So DDR I directly relatable to transient response, impulse response etc as X also stated.
Unfortunately in a majority of cases, clean impulse response and lack of resonances in cone structures don't occur in undamped materials, and is also directly related to smoothness of response.
Unless a polystyrene coned fullrange driver seems like a good idea?
So the two factors are, as usual in hifi and most things, in opposition, and thus middle ground has to be found.
Eg as Norman agreed with ribbons....lively settling time, when OFF a resonance. Poor transient response on resonances, which are multiple. I have some ribbons and they excel at realism, but only until they hit a peak in their response.
So what we seek is a driver with a very smooth response and light cone. Two attributes that are very much at odds with each other.
Unfortunately in a majority of cases, clean impulse response and lack of resonances in cone structures don't occur in undamped materials, and is also directly related to smoothness of response.
Unless a polystyrene coned fullrange driver seems like a good idea?
So the two factors are, as usual in hifi and most things, in opposition, and thus middle ground has to be found.
Eg as Norman agreed with ribbons....lively settling time, when OFF a resonance. Poor transient response on resonances, which are multiple. I have some ribbons and they excel at realism, but only until they hit a peak in their response.
So what we seek is a driver with a very smooth response and light cone. Two attributes that are very much at odds with each other.
Last edited:
Ribbons have been mentioned but not thin folded Mylar membrane AMT's like the Heil ESS AMT. My AMT's have a flat resonance free response above 700Hz all the way 20kHz (impedance curve is flat like a resistor) and there is no ringing, yet it sounds very much more detailed than any cone or dome tweeter. The AMT diaphragm is indeed very light and resonance free - could this be the ideal combo then for a driver with excellent low level detail and resolution?
I think that its not so much AMT diaphragm weight, as a dome is also very light, and neither dome or AMT come close to a ribbon for lightness.
What I think is special about some AMTs is the acoustic transformer effect, much like a decent horn can do the same for CDs.
And I like my ribbons since they are waveguide loaded used above 3.5k and the excel, for budget devices. I'm still aching to buy some goood AMTs
In short driver efficiency has a lot to do with, (advance apology for subjectivism...) effortless quality. Cone drivers alone are naff in that regard, even 'efficient' ones, unless theyre assisted....(I.e. Acoustically efficient radiators, rather than the usual electrical efficiency which most people value more)
What I think is special about some AMTs is the acoustic transformer effect, much like a decent horn can do the same for CDs.
And I like my ribbons since they are waveguide loaded
used above 3.5k and the excel, for budget devices. I'm still aching to buy some goood AMTs In short driver efficiency has a lot to do with, (advance apology for subjectivism...) effortless quality. Cone drivers alone are naff in that regard, even 'efficient' ones, unless theyre assisted....(I.e. Acoustically efficient radiators, rather than the usual electrical efficiency which most people value more)
Last edited:
How about play pink noise at a fixed (high) SPL, and then take a frequency/impulse response at a lower level, while the pink noise is still playing?
Of course, you'd have to dig out the low-level signals, but with today's computing power, it looks possible.
Then you'd be able to compare the low-level responses with/without louder pink noise, and get an idea of what the driver will be doing to tiny details in music.
Chris
Of course, you'd have to dig out the low-level signals, but with today's computing power, it looks possible.
Then you'd be able to compare the low-level responses with/without louder pink noise, and get an idea of what the driver will be doing to tiny details in music.
Chris
I think that its not so much AMT diaphragm weight, as a dome is also very light, and neither dome or AMT come close to a ribbon for lightness.
What I think is special about some AMTs is the acoustic transformer effect, much like a decent horn can do the same for CDs.
And I like my ribbons since they are waveguide loaded
In short driver efficiency has a lot to do with, (advance apology for subjectivism...) effortless quality. Cone drivers alone are naff in that regard, even 'efficient' ones, unless theyre assisted....(I.e. Acoustically efficient radiators, rather than the usual electrical efficiency which most people value more)
Is the mass of a ribbon tweeter (usually metal ribbon?) really less than that of the AMT? If you consider just the mass of one fold V channel which has to move to squeeze the air out, it is indeed very small. The fold depth is circa 1/4in deep x 4.75in high x 0.0005in thick of Mylar material (not metal). That is not much mass.
Last edited:
Yeah, I was thinking the same thing. Pink noise with a low level square wave, or something like that. Run the test signal below the pink by 30db or whatever.
As has been pointed out, this is not exactly a new issue, though the term is new. There may be some existing tests and analysis protocol.
Ribbons are typically aluminum foil, very thin. I'm sure you could find the thickness somewhere, and calculate the mass. I'm pretty sure it is less than AMT, but both are low. Somehow I haven't heard AMT, despite knowing people that had them. I've had a couple different ribbon tweeters.
Ribbons are weird. Superior detail in the top octave compared to anything else I've heard. The lower treble is OK if you can get it to behave, but somewhat prone to over-exciting into some distortion/resonance. This isn't that surprising considering the 'diaphragm' is basically a gossamer thin strip of foil swinging in the wind. I never really got my big ribbons to work until I gave up on high efficiency applications for them. Padded down 10db and crossed over with a series type crossover, they are excellent. In my big rig set ups, a $10 piezo horn blew them away.
BTW, x, I have to disagree with you regarding the idea of detail resolution being the result of ringing. I think that it is quite the opposite: ringing masks detail.
Certainly in conventional cone drivers, I have found that there is a correlation between overdamped low Qts drivers and high detail resolution.
