EnABL - Technical discussion

Jump to Latest
#821
BudP said:
Here's something for your brain John. Soongsc, quite far back in the original thread, discovered a mid cone position for the rows of pattern that completely dispersed the major resonant node on his test subject, a Jordan JX92. The test data is there in the thread.

I happened to have two pairs of the Jordan drivers on hand and I duplicated his placement of the toothpaste pattern bars, with acrylic paint. Of course I did not measure a thing, but the audible effect was interesting, revealing a lightness and agility to the reproduction of music I did not hear from the drivers that were untreated.

I have been experimenting with this effect, having found that the placement of the rings on the Jordan corresponded with an audible change in tone and a seeming change in decay direction, from towards the voice coil, to indeterminate and then towards the surround, when lightly struck with a small stick. This radial position for change in tone comes from a barely controlled taping on the driver surface, in a radial manner, with the tip of one of the pen nib holders, a narrow, relatively hard and light tool.

Once I learned what I was listening for, I found these change points on every driver I could use this method on. The pattern rows now adorn the entire surface of a treated driver and from an on axis view point, are very evenly spaced, even though their radial distance apart, on the actual surface of the cone, may not be, due to curvature. This distribution holds true for whizzer cones also and the first placement on a cone, under a whizzer cone, corresponds to the outer diameter of the cone.

The result of applying these distributed rings is an elimination of those radially dependent tone variations and elimination of the seeming change in decay vector, with all decay pointing toward the surround.

Audible changes area striking clarity, without superficial edge, and what appears to be theoretically correct dispersion, without audible combing. Much the same as the original pattern sets provided, but more effectively eliminating the shrieks and comparatively poor dispersion patterns, along with the extra downward dynamic clarity.

I offer this and invite you to experiment with a typical cone driver, though domes do show the same characteristics for the few I have dealt with in this fashion.

So, there you go, yet another "mystical treatise on the art of attaining truth".

Bud
Click to expand...

Hi Bud,
I'm ready now, finally, to try this on my JX92S. A couple of questions:
- do I have to take the drives out of their cabinets to do this tapping exercise or is it more effective to leave them in cabinets?

- this tapping exercise needs to be done in a ring all the way around the mid-cone, marking where the directionality of the decay changes from inner facing to outer facing?

- Are yo now saying that the ENaBL ring is placed following this line marked around the cone ie. it won't necessarily be a symmetrical ring pattern as in the template but may have some variation in placement dependent on the result of this test? Or are you saying that there is no longer a ring the whole cone gets tested & a pattern is applied at each of these transition points? If so, what is the unit pattern that gets applied at each of these points ?

- apart from the mid-cone, this also needs to be done for the other rings also? What is the sonic tapping test for these rings or is there one?

- I may just stick with the mid-cone one to start with to prove the efficacy of this treatment.
 
Last edited:
#822
Elias said:
Hi,

Is anyone yet tired to Enable his earflaps?

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



The pattern in the picture may not be correct, but you get the idea!
Click to expand...



Ok folks - this is the "EnABL - Technical discussion" thread - is it ?

Bullets don't seem to be as big around here...

🙂
 
Last edited:
#825
jkeny said:
Hi Bud,
I'm ready now, finally, to try this on my JX92S. A couple of questions:
- do I have to take the drives out of their cabinets to do this tapping exercise or is it more effective to leave them in cabinets?

- this tapping exercise needs to be done in a ring all the way around the mid-cone, marking where the directionality of the decay changes from inner facing to outer facing?

- Are yo now saying that the ENaBL ring is placed following this line marked around the cone ie. it won't necessarily be a symmetrical ring pattern as in the template but may have some variation in placement dependent on the result of this test? Or are you saying that there is no longer a ring the whole cone gets tested & a pattern is applied at each of these transition points? If so, what is the unit pattern that gets applied at each of these points ?

- apart from the mid-cone, this also needs to be done for the other rings also? What is the sonic tapping test for these rings or is there one?

- I may just stick with the mid-cone one to start with to prove the efficacy of this treatment.
Click to expand...

I know this is more an implementation question but since I didn't see it come up on the other thread, I thought I'd ask about a followup from Bud or someone. I had the same questions as above to Bud's earlier post (#817).
 
