Do you like AMT tweeters?

[DavidMcRoy]

Agreed. Judged from subjective listening test results, AMTs are among the most transparent and best sounding tweeters I have ever encountered.

This comes at a price, though, as they are very difficult for a speaker designer to handle. Due to their unique combination of providing a near-perfect even wavefront while being rectangular, flat diaphragm area types of tweeters not allowing really low x-over frequencies compared to other tweeter types of similar size, you run into a bunch of problems with combining them with other drivers. Their ´natural´ directivity is always in a way not constant over the band, they are comparably chunky for their x-over freq, and controlling them with compression chamber, waveguide or other measures leads to a mess in most of cases.

When listening to numerous speaker concepts employing AMT tweeters, in most of cases I had the feeling something was not homogeneous, either in terms of directivity or subjective dynamics. The best concepts surprisingly were those combining an AMT with a d´Apollito style midrange dipole or line array. Something like that:

View attachment 454131

That's how I ended up doing it:

 

[Arindal]

That's how I ended up doing it:

Looks like a reasonable concept. Is this an open baffle design? And how low is the x-over freq of the AMT?

 

[Audionaut]

not allowing really low x-over frequencies

Well, there are very few dome tweeters that can be used from 1 kHz.

The efficiency of domes is much poorer, the radiating surface much smaller, which has advantages and disadvantages.

For example

Beyma TPL 200 > 1 kHz
https://audioxpress.com/article/tes...a-tpl200-h-pro-sound-air-velocity-transformer

Dayton AMT Pro 4 > 800 Hz

Dayton Audio - AMTPRO-4 Air Motion Transformer Tweeter 4 Ohm

AMTPRO-4 Air Motion Transformer Tweeter 4 Ohm

www.daytonaudio.com

In terms of measurements, AMTs are particularly strong in terms of impulse fidelity, distortion and bandwidth.

It is important that they are selected and measured in pairs and that the manufacturing company knows its way around and manufactures with precision.

To something completely different:
Only Purifi can market an aluminum dome tweeter as the latest hot shit. They will be extremely well paid for this and the fanbase on ASR will be beating the advertising drum for free.

Aluminum domes are available in very good quality from SEAS, SB Acoustics, Bliesma, Beyma, Scan Speak, etc.

An absolutely exhausted topic.

 

[Audionaut]

If you want to build speakers that hardly distort even at high volume levels, you need a large cone area in the bass range and very high efficiency in the treble range.

A dome tweeter very quickly reaches its limits here.
Especially with impulse peaks, which are almost never measured in the normal test course, but are very present in some Symphonie, techno and metal for example.

I prefer loudspeakers that measure absolutely cleanly even beyond 110 db at 1m and for which such requirements like normal living room conditions are something like a relaxing vacation.

That sounds so clean, precisely because it doesn't distort. I had the Beyma TPL 200 for several Years. Amazing Sound.
Yes, you can measure that too.

If you choose the size of the drivers sensibly and select the crossover frequencies so that they complement each other in terms of dispersion behavior, you can nowadays adjust the frequency response very nicely with DSP.

 

[Arindal]

Well, there are very few dome tweeters that can be used from 1 kHz.

Domes maybe not, but some horns.

I was more referring to the combination of potential x-over freq and minimum c2c distance when combining an existing AMT, dome or horn with a midrange. Domes which cannot go that low might make it easier to choose a compact midrange driver and moving the two closer to each other, compared to horn and AMT.

For example

Beyma TPL 200 > 1 kHz

Dayton AMT Pro 4 > 800 Hz

Both excellent tweeters without any doubt. But they both occupy 20 (or 23cm respectively) of height on the baffle, and the combination with a bigger midrange driver (or preferably midrange array/horn) might be worth considering as the horizontal directivity is also rather narrow due to diaphragm area, compared to any dome. How big will this whole thing become?

In terms of measurements, AMTs are particularly strong in terms of impulse fidelity, distortion and bandwidth.

Fully agree, and the majority of them sounds accordingly offering superb transparency and clarity. But it comes at the cost of uneven/unusual directivity and size occupied on the baffle. I understand speaker designers shying away from such excellent drivers.

Tbh I am not aware of many conventional, convincing speaker concepts employing such chunky AMT. An exception is the FinkTeam Kim:

Only Purifi can market an aluminum dome tweeter as the latest hot shit. ..