At the same time, I agree with P10 in that the more highly damped cones tend to have less detail resolution. This is as far as I know why companies started moving away from polypropylene cones back in the 90s. They made pretty graphs, but at a cost.
GregB,
My suspicion of enhanced audibility of detail came about primarily because of the rave reviews of the aluminum cone A7.3. Have you looked at what its IR looks like? The first ring peak is about 60% of FS and it is a heavy ring - at about 10kHz. This frequency specific ring - for some reason makes quick sharp low level percussions sound more audible - but not smeared beyond recognition. I think it gives it emphasis. Because if we say that ringing drivers don't have detail then the A7.3 would fare badly - which, based on reports of others, is just the opposite. I think it can sound good with this type of enhanced audibility.
My suspicion of enhanced audibility of detail came about primarily because of the rave reviews of the aluminum cone A7.3. Have you looked at what its IR looks like? The first ring peak is about 60% of FS and it is a heavy ring - at about 10kHz. This frequency specific ring - for some reason makes quick sharp low level percussions sound more audible - but not smeared beyond recognition. I think it gives it emphasis. Because if we say that ringing drivers don't have detail then the A7.3 would fare badly - which, based on reports of others, is just the opposite. I think it can sound good with this type of enhanced audibility.
x, I think that's a valid argument and I while I agree that peaks in response at certain frequencies might enhance apparent detail audibility, my conjecture is that there is both real and apparent resolution, and that these are not the same. It would seem that this would be pretty easy to test with digital eq. If you eq out that peak, and the detail remains, that it would indicate that some other mechanism is at work.
I don't have the 7.3, but I do have the 10.3.
I don't have the 7.3, but I do have the 10.3.
XRK: I didn't think AMTs used 0.5 thou mylar, perhaps 5 thou? And yes i do believe some ribbons are maybe an order of magnitude lighter, in TOTAL diaphragm mass.
But that is my whole point, the force per unit mass for an AMT is far greater than traditional cone or dome drivers., and that is its main advantage over them.
EDIT: Air displacement relative to size is far greater, as part of their function.
I think its a similar case with small ribbons, although i wouldn't like to say if it is to the same degree or not, but either way both ribbon and AMT usually have a larger diaphragm area which contributes to greater acoustic efficiency. (for a given deflection)
My main reservation with AMT is having to use dipole, or use a back chamber and the associated resonances. A TL loaded AMT would be interesting though
But that is my whole point, the force per unit mass for an AMT is far greater than traditional cone or dome drivers., and that is its main advantage over them.
EDIT: Air displacement relative to size is far greater, as part of their function.
I think its a similar case with small ribbons, although i wouldn't like to say if it is to the same degree or not, but either way both ribbon and AMT usually have a larger diaphragm area which contributes to greater acoustic efficiency. (for a given deflection)
My main reservation with AMT is having to use dipole, or use a back chamber and the associated resonances. A TL loaded AMT would be interesting though
Last edited:
I agree. This specific driver acts as a hardware EQ. It migth sound good to some but one would have to ask why the mixing engineer didn't add that zing to the recording in the first place.
Last edited:
I can only speculate.
- He is young and still hears high frequencies quite well
- He didn't like it
- He didn't care
- He was listening in a control room with low reverberation time which didn't require such a boost because there was less auditory masking from room reverberation
Last edited:
If you guys are trying to convince me that a 5db (or whatever) rise at 10k is why I can hear footsteps and other small details with some drivers but not others, I'm still not buying it.
Look at the polar response of that particular driver. The peak can be seen at nearly all angles.
You can simulate this effect with an EQ. Try very narrow peaks at different frequencies. A lot of detail can be reveal. This depends on the recording though. That's why having it applied to each and every recording is an ill-conceived idea in my opinion.
Sibilance by any other name [emphasis, definition, pierce, etc.] is still sibilance, which some folks like and others don't [me and the friends who use 'FR' drivers], so while it may sound like more/better detail to some, it just sounds to others like the irritating distortion that it really is.
For folks like me then, the great bang/buck EL70, FRS125S with their so-called good 'detail' ultimately lost out to tweaked RS 40-1354 drivers by the several folks who tried them out in the same MLTL alignment for several months each.
In retrospect, I'm assuming it's more about whether one has tinnitus or not as to what we prefer.
GM
That sort of thing just sounds like sibilance/distortion/noise to me also.
Personally i can't stand any sort of HF peaking drivers, and will always attempt to damp it out to as flat as possible with physical tweaks, using EQ or passive filters for the remainder as needed.
pnix, i've done plenty of experimenting with EQ boost and cuts in the past, especially while mixing, recording, etc. I get your point, but this isn't what I'm hearing when I hear a very clean speaker system with good detail resolution. Actually 10k is a bit past the sibilance range - usually considered to be 8kish, but it's close enough to sound about the same.
Personally i can't stand any sort of HF peaking drivers, and will always attempt to damp it out to as flat as possible with physical tweaks, using EQ or passive filters for the remainder as needed.
pnix, i've done plenty of experimenting with EQ boost and cuts in the past, especially while mixing, recording, etc. I get your point, but this isn't what I'm hearing when I hear a very clean speaker system with good detail resolution. Actually 10k is a bit past the sibilance range - usually considered to be 8kish, but it's close enough to sound about the same.
The MA driver has a 6kHz peak ±15 deg around the 0 deg axis.
In my experience the best high frequency detail/transparency comes from low reverberation/ringing at low frequencies. It also makes a lot of sense when one looks at how auditory masking works.
Auditory masking - Wikipedia, the free encyclopedia