There you go. You have exactly what EnABL is supposed to provide, that extra 50 db down from signal level, below the more normal 40 db down, and all of it coherently. What ever the mechanical limits of the driver are will be expressed, right to that limit.
Nice that you are experienced enough with single channel listening to already know what you have. Just wait until you get both channels up and running and perhaps reduce the localization of the baffle they are on with further patterns, until all of the sound comes out of the wall in back of the speakers and they are sonically invisible. Then you will find yourself starting in on the rest of your system. Until, one day, you find yourself awakening from a listening session...literally... and feeling unusually refreshed and delighted. I find the very same set of refreshments from an afternoon spent in meditation, though I do not go to sleep.
As for another pattern set needed, you can bet on it. Best to take one of what you have in the PDF, measure it and determine how much bigger it needs to be in %. Then print another patterns page at that % of increase. Perhaps a couuple of pieces of paper later you will hve exactly what you want.
Bud
Nice that you are experienced enough with single channel listening to already know what you have. Just wait until you get both channels up and running and perhaps reduce the localization of the baffle they are on with further patterns, until all of the sound comes out of the wall in back of the speakers and they are sonically invisible. Then you will find yourself starting in on the rest of your system. Until, one day, you find yourself awakening from a listening session...literally... and feeling unusually refreshed and delighted. I find the very same set of refreshments from an afternoon spent in meditation, though I do not go to sleep.
As for another pattern set needed, you can bet on it. Best to take one of what you have in the PDF, measure it and determine how much bigger it needs to be in %. Then print another patterns page at that % of increase. Perhaps a couuple of pieces of paper later you will hve exactly what you want.
Bud
posted this from my other thread detailing my destruction of this driver. there has been a fairly substantial set of shifts in the FR as well as the detail increases. now to be sure i wasn't just detecting the extra brightness i EQ'd it down by the relative gain and the detail was still there.
Ali Farka Touré & Toumani Diabaté - Ruby
is my listening track and the EnABL process really brought out the detail in what was going on and a distinct separation betwen the instruments going on. really revealed the resonances and detail in the 'plucks' of the strings. its really weird that all the information was already there in the speaker, it just distorted itself out of reproducing it.
obviously its still a very flat sounding thing, hopefully with some tweeter help it will start to become something that i could actually use in my big room.
Don't hesitate to add a third row to any pattern set,even a fourth. Not a full pair per row, just one more row offset as usual, one at a time. Even a row that is misplaced will not degrade the sound noticeably. Once the driver is free of the factors that cover over much of what the driver is actually producing it is hard to ruin them. Too much gloss can do so.
The Zig 2 way glue pics are back a few dozen pages and have been mounted a few times in this thread, so, poke around a bit.
What other thread are you posting on for these experiments? Can we go see?
Bud
The Zig 2 way glue pics are back a few dozen pages and have been mounted a few times in this thread, so, poke around a bit.
What other thread are you posting on for these experiments? Can we go see?
Bud
how rude of me, terribly sorry.
http://www.diyaudio.com/forums/full-range/209490-phase-plugging.html
i just experienced what you said about the speaker disappearing. just changed the wool around the spider and then added a lining to the inside frame (suspected this has been causing problems from the start). i have the speaker currently on its own in 'near field', about 12 inches from my face. one passage on a track by the 'flaming lips' made me look over my shoulder as i thought there was something making a noise behind me. really quite odd what something this relatively simple does :S
Last edited:
and here's the stock VS phase plug mod VS phase plug + EnABL process (with some basket mods)
An externally hosted image should be here but it was not working when we last tested it.
I am always left wondering if I actually heard the differences that an FR plot provides. Mostly they seem to only be a specific tuning on an instrument that has altered, rather than something of importance.
In any event, here is something I occasionally enjoy doing. Take the EnABL'd driver, I assume it is still in free space, and sit it on the table in front of you, on it's magnet, with you off at a steep angle nearly looking across the top of the driver from one side to the other, but not able to see down inside, perhaps a foot away. Listen to a group of singers perform. What is the illusion? Now try it with the untreated driver.
