give them a blind listening test using four identical ported speakers they currently sell:
- Two will have EnABL in the port
- the other two will remain unchanged.

Cheers,

Alex
 
Alex from Oz said:
As I see things, the fastest and most profitable way for a company to introduce EnABL lies in application to speaker ports.

This would involve some re-tooling to get the EnABL pattern into the ports:
- either by incorporating it into the injection mould, or
- appliying it using some other material.

- It is easier to maintain consistent quality in manufacture of ports (eg. no moving parts, no assembly of discrete components).
- There is no visible change to the speaker.
- The drivers will not have to be re-engineered or undergo additional processing.
- The existing speaker designs will not have to be modified in any way.
- The EnABL ports can be easily introduced across the full range of ported speakers from sub woofers to multi-way.
- The audible improvement is significant enough to justify a premium price over an identical model without EnABL'd ports = more profit for minimal extra cost.

A very powerful demonstration to a prospective company would be to give them a blind listening test using four identical ported speakers they currently sell:
- Two will have EnABL in the port
- the other two will remain unchanged.

Cheers,

Alex

Putting the cart b4 the horse here, technically speaking... as there has been no confirmed valid analysis as to whether this even works...

maybe you can convince Floyd Toole though...;)

John L.
 
I don't know if this is the correct place to pose a question of this nature but if it should be somewhere else, some mod, please put it there. (disclaimer over)

I have been reading the EnABL threads for months and have tried to put my own thoughts into the process. There have been lots and lots and lots of testimony and argument over how it does work and how well it works. I worked in the automotive audio industry for a long time and I remember several speaker manufacturers building drivers from cheap plastic materials that included stamped shapes int he design. In fact, I am trying to recall if it wasn't pioneer that had a set of speakers that had what looked liked diamond plate right around the edges that would have been a very consistent way of putting "EnABL" on a speaker cone. Unfortunately, IIRC, these were just very cheap speakers that wouldn't have been anywhere close to comparable with the drivers most people have applied the process to.

In short order, the question I have is this. Could someone suggest a repeatable (controllable) way to apply the process to different types of cones? I see a lot of Fostex banana cone drivers done and I can think of no legitimate way to control the process here as any paint, ink, or other material will leech into the material at an undetermined rate. I would think this would cause inconsistencies similar to the differences in basic T/S paramaters. I could see it working very well in a metal or plastic cone driver, but who wants that material in a full range speaker? Or even a Hi-Fi speaker of any kind? (although I have heard some that sounded very good). Is it possible to have some sort of shaping to a paper cone? Or a wool cone? Or ...........

I have been in the business of business management and sales for a long time so I do agree with the above statements about profit. I am thankful that things such as EnABL are part of the DIY community and can be a part of my hobby where profit isn't a concern personally. I wouldn't mind helping someone to produce a driver that was consistently better than others though and I am good at such things as the financial analysis and marketing aspects of such a venture.

The objective part of this discussion I think Cal started the thread for would be related to measurable results and reliability of feedback on the process. Subjective opinions aside, I was just wondering if anyone had thought of a really consistent way to apply the process so that say 100 drivers could be modified and tested and the results controlled. Perhaps a robot or some sort of stationary stand that would hold the driver and the application apparatus? I could see making something like this. Although, from descriptions, the calligraphy pens et al aren't as consistent as we would like.

Just poking my nose in and stirring the pot, lol.

What have you guys seen or worked with that might accomodate that?

Take care,
Robert