Aluminum domes are available in very good quality from SEAS, SB Acoustics, Bliesma, Beyma, Scan Speak, etc.

Fully agree. But which of the aforementioned domes does really sounds that amazing in terms of clarity and transparency? It is purely anecdotal, but whenever I think to myself ´wow, what an amazingly transparent treble, like an AMT´, it is either an AMT, a very well-designed planar magnetostat, or some exotic dome like Beryllium.

If you choose the size of the drivers sensibly and select the crossover frequencies so that they complement each other in terms of dispersion behavior, you can nowadays adjust the frequency response very nicely with DSP.

Unfortunately you cannot DSP uneven directivity and interference/lobing issues. Which is what most of unexperienced speaker designers would get from implementing an AMT into their traditional concepts.

 

[Audionaut]

Unfortunately you cannot DSP uneven directivity and interference/lobing issues. Which is what most of unexperienced speaker designers would get from implementing an AMT into their traditional concepts.

I agree.
I'm not a native English Speaker.
Is the last sentence of my post easier to understand this way:
If you choose the size of the drivers sensibly and select the crossover frequencies so that they complement each other in their dispersion behavior - you can then in the final step of customization adjust the frequency response very nicely with DSP nowadays.

 

[Arindal]

I'm not a native English Speaker.
Is the last sentence of my post easier to understand this way:

All good, I understood the meaning of your sentence. Just wanted to highlight the level of difficulty implied in the condition ´...complement each other in their dispersion behavior´.

 

[2020]

Due to their unique combination of providing a near-perfect even wavefront while being rectangular, flat diaphragm area types of tweeters

I've always wondered why they can't make square ones. Is it some problem with the manufacturing process? Instead of making them wider, they could always make them shorter to reduce height compared to width. It would still have a larger area even with a reduction and be comparable to a 0.75-1" dome tweeter, so I don't think I'm asking something radical.

When listening to numerous speaker concepts employing AMT tweeters, in most of cases I had the feeling something was not homogeneous, either in terms of directivity or subjective dynamics. The best concepts surprisingly were those combining an AMT with a d´Apollito style midrange dipole or line array. Something like that:

I wonder if that also has to do with traditional listening equipment not employing them. There are way more non-AMT speakers out there involved in the recording, mixing, and mastering processes. If AMTs were the default like soft domes and existed as the good enough/cheap/mature technology, people would be more used to their voicing and work around them.

 

[Arindal]

I've always wondered why they can't make square ones. Is it some problem with the manufacturing process?

My guess after having witnessed the manufacturing process ,would be it is technically doable. And some manufacturers produce such, like the Dayton Mini AMT:

Instead of making them wider, they could always make them shorter to reduce height compared to width.

Increasing width is not very desirable, as the aforementioned issues with uneven, increasing directivity would be much bigger in the horizontal plane with a broader diaphragm area, in the end of the day limiting the usable listening window in the upper treble band. That's what seemingly no manufacturer would really want. The Dayton Mini is an exception as the width is anyways small enough, in the region of 3/4" or 2cm. The tradeoff in this case is higher x-over freq and limited max SPL being particularly disadvantageous compared to a dome tweeter of similar size.

If AMTs were the default like soft domes and existed as the good enough/cheap/mature technology, people would be more used to their voicing and work around them.

Maybe yes, and they exist as mature, available technology, but I do not think the pro market works like this. In studios there are a lot of bigger active nearfield or midfield 2-way monitors, and people being in the market for such a model demand high SPL, low distortion and broad listening window. That is a combination which is problematic with an AMT-equipped 2-way design (although Adam Audio gained some success). To make things right, you need an additional small midrange driver and 3-way amp, it gets expensive and chunky.

The other reason, speculatively, might be directivity unevenness causing the speakers to sound differently in different rooms. Combining an AMT with a proper waveguide is very complicated, so most of designs just ´let it flow´ with a curved or recessed baffle causing a very broad dispersion pattern hence too much of brilliance energy in the room (typically the lowest octave the AMT is playing alone). In my understanding that is why many people say they sound ´overly bright´.

 

[DavidMcRoy]

Looks like a reasonable concept. Is this an open baffle design? And how low is the x-over freq of the AMT?

