I've just watched Erin's Audio Corner review and measurments of the TAD CR1-TX.
He shows a compression measurment. At the highest levels there is a drop in output above 10KHz, but not much a the xover region. Given that excursion will be most at the lower range, why do we see compression in the tweeters top octave, but not the lower? It must be something unrelated to excusion.
Talking about passive speakers here or those without limiters.
Is it coil heating, back emf, crossover part non linearity?
He shows a compression measurment. At the highest levels there is a drop in output above 10KHz, but not much a the xover region. Given that excursion will be most at the lower range, why do we see compression in the tweeters top octave, but not the lower? It must be something unrelated to excusion.
Talking about passive speakers here or those without limiters.
Is it coil heating, back emf, crossover part non linearity?
Last edited:
Hmm.. I'm not sure that is what I'm looking at. Here is an example of the compression I mean. The TAD video is not public yet so I'll choose another speaker. Whardfale Super Denton. Now obviously there is lots of different compression going on below ~4KHz (which the TAD has none of), I'm really only curious about the top octave.
The wharfedale super denton has a soft dome. Soft domes operate in a constant break-up behaviour which easily falls apart at higher sound pressure levels.
Maybe it's also an acceleration problem which should show up in the impulse response
Btw, the dip at 150Hz is the perfect example of why to avoid small woofers whenever possible.
Maybe it's also an acceleration problem which should show up in the impulse response
Btw, the dip at 150Hz is the perfect example of why to avoid small woofers whenever possible.
The TAD has a beryllium dome and shows the same, so really don't think it's related to breakup. Perhaps acceleration as you said.
Depends on the tweeter but normally they can roll off, especially in a waveguide, and the lower band is attenuated. If that's the case it actually "sees" less power mid band than the last octave. So with no attenuation you could see power compression first in the last octave.
Rob 🙂
Rob 🙂
Well, no (in this case). What the OP describes happens on passive speakers and not with just soft domes. For example, check what happens to the Magico S5 beryllium dome at the measuring of 70 to 95dB linearity (compression). About 3dB compression at 10kHz and this tweeter is crossing about 2.5kHz in this speaker if I remember correctly!
https://www.soundstagenetwork.com/i...&catid=77:loudspeaker-measurements&Itemid=153
Sorry I see only 1.5dB difference in compression between 70 and 95dB which could be expected from this low efficiency design! (86dB 4 Ohm 1meter)
Still too much for such a reference grade speaker flying by powered with $30.000 engines 🙂
Diamond coated beryllium tweeters , what comes next , moon dust coating on aerogel roasted by virginz in the twilight zone 🙂
Btw - 86kg speaker weight , your backbones will give you such a thrill!
I need another less dangerous hobby ...
Still too much for such a reference grade speaker flying by powered with $30.000 engines 🙂
Diamond coated beryllium tweeters , what comes next , moon dust coating on aerogel roasted by virginz in the twilight zone 🙂
Btw - 86kg speaker weight , your backbones will give you such a thrill!
I need another less dangerous hobby ...
It's 3dB.
But the question is why happens at higher frequencies more than at lower frequencies (despite its excursion is much more there).
1) mass corner (i.e., acceleration limit of a direct radiating driver above a certain break-point frequency (called the mass corner)
2) thermal heating - it takes only seconds to heat a direct-radiating driver voice coil to very high temperature
You'd have to know the specifics of the Klippel sweep length to understand how much thermal heating is occurring. It could be a fraction of a dB, or it could be much more.
(See [PDF] TD / Introducing Thermal Distortion | diyAudio)
3) any magnetic saturation in the dividing network - up high.
_________________________________________________________________
If you think about it...102 dB for a tweeter (only) is a heck of a lot of output, since most music follows a pink-noise-like spectrum for the ear to judge it to be "normal music", for instance, a large integrated portion of a track from a pop CD:
![post-26262-0-56580000-1422364014[1].gif](https://www.diyaudio.com/community/attachments/post-26262-0-56580000-1422364014-1-gif.1317852/ "post-26262-0-56580000-1422364014[1].gif")
Chris
2) thermal heating - it takes only seconds to heat a direct-radiating driver voice coil to very high temperature
You'd have to know the specifics of the Klippel sweep length to understand how much thermal heating is occurring. It could be a fraction of a dB, or it could be much more.
(See [PDF] TD / Introducing Thermal Distortion | diyAudio)
3) any magnetic saturation in the dividing network - up high.
_________________________________________________________________
If you think about it...102 dB for a tweeter (only) is a heck of a lot of output, since most music follows a pink-noise-like spectrum for the ear to judge it to be "normal music", for instance, a large integrated portion of a track from a pop CD:
Chris
Last edited:
Is there any spec that allows us to estimate the 'mass corner'? Mms and BL maybe? So we can predict this behaviour?
Am I right in thinking magnetic saturation will not occur for air-core inductors?
I've seen the same compression behaviour from active speakers and I do not nessisarily think it can be contributed to limiting circuits as they would likely limit the full tweeter or focus on the lower tweeter range to protect from excursion.
Am I right in thinking magnetic saturation will not occur for air-core inductors?
I've seen the same compression behaviour from active speakers and I do not nessisarily think it can be contributed to limiting circuits as they would likely limit the full tweeter or focus on the lower tweeter range to protect from excursion.