#826
I went away on holidays & haven't had a chance to do this since I got back but I think I can answer these questions now & save Budp from writing the same information that he sent me - BudP will correct me if I'm wrong, I'm sure 🙂

- no you don't take the drives out of their cabinets
You will need to tap very lightly and just bounce the tip, of the pen holder preferably, off the surface and listen carefully to the decay. Move in a radial line, from VC to surround and note where the overall character of the tap changes and where the decay direction changes. After a bit this should become fairly obvious.

The patterns should be applied close to the outer edge of the pattern guide. As you can see on the photo, they are rather thinner than for some other drivers and the three on the cone are pretty close together.

After the paint has dried for a few hours, redo the tap test. You should find the decay going in one direction and only a very gradual shift in sound character as you tap on the radial line from VC to surround. You can apply the tings one after another, from outer to inner, in one sitting, if you can stand to be that controlled for that long. I cannot and so it takes me a couple of days to treat a driver.

You may need a set of magnifying lenses, likely from a local hobby shop, to allow you to see the printed pattern and resulting small droplets placed on the center dome.
Click to expand...
 
#827
Just a couple of added notes.

The tap test audible results are subtle in nature, until you have zeroed in on listening to the entire tap and decay combination. Your first take is likely to be, "nonsense, how can that tiny change be important?" What we are investigating is the nonlinearitiy in energy expression, apparently caused by the physical construction of the cone. These subtle shifts are fixed in place and are the cause of most of the "speaker" modifications to the sounds recorded on our favorite media. The tap test just allows you to zero in on each one and disperse the effect.

When I get Michael (mige0) his Selenium drivers with EnABL applied according to this scheme, we may see the differences in the test data he plans to provide.

You can leave the drivers in the cabinet for the tap test, but you will have to remove them to apply the pattern rings.
 
#828
Thanks for the notes Bud - forgot about removing the drivers for patter ring application.

I expected the tap test to be difficult but found it relatively easy to identify the transition area. It was surprising to hear it 🙂 I greatly appreciate this advance in how to recognise where to apply the rings, it gives us autonomy in doing this ourselves. It also begins to zone in on possible mechanisms of action.

I'm also excited by the prospect of some test data from mige0 & finally the beginnings of some measurement data for the effects. Even though I'm sure there will be plenty of the usual naysayers.

Full marks Bud for sticking with this, it was not an easy ride 🙂
 
#829
When I provided Dave Dlugos with the method, his comment was "Oh, so it is just one or two added rings in slightly different places?" And that is the truth of the matter. We were close to a correct solution with the original generic pattern placements, with Gloss used to extend the effect of the slightly misplaced patterns.

One of the reasons I provided EnABL to this forum was that I was certain that the sloppy thinking I am noted for would be refined. That Soongsc was responsible for the breakthrough that lead to what I suppose we should call EnABL 2.0 has been a great delight for me. The hammer and anvil work done by John K, dlr, Seretin and others is also deeply appreciated, I could not have found a better peer review.

So far, this test regimen has provided notably superior results and the use of Gloss has dropped by a factor of 4 or more. A recent Mark Audio Alpair 7 application utilized a 10% ratio of Gloss to clear water, to provide the pattern effect to those resonance points that are not discoverable by this still rather gross test.

SY has pointed me towards fractal pattern construction for these ever smaller resonance control mechanisms. At some point we might even collaborate on his specific ideas for an appropriate construct.
 
#830
Ah, so that's what the gloss is for 🙂 - I didn't know that. I understand the idea of extending the effect but the gloss is used all over the cone?

Yes, it has come a long way & maybe it will become a very exact & scientific approach eventually without any gloss?

I'm surprised SY even gives this any time as I thought he lived by the motto "if it can be heard then it can be measured in the analogue domain" I take it we are some way away from this goal still? Now a collaboration would be interesting.
 
#831
Looking at John K's blink http://planet10-hifi.com/johnK-test/ comparison and noting how the daughter nodes of that large resonant node are affected, by just the application of pattern rings at the voice coil / dust cap and surround positions, I find the following to be logical....

An extension of patterns to either side of the main EnABL 2 nodal pattern rings should provide dispersal of those ever smaller resonances that disturb the signal from a flat FR in the CSD.