Bud
In any event, here is something I occasionally enjoy doing. Take the EnABL'd driver, I assume it is still in free space, and sit it on the table in front of you, on it's magnet, with you off at a steep angle nearly looking across the top of the driver from one side to the other, but not able to see down inside, perhaps a foot away. Listen to a group of singers perform. What is the illusion? Now try it with the untreated driver.
Bud
is this where you can kinda hear them coming from the middle of the speaker through the frame over somewhere further away? i think i know what you mean if not that as i noticed something at the perimeter of the frame when listening.
here's an interesting thing for you to gaze your eyes upon.
i went back and did the tap test and looked again for anomalies and placed a single type ring at the coordinates. leaving them un-glossed i took some more frequency plots, here's what happened.
the blue line is the driver using the accepted EnABL protocol before i had at it again
the green, red and yellow are my attempts at removing variables. these were wool based wodding around the whizzer and some on the basket arms.
needless to say they sound horrible now, even by their usual standards! i just applied some gloss to the new EnABL spots as can be seen by the sheen in the picture.
any ideas?
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Well, what kind of horrible? Harsh, dull, edgy, screeching....?
Have to admit you have sort of backed into a corner here, single rings out in space as it were don't work at all like a pair or three or four. You could just slip some extra ring sets, rather thinner than what you have now into the spaces and make this a continuous pattern set. Certainly what I would do at this point, sort of a "how bad can it be?" attitude. There is a set of drivers, most recently deeply looked into with wavelet analysis, that required a continuous pattern on the entire cone. Astounding release of information was the audible result, but I really did not like listening to them.
Things to know. The outer pattern sets control the high frequencies, with inner ring sets controlling low frequencies. The whizzer is supposed to suppress the high frequencies coming off of the main cone and you may have released them instead. Get some Zig glue and describe what has changed and I suspect we can paint opalescent purse shapes over this ear after a bit.
Don't put any more gloss on for now and when you do, make it a 10% gloss to 90% water mix, using a syringe body to ensure that ratio.
Bud
Have to admit you have sort of backed into a corner here, single rings out in space as it were don't work at all like a pair or three or four. You could just slip some extra ring sets, rather thinner than what you have now into the spaces and make this a continuous pattern set. Certainly what I would do at this point, sort of a "how bad can it be?" attitude. There is a set of drivers, most recently deeply looked into with wavelet analysis, that required a continuous pattern on the entire cone. Astounding release of information was the audible result, but I really did not like listening to them.
Things to know. The outer pattern sets control the high frequencies, with inner ring sets controlling low frequencies. The whizzer is supposed to suppress the high frequencies coming off of the main cone and you may have released them instead. Get some Zig glue and describe what has changed and I suspect we can paint opalescent purse shapes over this ear after a bit.
Don't put any more gloss on for now and when you do, make it a 10% gloss to 90% water mix, using a syringe body to ensure that ratio.
Bud
Last edited:
i literally have no way to describe it. its just wrong. it was poor before but still 'right' and the EnABL process did its job and revealed more information. now its just wrong.
i seem to have lost a tonne of bass/low end from the noise and i'm not sure how?
being fairly captivated by this whole thing i usually take readings after each step so i know whats happened when i've cocked up etc. this is 2 identical recordings of eva cassidy - fields of gold, taken pre EnABL and post cluck up, under what is basically highly similar conditions. the spectrum is then plotted.
this is what has changed.
i could have just destroyed a driver but i'm actually quite happy, its really fascinating stuff. also fairly glad i did it on a $60 driver and not a $400 audio nirvana job.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
So am I!!!!!!!. By the way, we have the AN drivers pretty well figured out by now. If you buy them, splurge and get the 15's. They are rather more than impressive when treated and don't need ANY gloss at all.
Sounds like you have upset the phase relationship between the actual minimum phase portion of the output and the upper FR. Can you pull phase plots out of your previous data? Look here for Soongsc's experiments and measurements. Really caused a ruckus with the folks out to prove EnABL is "Bunk".
http://www.diyaudio.com/forums/multi-way/100399-enabl-processes-26.html#post1227789 multiple rings phase vs frequency response
These ridiculous rings have been beguiling me since 1973, so I get your fascination. So strange that so little mass applied in just the right place can cause such audible changes, while causing measurable changes that are well within the normal range over a run of drivers. Obviously many more things hidden within the FR test than we currently know about.
