The cones that Jordan drivers use have not really changed in the last probably 10 years. Ask Ted why. This is one thing that once you have optimized, you don't want to change with something as significant as coating change. It really takes lots of effort to optimize in design and manufacturing quality control. Not easily done by hand.BudP said:
How the cone and the VC former can be connected depends on specifics of application. Too much glue and you lose detail resolution and the added weight effects effeciency and bandwidth.
Re: easy to deal with ...
The next thing would be how you control the vibration modes of these. Too many vribration modes will let the drivers have a colored sound even when evenly spread. But if you can get the spectral decay to drop more that 12db in less that 0.3ms for frequencies above 1KHz, it will be quite clean sounding. Good luck.
moray james said:
The next thing would be how you control the vibration modes of these. Too many vribration modes will let the drivers have a colored sound even when evenly spread. But if you can get the spectral decay to drop more that 12db in less that 0.3ms for frequencies above 1KHz, it will be quite clean sounding. Good luck.
Trying to understand , again.
Hi Budp,
This enabl thing is very interesting
My last post regarding my understanding of enabl is probably not very good, since I missed the important point, which is removing standing wave, not damping it. Even now I'm still trying to understand the way it really works, so please be patient will my many questions :
I pretty much understand the transmission line analogy, at least in electrical sense, in this case, the transmission line is the cone surface.
and the enabl pattern act like a termination to remove reflection, is this correct ?
this is done by flinging ( I borrow this term from your? post ) the vibration/sound off the surface of the cone.
The part that I still don't have enough understanding is the skin effect.
let's assume paper cone in this case, so at the start of voice coil movement the cone will have vibration that start from cone near the voice coil former, then the higher frequency vibration will have skin effect ? so that most of it will travel inside the cone than the surface ?
Is that's why we need the conformal coating ? to make the vibration/sound travel mostly? on the surface of the cone and to make the vibration travel faster ?
then as the vibration goes to the outer end of the cone, it must first passes the diffraction grid (enabl pattern) where it will sort of dispersed, then to the outer end of the cone, then back to the diffraction grid but the reflected wave will meet some late/delayed dispersed wave caused by outer enabl pattern ?
then
as the vibration meets and they have nowhere to go, then it will be deflected to the air ?
is this visualization correct ?
Hi Budp,
This enabl thing is very interesting
My last post regarding my understanding of enabl is probably not very good, since I missed the important point, which is removing standing wave, not damping it. Even now I'm still trying to understand the way it really works, so please be patient will my many questions :
I pretty much understand the transmission line analogy, at least in electrical sense, in this case, the transmission line is the cone surface.
and the enabl pattern act like a termination to remove reflection, is this correct ?
this is done by flinging ( I borrow this term from your? post ) the vibration/sound off the surface of the cone.
The part that I still don't have enough understanding is the skin effect.
let's assume paper cone in this case, so at the start of voice coil movement the cone will have vibration that start from cone near the voice coil former, then the higher frequency vibration will have skin effect ? so that most of it will travel inside the cone than the surface ?
Is that's why we need the conformal coating ? to make the vibration/sound travel mostly? on the surface of the cone and to make the vibration travel faster ?
then as the vibration goes to the outer end of the cone, it must first passes the diffraction grid (enabl pattern) where it will sort of dispersed, then to the outer end of the cone, then back to the diffraction grid but the reflected wave will meet some late/delayed dispersed wave caused by outer enabl pattern ?
then
as the vibration meets and they have nowhere to go, then it will be deflected to the air ?
is this visualization correct ?
This is very close to how I view it. Please note that my explanation is only a theory. It is backed up with some objective data and a large number of experiments that appear to support this theory, but it is just a theory.
My understanding is this. In a cone, for example, a shear wave is generated at the cone /voice coil junction. This becomes a transverse energy wave. One that looks like an ocean wave due to the energy in the pulse being stretched out in time but iis really a pulse reaching through the cone material to the other side, where it also looks like a wave. As this pulse travels out to the cone edge (for a 3 inch long cone surface, in 2.35 -4 seconds if it were air and faster because it is not) the pulse is continually shedding energy into the air.
