I can't see where have more room reflections make anything more meaningful. Just me I suppose,
Dan
I was really scratching my head about this wiggle of the TD15M. Looked up half a dozen responses of 12" & 15" drivers at Brandon's Driver Measurements - drivervault and all had that same thing more or less. 
I think it is a delayed damped back-bounce from the shockwave through the type of spider used for that sort of pro drivers, isn't it? Or is that too simple?
If it were stable (linear) and predictible in character pre-distortion of the driving current (per FIR-filter in the digital domain) should be able to correct for this, trying to terminate the reflection mechanically. Essentially the same strategy that digital room-equalization tries to use, but this time the dimension of the problem is 1, it's 1D and therefore much better suited to impulse response correction techniques, to the extend that the effects are not too non-linear (drifts included).
Ony may argue though if it's worth the effort, or favorable at all, besides the "digital stuff" involved. Well, maybe that delayed wiggle is part of the trademark sound of this class of drivers...
I think it is a delayed damped back-bounce from the shockwave through the type of spider used for that sort of pro drivers, isn't it? Or is that too simple?
If it were stable (linear) and predictible in character pre-distortion of the driving current (per FIR-filter in the digital domain) should be able to correct for this, trying to terminate the reflection mechanically. Essentially the same strategy that digital room-equalization tries to use, but this time the dimension of the problem is 1, it's 1D and therefore much better suited to impulse response correction techniques, to the extend that the effects are not too non-linear (drifts included).
Ony may argue though if it's worth the effort, or favorable at all, besides the "digital stuff" involved. Well, maybe that delayed wiggle is part of the trademark sound of this class of drivers...
Last edited:
Delay time of 2.3ms seems to be too long for that to enter the picture IMO
*If* its in the driver itself, I'm rather with John K that it may be from cone modes in break up.
There is another reflection to be identified earlier on (and slighty higher in frequency) but not exactly half of that 2.3ms - so it might or might not be part of the mechanism - for sure its *not* a simple looped reflection, otherwise time would be pretty exactly halved and also we should see pronounced subsequent reflections later on.
There is also a specialty with those speakers in general - the are more of a horn . actually a moving horn - than other speakers, as they have no dust cap and hence more "horn length"
It could well be we see an interference (beat) of the sound created at the throat+ "amplified by the horn" with the sound created at the rim.
Michael
Last edited:
Are you referring to this one ?
http://www.diyaudio.com/forums/multi-way/100392-beyond-ariel-138.html#post2257478
http://www.diyaudio.com/forums/multi-way/100392-beyond-ariel-138.html#post2257509
http://www.diyaudio.com/forums/multi-way/100392-beyond-ariel-138.html#post2257524
If so - what are you telling us ?
I didn't got the point and also how its related to Dan's measurements ?
Michael
Ok – here we go:
Full dampening
Less dampening
Zero dampening
Michael
I think just the full 5 msec window tells a lot. That thing is nasty above 1k Hz. Time to look at a measurement with a 500 Hz 4th order LP filter.
Here is an implse response of a simple system:
Want to see what the FR looks like?
An externally hosted image should be here but it was not working when we last tested it.
Want to see what the FR looks like?
Michael, look here: http://www.diyaudio.com/forums/multi-way/100392-beyond-ariel-695.html
I swear I think you guys try to make things difficult.
Dan
I swear I think you guys try to make things difficult.
Dan
Is this a test? Yes. I'd like to see the FR plot.
I'm guessing it looks like a dominant peak at a frequency in the neighborhood of 1.5 KHz, modulated by something about one-seventh of that frequency. I'm tempted to cheat and do some simulations to check my guesses and avoid exposing my ignorance, but on a Friday night I'll go for broke and place my bet.
Few
Just having a little fun. Pretty good guess. Here is the FR:
An externally hosted image should be here but it was not working when we last tested it.
What this is, and most likely what the driver problem is, is 3 high Q resonances closely spaced, as shown in the FR form the windowed driver impulse. The driver is radiating three closely spaced frequencies at elivated amplitrudes which give rise to beats. That is why the impulse appears to die out then rise again. It is not a reflection or diffraction, it is just the manifestation of beats due to the poorly damped, resonant maechanical system. The impulse reflects the beat frequency which the rise and fall is the envelope. The envelope decays over time because the resonances are (perhaps only slightly) damped.
Looking at an enlarged picture (100ms at 60dB resolution) I'd say your description of the sonic pattern form mere listening to the IR response was pretty spot on - no ?
Though we are looking at - and you were listening to - a loooooot of room interaction here too.
normalization not needed - hence a non-normalized plot..
As seen it takes quite some tome to decay to -40dB -60 dB in such a room - a looot of bouncing around until then....
Michael
Last edited:
In the US, I would recommend talking with McKenzie Acoustical Design.I was really scratching my head about this wiggle of the TD15M. Looked up half a dozen responses of 12" & 15" drivers at Brandon's Driver Measurements - drivervault and all had that same thing more or less.
I think it is a delayed damped back-bounce from the shockwave through the type of spider used for that sort of pro drivers, isn't it? Or is that too simple?
If it were stable (linear) and predictible in character pre-distortion of the driving current (per FIR-filter in the digital domain) should be able to correct for this, trying to terminate the reflection mechanically. Essentially the same strategy that digital room-equalization tries to use, but this time the dimension of the problem is 1, it's 1D and therefore much better suited to impulse response correction techniques, to the extend that the effects are not too non-linear (drifts included).
Ony may argue though if it's worth the effort, or favorable at all, besides the "digital stuff" involved. Well, maybe that delayed wiggle is part of the trademark sound of this class of drivers...
I'm pretty sure this company will be more than happy to sell miracle boxes developed by "innovative science" to any believers.
LOL
If you would like to do some investigation that proves otherwise - reread about CMP, only few pages back...
Saves you money by dampening too high expectations.
Michael
Last edited:
Ok - by reading some further I've seen he does *not* sell magic boxes (EQing stuff), so I was plain wrong - sorry .
Treating the diaphragn itself is an other thing all together!
Could you possibly point me there ?
Michael
Last edited: