Do you like AMT tweeters?

[Lux_ury]

I love my ribbons only thing if your source is bad it will pick up the imperfections and rub your face into it.

 

[DavidMcRoy]

The handful of AMT tweeters and mid-tweeters I’ve experimented with mostly have very low distortion, which I tend to associate with a “lack of bite,” as it were, if you’re not used to it. Their sound seems on par with electrostatics and planar magnetics in that characteristic. If they don’t “excite enough,” just crank up the EQ a little to taste.

 

[Ian Knight]

the ml's i have heard tend to have elevated f.r. ..... the emotiva b1 mods i have had the treble flattened by Dennis Murphy , they are in no way harsh on good recordings (although sibilant recordings can get "screechy"), i think frq. response is key here not the type of tweeter...

It is the implementation and voicing of the crossover then in your opinion ?. The curveball for me was how different they sounded in the high frequencies (on the same tracks from Qobuz ) in my room than when I heard them in the shop. Admittedly the hifi Rose amp and streamer was used instead of my NAD C368 and DRC bluos .....the levels achieved in the shop in a much larger compared to my smaller room were a lot higher without apparent problems in the hf and mid that I am getting here.

 

[JohnnyAudio]

I have 6 of them as well, and you might want to look into stacking 3 on each side instead of one at a time

0726240757_HDR by drjlo2, on Flickr

Wow an AMT line array!

 

[tw 2022]

It is the implementation and voicing of the crossover then in your opinion ?. The curveball for me was how different they sounded in the high frequencies (on the same tracks from Qobuz ) in my room than when I heard them in the shop. Admittedly the hifi Rose amp and streamer was used instead of my NAD C368 and DRC bluos .....the levels achieved in the shop in a much larger compared to my smaller room were a lot higher without apparent problems in the hf and mid that I am getting here.

that would be my best guess...that elevated treble can make many tweeters sound harsh at higher volume from my understanding..my best understanding is that "v" shaped f.r. will give extra low volume "oomph" and dynamics while a proper sloping speaker may lack a bit at low(er) volume...this is all from my experience and reading for research...

 

[ThatSoundsGood]

that would be my best guess...that elevated treble can make many tweeters sound harsh at higher volume from my understanding..my best understanding is that "v" shaped f.r. will give extra low volume "oomph" and dynamics while a proper sloping speaker may lack a bit at low(er) volume...this is all from my experience and reading for research...

It's certainly the implementation of the tweeter that makes the most difference. If the frequency response is not flat (or close to it) then that's the easiest difference to hear. But there are other differences; AMT's normally have different vertical and horizontal dispersion characteristics which make them interact with the room differently than a dome tweeter. Secondly, distortion can be low or high on either of them depending on the model. Sometimes a well designed crossover/design sounds worse since it reveals the problems in a recording so you might think a song sounds better on a given speaker but that speaker is actually designed worse and is masking the problems in the recording. A great quote I heard from a friend when he bought Genelecs was "Oh, those speakers are amazing, everything sounds bad on them." Accurate speakers do not always sound better but they do sound more accurate. The problematic details that you hear might not be the speaker.

 

[JaapD]

Here is what ‘my’ ears think of it:
- Soft-domes: a bit dull to my taste.
- Titanium domes: more detailed, appreciated them before moving over to AMT.
- AMT: very transparent extremely detailed, listening to them over the last 5 years.
I’m hugely generalizing here so take it with a grain of salt, okay?

I very much appreciate AMT’s in my 3-way design (4th order active cross-over from 2800Hz). I’ve never experienced the sound of for instance cymbals and breaking glass so realistic, in comparison to the more regular dome tweeters. This might be related to its design aspects, among others the very low moving mass, fast impulse response, and high bandwidth.

Cheers,
JaapD.

 

[JohnnyAudio]

Here is what ‘my’ ears think of it:
- Soft-domes: a bit dull to my taste.
- Titanium domes: more detailed, appreciated them before moving over to AMT.
- AMT: very transparent extremely detailed, listening to them over the last 5 years.
I’m hugely generalizing here so take it with a grain of salt, okay?

I very much appreciate AMT’s in my 3-way design (4th order active cross-over from 2800Hz). I’ve never experienced the sound of for instance cymbals and breaking glass so realistic, in comparison to the more regular dome tweeters. This might be related to its design aspects, among others the very low moving mass, fast impulse response, and high bandwidth.

Cheers,
JaapD.

" cross-over from 2800Hz " I think this is perfect for AMT's

 

[Chromatischism]

Here is what ‘my’ ears think of it:
- Soft-domes: a bit dull to my taste.
- Titanium domes: more detailed, appreciated them before moving over to AMT.
- AMT: very transparent extremely detailed, listening to them over the last 5 years.
I’m hugely generalizing here so take it with a grain of salt, okay?

I very much appreciate AMT’s in my 3-way design (4th order active cross-over from 2800Hz). I’ve never experienced the sound of for instance cymbals and breaking glass so realistic, in comparison to the more regular dome tweeters. This might be related to its design aspects, among others the very low moving mass, fast impulse response, and high bandwidth.

I don't think there's an appreciable difference in sound between a square AMT and a soft dome tweeter measuring the same frequency response. I welcome any evidence, though.

 

[JaapD]

I don't think there's an appreciable difference in sound between a square AMT and a soft dome tweeter measuring the same frequency response. I welcome any evidence, though.

The frequency response is only part of the story. We have the frequency domain and also the time domain. The last one telling us more about the impulse response.

Think for instance a tweeter with a relatively heavy cone and another one with a very small mass, both capable of delivering 20 KHz in the frequency domain. The time domain and their impulse response may be very different though (relatively long for the heavy cone, short for the other one).