You may be able to "erase" those extra rings if you want to.
Bud
the CSD plots didn't work very well when i tried them, if that is what you're looking for.
i'm happy to try and fix them, i figure its a good way to learn how to not ruin good drivers! i'm fairly well convinced that the EnABL process does indeed bring about some fairly obvious changes. the reason i chose a big ugly driver to use as my first play thing and not a 3 incher is because i figured it was like cabinets, the bigger they are the more they suffer from resonances and distortions. thus any attempt to counter these is going to show up more on a driver of 12 inches rather than 3.
what should i try next? i've got room to pair up those rogue rings on the main cone up with smaller efforts. is the pattern on the very lip of the whizzer a problem? they are paired sequentially, just on different faces.
i had my eye on the audio nirvana super 8's as its fairly similar to more expensive fostex drivers like the FE206. so once again carries less financial immorality if i ruin them. the super 10's also seem well received. the super 12's i've read some poor opinions with regards to the sound being all over the place (similar to these fanes and its not something i enjoy) and not found anything about the 15's. nothing concrete yet as i've also seen some sonido drivers that look pretty, although for considerably more money.
update this info.
this is a plot of the frequency response after that coat of gloss had dried. you can't see at that resolution but the bass has recovered somewhat.
what exactly is the gloss doing in the entire process? its impact is almost as profound as the culprit rings!
this is a plot of the frequency response after that coat of gloss had dried. you can't see at that resolution but the bass has recovered somewhat.
what exactly is the gloss doing in the entire process? its impact is almost as profound as the culprit rings!
An externally hosted image should be here but it was not working when we last tested it.
The Gloss is a conformal coating material that helps to suppress individual loop back zones. These zones can be seen in the following four u tube resonance movies.
resonance - YouTube
resonance2 - YouTube
resonance4 - YouTube
resonance3 - YouTube
The material appears to take out the "monkey nuts" you will shortly learn about....
Then there is a wavelet analysis for untreated and treated drivers of the same type. Here is a blink comparison provided by Dave of Planet 10
Enable Tests
Roll your pointer on and off of the edge to get the blink going. Notice that the untreated driver has a pattern that is repeated at large amplitude over the first 4 ms or so. The treated driver also has a repeated pattern, but the amplitude is much lower and the pattern is longer in duration. This is due to both EnABL and gloss applied to a truly difficult driver.
The full report is here.
http://www.diyaudio.com/forums/multi-way/119677-enabl-technical-discussion-88.html#post2350234
Scroll up from where the link deposits you. This is in the technical argument thread.
Bud
resonance - YouTube
resonance2 - YouTube
resonance4 - YouTube
resonance3 - YouTube
The material appears to take out the "monkey nuts" you will shortly learn about....
Then there is a wavelet analysis for untreated and treated drivers of the same type. Here is a blink comparison provided by Dave of Planet 10
Enable Tests
Roll your pointer on and off of the edge to get the blink going. Notice that the untreated driver has a pattern that is repeated at large amplitude over the first 4 ms or so. The treated driver also has a repeated pattern, but the amplitude is much lower and the pattern is longer in duration. This is due to both EnABL and gloss applied to a truly difficult driver.
The full report is here.
http://www.diyaudio.com/forums/multi-way/119677-enabl-technical-discussion-88.html#post2350234
Scroll up from where the link deposits you. This is in the technical argument thread.
Bud
feels like the driver is shrieking a bit now.
i'm sorry to be a pain but i can't find in this thread the description of what i need to do with zig zags and PVA. if you have the spare time could you please give me a quick description of what i'd need to do please?
i've updated my main diary of the modifications in the full range forum to include the latest FR reading and i managed to get the CSD responses to work properly.
http://www.diyaudio.com/forums/full-range/209490-phase-plugging-4.html#post3018451
An externally hosted image should be here but it was not working when we last tested it.
judging by that, would i be able to say its like the 6khz range causing the unpleasant high end?
any ideas on how to improve the bottom end, as that doesn't look so great and perhaps even has worsened since i got my hands on it.
thanks for your guidance.
Last edited:
ok some more feedback on the EnABL process.
music created by purely electronic processes, or containing mostly electronically fed/created signals sound completely 2d, flat and lifeless. the visual representation would be comparing a picture of a life to the actual real life event. both look pretty much identical but they lack that depth that makes it feel 'real'. highly lossless MP3 etc type music sounds truly horrible. compared then to music recorded through multiple studio mics and separate channels now sound 3d, with depth and position.
anything less than a 320kbs electronic file will make you cry through EnABL'd speakers. i have quite a lot of stuff ripped into V0 (donated most the hard copies to charity as it was taking up an entire room) that is going to have to be purchased again.
music created by purely electronic processes, or containing mostly electronically fed/created signals sound completely 2d, flat and lifeless. the visual representation would be comparing a picture of a life to the actual real life event. both look pretty much identical but they lack that depth that makes it feel 'real'. highly lossless MP3 etc type music sounds truly horrible. compared then to music recorded through multiple studio mics and separate channels now sound 3d, with depth and position.
anything less than a 320kbs electronic file will make you cry through EnABL'd speakers. i have quite a lot of stuff ripped into V0 (donated most the hard copies to charity as it was taking up an entire room) that is going to have to be purchased again.
Flac files through a good asynchronus usb to spdif twist loc BNC, out to a dac sound marvelous. Far better than the expensive Sony I borrowed for a comparison, though the most recent Opus sounds very close to the Berkley Alpha Dac I also have on loan.
Zig 2 way glue is used to quell shrieks, specifically those found hiding out at the surround to cone glue joint, the surround itself and upper and lower portions of most whizzer cones both front and back sides and the outer whizzer cone edge. It will also kill the Raleigh wave that circulates on the outer 5% of the cone in almost all cone drivers. Just apply to the back side until the blurring and twisting sensation go away. Typical two to three coats, full strength, with one to two coats on the backside, beginning portion of the surround.
The whizzer often needs alternating coats of Zig and 10% gloss to water, down at the bottom 20% and up at the top 10%, on the front face. That combined with Zig on the back, with a final coat of 10 % gloss contains most of the problems. You do need to go slowly here, over a number of days, just to be sure you have each coat cured and are secure in what has changed.
Not much to be done for the Bass. The loss you show is always there for treated drivers. More with added mass of course. I think most of the energy that has disappeared is just rumble that has been created by incoherent wave forms created by loop back mechanisms. These actually appear to be room related, in that with their dispersal a number of room resonances, hot spots and nulls just disappear.
Bud
Zig 2 way glue is used to quell shrieks, specifically those found hiding out at the surround to cone glue joint, the surround itself and upper and lower portions of most whizzer cones both front and back sides and the outer whizzer cone edge. It will also kill the Raleigh wave that circulates on the outer 5% of the cone in almost all cone drivers. Just apply to the back side until the blurring and twisting sensation go away. Typical two to three coats, full strength, with one to two coats on the backside, beginning portion of the surround.
The whizzer often needs alternating coats of Zig and 10% gloss to water, down at the bottom 20% and up at the top 10%, on the front face. That combined with Zig on the back, with a final coat of 10 % gloss contains most of the problems. You do need to go slowly here, over a number of days, just to be sure you have each coat cured and are secure in what has changed.
Not much to be done for the Bass. The loss you show is always there for treated drivers. More with added mass of course. I think most of the energy that has disappeared is just rumble that has been created by incoherent wave forms created by loop back mechanisms. These actually appear to be room related, in that with their dispersal a number of room resonances, hot spots and nulls just disappear.
Bud
i applied the Zag and it didn't do anything to the FR or that shriek. i located it to the gap between the whizzer cone and the phase plug and lightly stuffed it with cotton wool and suppressed it by quite a bit and removed a few other wobbles.
on a side note, given your experience with audio nirvana drivers can you add anything to this thread (if you have time to spare).
http://www.diyaudio.com/forums/full-range/212474-any-updates-accuracy-audio-nirvana-t-s-specs.html
Here are the pattern guides for the Audio Nirvana Super 15. Easily the best large format full range driver ever and once EnABL'd easily one of the best drivers of any kind ever. Seriously, these things are better than the best flat panel drivers.
Bud
Bud
Hi Bud, there appears to be 3 variations of the AN Super 15; ferrite, neo, and alnico, which one were you referring to? And if you have heard the different variants what are the differences? Thanks.