As it travels across the cone, in time, the shed energy has become a pressure wave in the air and is also moving up the cone. If the speed of the two is equivalent then a flat energy wave is formed. All of this energy transform is occurring within the boundary layer or media intermix layer at the cone's surface. That the shear wave pulse is diminishing in "amplitude" during this traverse is obvious. Ideally the energy waves "amplitude" is effectively zero at the cone edge and the parting of flat wave in air from the cone is completed at the cone edge, with zero remanent energy left over. Reality is different of course.
When the incomplete transform of transverse wave made of a shear wave pulse leaves energy on the cone it is still a shear wave pulse but one of greatly reduced amplitude and cannot completely reenter the cone. Think of this as a wave traveling on the surface of a pail of water and rising at the wall only to fall back and create a "reflection". This reflection returns across the cone surface as a shear wave of unknown angle, in the boundary layer.
The EnABL pattern and where needed, the conformal coating, appear to allow, or enforce a greater "amplitude" transform during this first pass on the cone , of the shear wave pulse. Then it appears to effect a more complete transform right at the pattern by squeezing the initial shear wave pulse and thus forcing more energy to enter the flat wave than otherwise would happen at this point Then it provides enough of a barrier to the reflected energy from the cone edge to reduce what is allowed to re traverse the cone, in boundary layer, to a very limited amount in very discreet frequencies. The two interferometer vies I have had of this does show a remarkably small amount of energy beyond the EnABL pattern, when compared to what is happening before it. I assume that this left over energy is then damped by the surround.rather than ringing within this tiny area.
That some energy does reenter the cones boundary layer is unarguable. That most of the energy that would have reentered, does not, is also unarguable. That this is an important event, one to encourage, will become unarguable the first time you listen too a treated cone.
Your thought about the importance of a more abrupt gradient change at the outer surface of the boundary layer is a clear one. It is shown to be correct by the lack of need for a conformal coating on the surface of metal cones and polycarbonate cones and domes. Though an aluminum cone is usually aided slightly from a qualitative point of view.
Your understanding of what the mechanism is at the EnABL pattern is just as likely to be correct as mine is since the time of passage for the shear wave pulse will be long enough to effect this inability for the reflected energy to "get back through" the EnABL passages and back down the cone boundary layer.
You still need to add the energy exiting into a pressure wave in the air, over time, to your mental model. The use of "fling into the air" is just an attempt to overcome the 2D model of a diffraction tank that we are all loaded with from our education.
LOWTHER,
For those of you awaiting the last bits of this treatment, it will be delayed by about a week, as I await the arrival of appropriate pen tips.
Bud
My understanding is this. In a cone, for example, a shear wave is generated at the cone /voice coil junction. This becomes a transverse energy wave. One that looks like an ocean wave due to the energy in the pulse being stretched out in time but iis really a pulse reaching through the cone material to the other side, where it also looks like a wave. As this pulse travels out to the cone edge (for a 3 inch long cone surface, in 2.35 -4 seconds if it were air and faster because it is not) the pulse is continually shedding energy into the air.
As it travels across the cone, in time, the shed energy has become a pressure wave in the air and is also moving up the cone. If the speed of the two is equivalent then a flat energy wave is formed. All of this energy transform is occurring within the boundary layer or media intermix layer at the cone's surface. That the shear wave pulse is diminishing in "amplitude" during this traverse is obvious. Ideally the energy waves "amplitude" is effectively zero at the cone edge and the parting of flat wave in air from the cone is completed at the cone edge, with zero remanent energy left over. Reality is different of course.
When the incomplete transform of transverse wave made of a shear wave pulse leaves energy on the cone it is still a shear wave pulse but one of greatly reduced amplitude and cannot completely reenter the cone. Think of this as a wave traveling on the surface of a pail of water and rising at the wall only to fall back and create a "reflection". This reflection returns across the cone surface as a shear wave of unknown angle, in the boundary layer.