I'm sure you can develop a formula that can estimate this using T/S parameters of the driver. I do note that there are a lot of drivers that the manufacturer/vendor do not provide those values, i.e., they have to be measured by someone else.
Yes, essentially zero. With the price of copper going up over the past 10-20 years, I do see more manufacturers using iron cores in their chokes. YMMV.
Then it's probably (and predominantly) a driver issue instead of dividing network issue.
I am amazed at how little dynamic headroom that many manufacturers provide in their "small direct-radiating loudspeakers". It's clearly a choice, but I don't believe most consumers realize just how limited these loudspeakers/drivers really are in terms of headroom.
Chris
The only reason I could think of for only the top end being thermally compressed is the tweeter having a waveguide and a corresponding filter that reduces level for lower frequencies.
In the images I found there is some kind of conical shape, maybe that acts as a waveguide?
See Erin's measurements, these are instant compressions, the long term measurements usually don't show significant worsenings (over the instant compression). I would call (maybe wrong) that power compression instead of thermal compression.
Higher efficiency drivers compress less, check JBL or Klipsch speakers equippped with compression drivers for example.
Which Acoustic Music Instruments produce extremely large amplitude harmonics above 10kHz?
Please, can anybody list them?
Well, I guess if the sound system is trying to accurately reproduce the harmonics of acoustical music, and do that for an outdoor audience that are all 100 Meters from the loudspeakers, those tweeters need to be putting out very high level harmonics.
When setting up such a sound system, do not forget that the attenuation rate due to distance, is higher for 10kHz than it is for 100Hz.
Woodstock 7 anybody?
I think the tweeter compression issue does not apply to most inside-the-home Hi Fi Stereos.
What do we want the tweeter to do for Electronic Music?
Reproduce full sound levels of 90dB or more from an oscillator or square wave generator?
Or from a new fangled Moog Synthesizer?
That might be a possible reason to worry about the compression.
Lets call a square wave a valid synthesizer "music note" . . .
For a 6kHz Square wave fundamental that is reproduced at 90dB, then the third harmonic at 18kHz has to be 9.54 dB less, at 80.45 dB,
according to the mathematical definition of the harmonic levels of a square wave.
The fifth harmonic of a 6 kHz square wave is 30kHz, it is 13.98 dB less (76.02dB) versus the 6kHz 90 dB amplitude.
Most Humans do not hear 30kHz, some animals and bats might hear it.
I sleep at night, partly because I am not worried that my tweeters compress high sound levels above 10kHz . . . on "Music" CDs that I do not own.
Oh . . . Did I forget the CD Nyquist that only allows harmonics up to 22.05kHz?
Try peeling the outer layers off of the Onion first, before worrying about the inner layers of the onion.
Those inner layers are really packed tight (compressed?).
If you have read all the above, you are a better person than many others.
Question:
Given a specific drive level to a tweeter, 2.828 Vrms, what electrical parameters most effect the tweeter sound level?
At the tweeter's low frequency limit, is DCR the primary determinant?
At the tweeter's high frequency limit, is the voice coil Inductance the primary determinant?
Of course, there are other factors that affect the tweeter sound level, area, mass, shape, etc.
The above are just a compilation of my opinions, ideas, calculations,
and a Question about tweeter sound levels based on tweeter impedance.
Please, can anybody list them?
Well, I guess if the sound system is trying to accurately reproduce the harmonics of acoustical music, and do that for an outdoor audience that are all 100 Meters from the loudspeakers, those tweeters need to be putting out very high level harmonics.
When setting up such a sound system, do not forget that the attenuation rate due to distance, is higher for 10kHz than it is for 100Hz.
Woodstock 7 anybody?
I think the tweeter compression issue does not apply to most inside-the-home Hi Fi Stereos.
What do we want the tweeter to do for Electronic Music?
Reproduce full sound levels of 90dB or more from an oscillator or square wave generator?
Or from a new fangled Moog Synthesizer?
That might be a possible reason to worry about the compression.
Lets call a square wave a valid synthesizer "music note" . . .
For a 6kHz Square wave fundamental that is reproduced at 90dB, then the third harmonic at 18kHz has to be 9.54 dB less, at 80.45 dB,
according to the mathematical definition of the harmonic levels of a square wave.
The fifth harmonic of a 6 kHz square wave is 30kHz, it is 13.98 dB less (76.02dB) versus the 6kHz 90 dB amplitude.
Most Humans do not hear 30kHz, some animals and bats might hear it.
I sleep at night, partly because I am not worried that my tweeters compress high sound levels above 10kHz . . . on "Music" CDs that I do not own.
Oh . . . Did I forget the CD Nyquist that only allows harmonics up to 22.05kHz?
Try peeling the outer layers off of the Onion first, before worrying about the inner layers of the onion.
Those inner layers are really packed tight (compressed?).
If you have read all the above, you are a better person than many others.
Question:
Given a specific drive level to a tweeter, 2.828 Vrms, what electrical parameters most effect the tweeter sound level?
At the tweeter's low frequency limit, is DCR the primary determinant?
At the tweeter's high frequency limit, is the voice coil Inductance the primary determinant?
Of course, there are other factors that affect the tweeter sound level, area, mass, shape, etc.
The above are just a compilation of my opinions, ideas, calculations,
and a Question about tweeter sound levels based on tweeter impedance.
Last edited: