EnABL Processes

[ronc]

but am happy to oblige in future if you do so.


Thank you.

ron

 

[dlr]

When you try to reduce standing wave of a specific frequency like the strong cone breakup modes, you have to design patterns specifically for that mode. Based on my experience is that this is the last resort in treating the driver. Generally if we use the patterns so that the energy in the overall spectrum can be dissipated more quickly, we can accomplish cleaner sound. Once you get cleaner sound, defficiencies at specific points will become dominating factors effecting the quality of the reproduction. If these specific points are not there in the first place, then you really have a good driver.

From that perspective I completely agree.

Dave

 

[soongsc]

john k,

The abrupt nature of your response is perplexing.

I simply asked for clarification on whether you applied EnABL inside the ports.
At this stage there is no ideal or correct method for doing this.

If you did put EnABL inside the ports and there was no audible difference, that's fine.
But, why are you refusing to answer any of my questions?

I think the magnitude of effects will depend on the quality of the port design itself. If there was sifnificant port noise to begin with, then EnABL might help reduce it. If the port was designed originally with port noise reduction in mind, then improvement might not be audible. Since John is quite experienced in enclosure design, I would assume his designs might already have port noise and damping issues optimized such that EnABLing might not be able to improve it further.

 

[soongsc]

From that perspective I completely agree.

Dave

That was a surpise.

 

[BudP]

Alex and John K

My one treated port test, with paint, did not produce any particular effect either. I do still have the speakers, some very small RS metal box two way systems. My memory is of a "four inch" woofer and a port length of about 1 1/2 inches in length. I will dig them up and give a re-listen to see if I hear any difference, and confirm all of those mis remembered dimensions.

Bud.

 

[Alex from Oz]

G'day Bud,

I suspect that exploring the EnABL pattern applied to ports and baffles will eventually yield enormous returns.
Let me know what the diameter is and I'll propose a block size to start with.
Maybe time to break out the duct tape?


soongsc,

There are too many unknowns when it comes to the EnABL pattern on ports and baffles to make any conclusions.
This is something that anyone with a ported speaker design can get involved in - should they desire to participate.

Cheers,

Alex

 

[ttruman]

Initially the passages had smooth walls. In later measurements the passage was “enabled”. That is the walls were ribbed.

Interesting. By chance did the outcome of that test reveal ribbed walls reduced turbulance?

 

[Graham Maynard]

Aha.
After comments on page 117 it appears to me as if the odd writer here has taken umbridge at Bud's Patent, and then decided it was their responsibility to take an authoritarian stance (on who's behalf ?), even though all necessary 'hands-on' information has still not been gathered !

Also:
Surely when Patents are drawn up, the intention is to cover all imaginable future applications whether apparent or proven at the time of application or not ?
________________________________________________

Hi John K,

Re the first of your two sine illustrations leading to your *personally* aimed comments of obfuscation and technobabble.

A transient cannot be represented by a burst of single frequency sine wave suddenly being turned on and off.

Yes a transient may be deconstructed into a series of concidentally *starting* sines, but where an overall response becomes reactively modified at any frequency(ies) and the transient itself becomes (audibly) distorted, the wave distortion does not arise at the same frequencies as those which can become established as phase shifted steady state standing waves due to all of the other components still being present in real time !
_______________________________________________

Hi RonC,

Looks like some interesting testing going on there.

Appears to be subtractive. Is your X-Y display a diagonal line with distortion/noise appearing as amplitude/time related blips along that line, or building up to 3D ?
________________________________________________

?? Maybe Bud's surface coating tightens a cone's surface layer and reduces transduction delay at higher frequency, though could introduce ringing if overdone ??


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

 

[soongsc]

G'day Bud,

I suspect that exploring the EnABL pattern applied to ports and baffles will eventually yield enormous returns.
Let me know what the diameter is and I'll propose a block size to start with.
Maybe time to break out the duct tape?


soongsc,

There are too many unknowns when it comes to the EnABL pattern on ports and baffles to make any conclusions.
This is something that anyone with a ported speaker design can get involved in - should they desire to participate.

Cheers,

Alex

Alex,

Sometimes it just seems funny why experienced people take time to devalue improvement methods presented here, while similar aspects in commercial designs like B&W go unchallenged. It seems like an old Chinese saying "Know whom the master of a dog is before you annoy it".

 

[Alex from Oz]

soongsc,

I agree with you.

"Those who have free seats at a play hiss first." (Chinese proverb)

Cheers,

Alex

 

[john k...]

We are using a monochromatic light source and a reciever that both are collimated and focused at a given point. The frequency is input to the driver. The cone is measured across the horiz plane. The cone is rotated 90 degrees and another measurement is made. The results are filtered to are to exclude the input frequency response.
The only problem i have had is the stepper motor noise, so i had the techs measure in between the step action. We are measuring the residual energy in the cone at different frequencies and position. Note! this position of noise moves across the cone to different positions as the frequency changes.

As stated the only pickle in the results is the difference between the X/Yscans. I can only attribute this to the varing height/mass of the dots.

ron

Ron,

I am perplexed as to why you filter out the frequency components associated with the input. In that case all you would be looking at is nonlinear effects and noise. That is, you would be looking at distortion resiruals plus noise, no? That may be useful information but does it tell us anything about damping of standing waves?

For everyone in general:

Here are some links to KEF on the use of FEA and laser vibrometry


FEA:

http://www.kef.com/kefamerica/technology/new_uniq/finite.asp


Lase Vibrometry:

http://www.kef.com/kefamerica/technology/new_uniq/laser.asp

http://www.kef.com/kefamerica/technology/new_uniq/lasergallery.htm


frequency.

 

[john k...]

Interesting. By chance did the outcome of that test reveal ribbed walls reduced turbulance?

The intent of ribs is to induce trubulance not reduce it. The study was about the cooling passages in jet engine turbine vains. The vains have internal passages through which cool air is pumped to cool the vains so they don't burn up. The passages are ribbed to induce turbulaent and thus increasing heat transfer yielding better cooling. The ribs are quite effective at inducing turbulence.

 

[Ed LaFontaine]

Looks like some Looks like some interesting testing going on there.

Appears to be subtractive. Is your X-Y display a diagonal line with distortion/noise appearing as amplitude/time related blips along that line, or building up to 3D ?interesting testing going on there.

Also, what is the orientation of any ribs of the speaker frame, relative to the testing axis?

 

[dlr]

Ron,

I am perplexed as to why you filter out the frequency components associated with the input. In that case all you would be looking at is nonlinear effects and noise. That is, you would be looking at distortion resiruals plus noise, no? That may be useful information but does it tell us anything about damping of standing waves?

For everyone in general:

Here are some links to KEF on the use of FEA and laser vibrometry


FEA:

http://www.kef.com/kefamerica/technology/new_uniq/finite.asp


Lase Vibrometry:

http://www.kef.com/kefamerica/technology/new_uniq/laser.asp

http://www.kef.com/kefamerica/technology/new_uniq/lasergallery.htm


frequency.

Thanks for the links. I had seen another of them on tweeter movement and surround deformation, but not these.

One thing I noticed in reading the page on Laser Vibrometry is that it's based on surface velocity, not displacement, using Doppler effect changes. Maybe this is newer than some of the systems that may have been based on displacement.

It would be very interesting to see a set of vibrometer measurements for doped paper cone drivers that are highly damped and intended to have a decreasing area of radiation as frequency increases, moving from full surface radiation to an effective ring radiator.

Dave

 

[john k...]

Aha.


Hi John K,

Re the first of your two sine illustrations leading to your *personally* aimed comments of obfuscation and technobabble.

A transient cannot be represented by a burst of single frequency sine wave suddenly being turned on and off.

Yes a transient may be deconstructed into a series of concidentally *starting* sines, but where an overall response becomes reactively modified at any frequency(ies) and the transient itself becomes (audibly) distorted, the wave distortion does not arise at the same frequencies as those which can become established as phase shifted steady state standing waves due to all of the other components still being present in real time !
_______________________________________________



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

What the heck Graham. When Bud states that he isn't talking about symmetric standing waves but rather transient standing waves, what would you call it? If you don't think the statement was made to misdirect attention away from the demonstrated inability of enable to damp standing waves, what was it? There are only two choices. Either Bud doesn't understand the physics, or Bud is intentionally misleading. Take you pick. My figures serve well to demonstrate the transient behavior of a resonance and the development of a standing wave to excitation by the sudden turn on/off of an input at the resonant frequency. If there was no resonance the output would follow the input.

Here is the same system subject to a more complex input. Left is the system with a single resonance. Right is with the resonance damped. Red is input, blue is output.

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

Obviously the resonance in the system at the left severely colors the reproduction of the input over the entire frequency range with severe amplitude and phase distortion at and around the resonant frequency, and shows a long hangover after the input stops. Damp that single resonance and the coloration is removed, as shown at the right.

But you point is well taken as it is important to recognize that a single resonance has a much broader effect than simply altering the response at the resonant frequency. And while it has been shown that Enable doesn't effectively damp standing waves, it has been shown that it does alter them by shifting the frequencies and Q of the (many) standing waves present in the drivers tested and this could well result in very significant audible effects as the coloration of the reproduction may change significantly. Subjectively these could be viewed as improvements. So maybe the correlation between what is measured and what is heard isn't all that mysterious. But if the standing waves are present, then the reproduction is inaccurate and colored.

Frankly this makes the discussion on Enable even less relevant. If you accept that Enable doesn't or only slightly modifies the frequency response, but in some other way improves the sound, then you are still left with the glaring fact that drivers like the Fostex and Lowether, and other full range drivers exhibiting serious breakup above 2k or 3k are highly inaccurate transducers which introduce their own signature coloration on the sound and, as I have said, not worthy of being considered high fidelity. You yourself effectively state this just above, "...but where an overall response becomes reactively modified at any frequency(ies) and the transient itself becomes (audibly) distorted..."

So maybe we should all just pack up and go home because the elimination of standing waves in a cone is by far the more significant than what ever Enable does. And if the damping of standing waves isn't observed in the frequency response of the driver, it ain't happing under transient conditions either. There is no beating around the bush.

 

[john k...]

One thing I noticed in reading the page on Laser Vibrometry is that it's based on surface velocity, not displacement, using Doppler effect changes. Maybe this is newer than some of the systems that may have been based on displacement.


Dave

As I recall now, you always get velocity. Displacement is just post processing. Passing the U(t) data through an integrator to get displacement.

 

[dlr]

Help me a bit here, John

But you point is well taken as it is important to recognize that a single resonance has a much broader effect than simply altering the response at the resonant frequency. And while it has been shown that Enable doesn't effectively damp standing waves, it has been shown that it does alter them by shifting the frequencies and Q of the (many) standing waves present in the drivers tested and this could well result in very significant audible effects as the coloration of the reproduction may change significantly. Subjectively these could be viewed as improvements. So maybe the correlation between what is measured and what is heard isn't all that mysterious. But if the standing waves are present, then the reproduction is inaccurate and colored.

In addition to the linear distortions generated by the FR aberrations, there would be change to the distortion profile.

If a resonance, say at 2K that is creating significant IMD (especially prevalent in full-range drivers, I would think) were shifted higher, then even though still present, it's impact to the IMD would effectively be shifted higher as well, unless I'm misunderstanding how it impacts IMD. Am I right or wrong in this?

If the resonances are shifted, then the relative effects of motor-induced non-linear distortion amplification would be affected as well. Drivers without copper or shorting rings are going to have more noticeable changes, I would think, again being even more noticeable in full-range drivers. All of this could be measured and quantified, as we know. Then correlations could be made between objective changes and whatever changes occurred in perception, of course.

In the end, as you point out, if resonances remain, then the response is still colored and it becomes a question of which colorations are more acceptable.

Dave

 

[Ed LaFontaine]

from post 2793:

Further agreed that the EnABL bands are producing reflections/transmissions.

My curiosity becomes focussed (sp?) on the qualities of those transmissions, particularly in comparison with those transmissions from the junction of the VC and/or surround with the cone.

from dlr:

In the end, as you point out, if resonances remain, then the response is still colored and it becomes a question of which colorations are more acceptable.

yes...yes

 

[planet10]

the glaring fact that drivers like the Fostex and Lowether, and other full range drivers exhibiting serious breakup above 2k or 3k are highly inaccurate transducers which introduce their own signature coloration on the sound and, as I have said, not worthy of being considered high fidelity.

Maybe if amplitude distortion is your primary measure. There are a lot of people that would argue as to their fidelity... they do things that most multi-ways jusy don't have any hope of doing. Too many people are enjoying these to discount that. Seems to indicate to me that there is something that is not getting characterized in existing measurement schema.

dave

 

[Graham Maynard]

Hi John,

"What the heck"? Nice !
Is that what I get for not being hung-up about the wording of Bud's Patent ?

I still don't know what your Post#2935 waveform is supposed to show in relation to a transient. Again this is five cycle burst more like a sibilant event, with expectable res/non-res waveform modifications - not the difference between an original signal voltage and reproduced waveforms due to error during the transduction of a transient.

Either way, I have not assumed that what EnABL is doing to a driver's response is related to cone resonance, though of course any wave-launch modification of air-side resonances must present 'equal but opposite' change to the cone and be measurable through same.

So suppose two test membranes can be constructed which tests EnABL in reverse. That is one with a dot pattern and one without, using both as microphones and exposed to sound along the plane of the membrane, not axially as is normal. Subtract to observe any difference. Maybe even two small loudspeakers with a common single source.

Similar subtractive comparison might be possible with microphones beside treated and untreated ports.

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