The EnABL pattern and where needed, the conformal coating, appear to allow, or enforce a greater "amplitude" transform during this first pass on the cone , of the shear wave pulse. Then it appears to effect a more complete transform right at the pattern by squeezing the initial shear wave pulse and thus forcing more energy to enter the flat wave than otherwise would happen at this point Then it provides enough of a barrier to the reflected energy from the cone edge to reduce what is allowed to re traverse the cone, in boundary layer, to a very limited amount in very discreet frequencies. The two interferometer vies I have had of this does show a remarkably small amount of energy beyond the EnABL pattern, when compared to what is happening before it. I assume that this left over energy is then damped by the surround.rather than ringing within this tiny area.
That some energy does reenter the cones boundary layer is unarguable. That most of the energy that would have reentered, does not, is also unarguable. That this is an important event, one to encourage, will become unarguable the first time you listen too a treated cone.
Your thought about the importance of a more abrupt gradient change at the outer surface of the boundary layer is a clear one. It is shown to be correct by the lack of need for a conformal coating on the surface of metal cones and polycarbonate cones and domes. Though an aluminum cone is usually aided slightly from a qualitative point of view.
Your understanding of what the mechanism is at the EnABL pattern is just as likely to be correct as mine is since the time of passage for the shear wave pulse will be long enough to effect this inability for the reflected energy to "get back through" the EnABL passages and back down the cone boundary layer.
You still need to add the energy exiting into a pressure wave in the air, over time, to your mental model. The use of "fling into the air" is just an attempt to overcome the 2D model of a diffraction tank that we are all loaded with from our education.
LOWTHER,
For those of you awaiting the last bits of this treatment, it will be delayed by about a week, as I await the arrival of appropriate pen tips.
Bud
Started work on the JX92S.
This site shows the publihsed data
http://www.ejjordan.co.uk/drivers/jx92.html
Attached is a measurement I took in March last year.
CSD
SPL
Picture of EnAbled JX92 that will be tested.
Toothpaste is used for this pattern. Although not really accurate, but if it effects the response, it will likely show up. I can also scrap it off after the test to measure weight.
This site shows the publihsed data
http://www.ejjordan.co.uk/drivers/jx92.html
Attached is a measurement I took in March last year.
CSD
An externally hosted image should be here but it was not working when we last tested it.
SPL
An externally hosted image should be here but it was not working when we last tested it.
Picture of EnAbled JX92 that will be tested.
An externally hosted image should be here but it was not working when we last tested it.
Toothpaste is used for this pattern. Although not really accurate, but if it effects the response, it will likely show up. I can also scrap it off after the test to measure weight.
Hi soongsc.
The test should be interesting. After your first tests are done you might look at removing overlap between adjacent block ends on the two different rows. This does change sonics in the painted cones and may or may not be an issue here.
Haven't ever worked with toothpaste on metal before. On the metal domes and cones (Ohm F, Hartkey & many domes) the amount of material was pretty critical for qualitative judgments. I realize you are just trying to get any kind of measureable response, but the process, as applied here, may be a bit over the top.
The Jordan test results shown are quite excellent for resonance control and were I treating a pair I would be using a cut paint and probably no coating at all. I would not expect a big improvement either, in qualitative results, perhaps only a reduction in fricative and sibilance artifacts.
Your pictures are good, but your hosting site doesn't mount them, at least when I go there.
Bud
The test should be interesting. After your first tests are done you might look at removing overlap between adjacent block ends on the two different rows. This does change sonics in the painted cones and may or may not be an issue here.
Haven't ever worked with toothpaste on metal before. On the metal domes and cones (Ohm F, Hartkey & many domes) the amount of material was pretty critical for qualitative judgments. I realize you are just trying to get any kind of measureable response, but the process, as applied here, may be a bit over the top.
The Jordan test results shown are quite excellent for resonance control and were I treating a pair I would be using a cut paint and probably no coating at all. I would not expect a big improvement either, in qualitative results, perhaps only a reduction in fricative and sibilance artifacts.