That the Gloss coating has been such an important part of the effects of applying the patterns as John K did, and that the EnABL 2.0 patterns reduce this effectiveness to the degree they do, supports the thought that smaller patterns arrayed around the main nodal dispersion could eliminate the Gloss entirely. Not sure that this is the best plan for some ultimate goal, but, for most of us I think there will be a threshold beyond which the improvements wrought by ever smaller patterns sets will only be important to folks like Romy the Cat.

Aside from that, I have found SY to be an enthusiastic investigator of all phenomena. I find his rigor to be refreshing. If he says something is not working in a particular set of circumstances I am pretty certain that is what I will also find. He is always careful to outline just now extensively his comments are to be applied.

I am actually also of the persuasion that if it can be heard it can be measured. My problem has been that my skill set was not up to discovering what John K finally uncovered in his blink tests. And I still wish the other three tests were available, as they were enlightening to say the least. But John K and SY are rigorous in their approach. My skill set is off in a different direction and I could not do without them.
 
#832
Do you have a link to John K's blink tests?
What you say makes sense about eliminating the gloss entirely - it may be a case that at a certain threshold a less exact & more blanket approach using the gloss is warranted!

I've no doubt that anything that can be heard SHOULD be able to be measured BUT I just thought that the measurement techniques or instruments weren't sophisticated enough yet! Is enaBLe not an example of this shortfall between what can be measured & what can be heard? I will always lean towards the side of hearing rather than measurements because of this but it is delightful to see measurements supporting what is heard.
 
#833
I think the measurements do show what EnABL is about. I did insert the link to John K's blink in the preceding post, first sentence pointing to planet10hifi.

The problems have been interpretation and a lack of low level information, hiding under the loudest. Not sure that the current favorite signal generation method is capable or incapable of providing this information. Certainly the processing software is not yet looking clearly at downward dynamic information, though the post processing that mige0 is doing with wavelets might be just the tool.

When you look at the blink comparison, look specifically at the resonance node and it's increase in orderliness of decay resonant structures. And in general how much more orderly the FR band above the loss of strict minimum phase behavior is. This being generally referred to as the cone breakup point in FR.

EnABL is the devil in these detail changes and it is no wonder to me that I never found it on my own, just not rigorous enough.
 
#835
Sorry, Bud, I missed that link - the links just don't stand out from text for me!
Yes it looks like it's a lack of interpretation, rather than lack of measurements.

BTW, I see Soongsc mentioning flattening out the impedance curve of the JX92S speakers before applying enaBLe treatment - is this a BSC or frequency compensation that he is speaking of? I can't find any other reference to it.
 
Last edited:
#837
jkeny said:
...
BTW, I see Soongsc mentioning flattening out the impedance curve of the JX92S speakers before applying enaBLe treatment - is this a BSC or frequency compensation that he is speaking of? I can't find any other reference to it.
Click to expand...
Flattening of the impedance curve is nor related with any cone treament. But if you use zobels to flatten the impedance curve, the sound will be more balanced and the image depth will be more correct. I've been thinking about doing more work with the JX92S, but simply had not had the time to do it. I'm sure that once you get the JX92S properly treated, and with the proper zobels, you will find that there are other areas to explore since the speaker is now more transparent. I've climbed the signal path up to the interconnects, and one person doing the listening comparison using the Exaltation (active speaker currently still in certification process) call me up and said that even his wife could hear the improvement. His son also got real excited.

What I'm saying is that anything you can do to get the speakers to sound clean and detailed, the better you will hear improvements/problems in other areas.

I would recommend placing speakers back against a stiff wall with sound absorbing material instead of using BSC.
 
#838
Thanks Soongsc,
I know it's independent of cone treatment so what you are saying is that it's probably needed for best results!

I went back to the original mention of this by Erictoucan & emailed him
It appears to be a parallel combination of LRC : L~0.5 mH, C~10uF, R~10 ohm. adjusted to taste.

My JX92S are in a different sort of enclosure - called 3D Spiral Horns http://www.diyaudio.com/forums/multi-way/119903-3d-spiral-horns-anyone.html
I guess based on their construction that they are more TL than Horns but they are bookshelf size (about the size of LS3/5a) but kick out a nice bass.

I don't use a BSC are you calling the above zobel a BSC & it's not needed - can use placement & sound absorption instead? I'm confused now!
 
You must log in or register to reply here.
Top Bottom