Similar example: put a 20 KHz tone burst on both tweeters. After the tone burst stops the one with the heavy mass will slowly play quieter (exaggerating a bit, to clarify things) while the other one immediately stops playing.

In order not to make things too complicated I'm not mentioning damping factors here.


Apart from this: My AMT delivers up to 50KHz. Many soft dome tweeters already have their -3dB point at 20KHz, so 3dB down => 45 degrees phase shift, no flat group delay.


Cheers,
JaapD.

 

[tw 2022]

AMT's normally have different vertical and horizontal dispersion characteristics which make them interact with the room differently than a dome tweeter.

yes , i notice the vertical dispersion is limited...i think mostly to combat floor/ ceiling bounce...

 

[tw 2022]

Accurate speakers do not always sound better but they do sound more accurate

my little b1's prove this point... the Dennis Murphy mod cleaned the mids/ treble up to the point of almost too much clarity ...

 

[ThatSoundsGood]

yes , i notice the vertical dispersion is limited...i think mostly to combat floor/ ceiling bounce...

Precisely that. The taller the AMT then the more limited the vertical dispersion (generally). Which I really like with mine.

 

[ThatSoundsGood]

The frequency response is only part of the story. We have the frequency domain and also the time domain. The last one telling us more about the impulse response.

Think for instance a tweeter with a relatively heavy cone and another one with a very small mass, both capable of delivering 20 KHz in the frequency domain. The time domain and their impulse response may be very different though (relatively long for the heavy cone, short for the other one).

Similar example: put a 20 KHz tone burst on both tweeters. After the tone burst stops the one with the heavy mass will slowly play quieter (exaggerating a bit, to clarify things) while the other one immediately stops playing.

In order not to make things too complicated I'm not mentioning damping factors here.


Apart from this: My AMT delivers up to 50KHz. Many soft dome tweeters already have their -3dB point at 20KHz, so 3dB down => 45 degrees phase shift, no flat group delay.


Cheers,
JaapD.

This plus distortion and ultimately the on and off-axis response of the crossover. We spend so much time with on-axis measurements but off-axis is really important here. An AMT can be limited vertically which means that the crossover has to work with this. It also has to pair well with the mid range driver and be crossed over in a way that works both horizontally and vertically. I personally, wouldn't worry much about what happens above 15KHz since most people don't hear that and there is very limited information up there.

 

[JohnnyAudio]

This plus distortion and ultimately the on and off-axis response of the crossover. We spend so much time with on-axis measurements but off-axis is really important here. An AMT can be limited vertically which means that the crossover has to work with this. It also has to pair well with the mid range driver and be crossed over in a way that works both horizontally and vertically. I personally, wouldn't worry much about what happens above 15KHz since most people don't hear that and there is very limited information up there.

We may need a few of these to improve off axis...

 

[Chromatischism]

The frequency response is only part of the story. We have the frequency domain and also the time domain. The last one telling us more about the impulse response.

Impulse is used to calculate the frequency response.

https://en.m.wikipedia.org/wiki/Fourier_transform#:~:text=In%20physics%2C%20engineering%20and%20mathematics,present%20in%20the%20original%20function.

 

[JaapD]

Impulse is used to calculate the frequency response.

https://en.m.wikipedia.org/wiki/Fourier_transform#:~:text=In%20physics%2C%20engineering%20and%20mathematics,present%20in%20the%20original%20function.

Indeed. And vice versa through the inverse Fourier transformation.

But do take both amplitude (by you mentioned as 'frequency response') and phase response into account here => complex Fourier transformation.

Last but not least, the usual frequency response plot never describes inrush or settling aspects, it is missing any dynamic behavior so to speak. This is simply because the applied sinewave sweep is a relatively ‘easy’ signal, not containing any dynamic related contents (just a constant voltage and a slowly changing frequency).


Cheers,
JaapD.

 

[ThatSoundsGood]

We may need a few of these to improve off axis...
View attachment 384855

I don't want my speakers to eat me. Those are somehow majestic and frightening all at the same time.

 

[Chromatischism]

Indeed. And vice versa through the inverse Fourier transformation.

But do take both amplitude (by you mentioned as 'frequency response') and phase response into account here => complex Fourier transformation.

Last but not least, the usual frequency response plot never describes inrush or settling aspects, it is missing any dynamic behavior so to speak. This is simply because the applied sinewave sweep is a relatively ‘easy’ signal, not containing any dynamic related contents (just a constant voltage and a slowly changing frequency).

If you subscribe to Dr. Floyd Toole's research, what you hear is represented by the frequency response. Being calculated from the impulse, there's nothing "left out". Even EQ systems such as Dirac and Audyssey filter the impulse. That also includes things like decay and phase. When those change, so too does the frequency response.

 

[JaapD]

If you subscribe to Dr. Floyd Toole's research, what you hear is represented by the frequency response. Being calculated from the impulse, there's nothing "left out". Even EQ systems such as Dirac and Audyssey filter the impulse. That also includes things like decay and phase. When those change, so too does the frequency response.

Thanks for your reply. From the frequency plots that I’m aware of (created through a frequency sweep with constant amplitude) I can’t read its dynamic behavior, i.e. identify if its a tightly controlled system or not, low or high Q factor, amplitude- and phase margin, phase shift, group delay, etc.

Off topic: some (but not all !) amplifiers specify its frequency range, i.e. 20Hz-20KHz. If you read closely then the measurement has been carried out with a relatively small signal amplitude. The amplifier could very well be slew rate limited and not perform same bandwidth at full power. Why I’m mentioning this? As another example that to me just a frequency plot does not tell the full story. I have no problem what so ever if you think otherwise though.

Cheers,
JaapD.