Your pictures are good, but your hosting site doesn't mount them, at least when I go there.
Bud
Hi Bud
Actually, what I will do next depends on the test results. I've always used toothpaste as a method for test-of-concept which allows multiple turnarounds in a very short time. If you start adding things to the cone, it is necessary to know the added weight to original weight ratio. What exact models have you tested?
The JX92S is known to be very good up to the very high frequencies, I suspect it's the region above the 10KHz region. On other drivers, once this region gets smoothed out, the performance is exceptional. I have not worked the JX92S yet, because that model designed for the JX92 is still on backlog, but I'm trying to slip in some time to do some test-of-concept with these.
Actually, what I will do next depends on the test results. I've always used toothpaste as a method for test-of-concept which allows multiple turnarounds in a very short time. If you start adding things to the cone, it is necessary to know the added weight to original weight ratio. What exact models have you tested?
The JX92S is known to be very good up to the very high frequencies, I suspect it's the region above the 10KHz region. On other drivers, once this region gets smoothed out, the performance is exceptional. I have not worked the JX92S yet, because that model designed for the JX92 is still on backlog, but I'm trying to slip in some time to do some test-of-concept with these.
LOWTHER
Here are the Rev a treatment rings and conic sections for the DX 4. These allow you to put the first or lower whizzer ring just above the ring radiator triangles, as Jon of Lowther America has suggested. Only the relevant conic section and ring guide have changed, from the originals.
These rings should be applied and pictures put up tomorrow, late afternoon left coast USA time. See you then.
Bud
Here are the Rev a treatment rings and conic sections for the DX 4. These allow you to put the first or lower whizzer ring just above the ring radiator triangles, as Jon of Lowther America has suggested. Only the relevant conic section and ring guide have changed, from the originals.
These rings should be applied and pictures put up tomorrow, late afternoon left coast USA time. See you then.
Bud
Hi Bud,
I enabl'ed the area where the dots were. Ofcourse the sound was changed because the dots were gone. So it is hard to say what influence the inside enabl'ement had.
After listenening for a week without the dots I decided to see what would happen if I put them back. The result is that my soundstage is back. What happens with the dot is that the voices are directly in front of me. Without the dot this soundstage is very different (less of a sweet spot). But I do prefer the voices in the middle.
The end result of enabl'ement is in any case with my drivers very very subtle. At times I think it has improved clarity a little...and maybe a little more highs because of the many hard acrylic surfaces. Certainly it has not worsened the sound...
Regards,
Bas
I enabl'ed the area where the dots were. Ofcourse the sound was changed because the dots were gone. So it is hard to say what influence the inside enabl'ement had.
After listenening for a week without the dots I decided to see what would happen if I put them back. The result is that my soundstage is back. What happens with the dot is that the voices are directly in front of me. Without the dot this soundstage is very different (less of a sweet spot). But I do prefer the voices in the middle.
The end result of enabl'ement is in any case with my drivers very very subtle. At times I think it has improved clarity a little...and maybe a little more highs because of the many hard acrylic surfaces. Certainly it has not worsened the sound...
Regards,
Bas
Jx92s
There is a difference between the clean and EnABLed JX92S.
Please note that the measurement sound card is different, and so is the driving amplifier. Black is original, blue is enabled.
I am going to assume that EnABLE process is responsible for the difference and work the other end of the cone.
Here is the CSD
and the original CSD
There is a difference between the clean and EnABLed JX92S.
An externally hosted image should be here but it was not working when we last tested it.
Please note that the measurement sound card is different, and so is the driving amplifier. Black is original, blue is enabled.
I am going to assume that EnABLE process is responsible for the difference and work the other end of the cone.
Here is the CSD
An externally hosted image should be here but it was not working when we last tested it.
and the original CSD
An externally hosted image should be here but it was not working when we last tested it.
Enabled CSD to the same time scale
I think this indicates a cleaner sound.
After all the changes, it would be interesting to see the measurements with everything scaped off.
An externally hosted image should be here but it was not working when we last tested it.
I think this indicates a cleaner sound.
After all the changes, it would be interesting to see the measurements with everything scaped off.
soongsc,
Thanks for that!!!!!. Even epastel works as it should.
The frequency response does not change greatly, so the "character" remains the same, but the energy appears to show an initial drop in about half the time that it takes the untreated cone to loose enough of it's original signal energy to do the same.
The middle period doesn't appear very different in level though it is frequency shifted to a degree. I do note about a 15 dB decrease in the worst of the systemic ringing, out at 0.75ms which is encouraging.
Can you run some tests that look at low level ringing to see if that portion of improvement is also shown? I am not sure how the metal cone will handle low level signals and am certainly interested to see, perhaps out as far as 2.0 ms or further?
Regardless, thanks again for this much work it is very helpful.
Bud
Thanks for that!!!!!. Even epastel works as it should.
The frequency response does not change greatly, so the "character" remains the same, but the energy appears to show an initial drop in about half the time that it takes the untreated cone to loose enough of it's original signal energy to do the same.
The middle period doesn't appear very different in level though it is frequency shifted to a degree. I do note about a 15 dB decrease in the worst of the systemic ringing, out at 0.75ms which is encouraging.
Can you run some tests that look at low level ringing to see if that portion of improvement is also shown? I am not sure how the metal cone will handle low level signals and am certainly interested to see, perhaps out as far as 2.0 ms or further?
Regardless, thanks again for this much work it is very helpful.
Bud
Hi Bas,
I have had occasional problems with the center between two drivers being somewhat weaker after treatment. Not in all drivers, but certainly some systems I have treated had this result.
Did you treat the center dome to a ring of blocks at cone join and out at the very center? This is usually helpful in these circumstances. In some drivers there appears to be some sort of phase or time or energy redirection change in the relationship between cone and dome. The major reason I truncated the two sets of rings at the cone dome junction down into a single three ring set at the that junction was to help eliminate this difficulty.
The coated Vifa 6 inch cones are particularly problematic. You can literally suck the center out from between a pair and extend the outer "sound stage " boundaries, with relative thickness of conformal coating materials, after the pattern is on both locations.
It is good that it does not sound worse, though I have always thought of the process as removing unwanted noises rather than making it sound "better" in some way. Perhaps if you do try the center pattern, even if the dome is an open weave mesh, some of that middle ground will return, if you want to experiment further. Or, leave it like it is if it sounds good to you. I certainly see nothing wrong with the dots.
Bud
I have had occasional problems with the center between two drivers being somewhat weaker after treatment. Not in all drivers, but certainly some systems I have treated had this result.
Did you treat the center dome to a ring of blocks at cone join and out at the very center? This is usually helpful in these circumstances. In some drivers there appears to be some sort of phase or time or energy redirection change in the relationship between cone and dome. The major reason I truncated the two sets of rings at the cone dome junction down into a single three ring set at the that junction was to help eliminate this difficulty.
The coated Vifa 6 inch cones are particularly problematic. You can literally suck the center out from between a pair and extend the outer "sound stage " boundaries, with relative thickness of conformal coating materials, after the pattern is on both locations.
It is good that it does not sound worse, though I have always thought of the process as removing unwanted noises rather than making it sound "better" in some way. Perhaps if you do try the center pattern, even if the dome is an open weave mesh, some of that middle ground will return, if you want to experiment further. Or, leave it like it is if it sounds good to you. I certainly see nothing wrong with the dots.
Bud
I think we should be aware of differences, but not jump to conclusions yet. Bear in mind that the driver might not be the same one I used with the first test which was a year ago. The room is different as well.BudP said:
The second test is probably at half the amplitude if the first test. So I'm not sure what level you are looking at. Normally it's days or a week till I go to testing again. So if there are specifics you all would like to look at, please be as specific as you possibly can.
I am planning to move the rings a little inward to see how it will change the results.
- Status
- Not open for further replies.