Yes, it's open baffle with DSP and an a dedicated amplifier channel for each driver. Crossovers are 40Hz to outboard subwoofers, 100Hz, 300Hz and 3,000Hz, all 24db/octave Bessel.

 

[Arindal]

Yes, it's open baffle with DSP and an a dedicated amplifier channel for each driver. Crossovers are 40Hz to outboard subwoofers, 100Hz, 300Hz and 3,000Hz, all 24db/octave Bessel.

Very promising and coherent concept, congrats! The only aspect I would pay closer attention to is the 3,000Hz x-over point which looks on paper a bit high and might be contributing to lobing and cancellation issues between the two small cones.

 

[Frunse]

Hello, here you can find my aktualy AMT Wavguide project, i like them too a lot Use the Harwood AM25 since probably 2015 or when ever they released it first , not sure.

https://www.diyaudio.com/community/...aveguides-for-cnc-3d-printing.318190/page-127

 

[Audionaut]

here you can find my aktualy AMT Wavguide project

Interesting

Correct Link

Open source Waveguides for CNC & 3D printing!

This is Satori 6 inch scaled uniformly to 1.18 to fit 2 inch broadband driver. Horizontal directivity of 2 inch driver raw and then quick cross with 8 inch but no shelf yet. Might be worth pursing but not sure yet. will probably add the shelf or notch and listen again.

www.diyaudio.com

 

[gamerpaddy]

i found AMT to be less fatigueing to my ears especially at the volumes im listening to.

i tried many speakers in the past, dome or horn, one even with a servo dome tweeter (backes & müller)... and some ring-radiator.
they all had the same issue to me, they start screaming at certain tones.
havent had that issue once with an AMT

right now im using some old ELAC BS244 after trying out the KEF R3 (which i sent back immediately due to the same issue),
i just couldnt find a single speaker that came close to what im getting with the ELAC yet

 

[Pearljam5000]

How good are the Adam AMT tweeters?

 

[Arindal]

they all had the same issue to me, they start screaming at certain tones.

Know exactly what you mean. I found a number of flat, light and stiff 1" domes to offer transparency and clarity on the level of the best AMTs available, but at the price of much broader dispersion, so it is not really comparable and might result in problematic interaction with the room.

Experience is hinting that the tendency to scream can only be successfully avoided if you combine such dome with a midrange driver of very similar behavior in terms of subjective dynamics and directivity. Example by Scanspeak:

How good are the Adam AMT tweeters?

The drivers itself can be excellent, but how it translates to overall sound quality, is in my understanding vastly dependent on x-over freq, baffle geometry, waveguide and combination with a midrange driver. Particularly models with a waveguide cannot be judged from afar or solely based on measurements, you have to really give them a listen.

 

[waofner]

How good are the Adam AMT tweeters? View attachment 454868

for me nothing compares to AMT tweeters. I am runnng Adam S3v at main listening place and Adam A7V at my desk.
integrated DSP is very helpful( only needed below 400Hz) Tweeter history: Mivoc..Viva..Wharfedale..Briston..Morel..Seas Millenium...etc, do not know about Beryllium SB Acoustics

Graph: REW Adam S3V listening place; average of 6 measurements.

 

[Audionaut]

How good are the Adam AMT tweeters?

Adam A5X Review (Powered Studio Monitor)

This is a review and detailed measurements of the Adam A5X powered speaker. It was kindly sent to me by a member for review and costs US $499 each. Despite its small size, the A5X is quite heavy: The back of it got warm so I am assuming it uses class AB amplification or at least a linear...

www.audiosciencereview.com

Adam T8V Studio Monitor Review

This is a review and detailed measurements of the Adam T8V active monitors (speaker). It was kindly sent to me by the company a few months ago. I was surprised at the very low cost of these speakers at US $299 (each) considering that they are DSP speakers with dual amplification. This is one...

www.audiosciencereview.com

Adam S2V Studio Monitor Review

This is a review and detailed measurements of the Adam S2V Professional Studio Monitor (powered speaker). It is on kind loan from a member. The S2V costs US $1,750 (each, NOT a pair). This is one beefy, heavy and deep bookshelf speaker: Yes, the cone was a bit dusty. :) The back panel...

www.audiosciencereview.com

 

[DavidMcRoy]

Very promising and coherent concept, congrats! The only aspect I would pay closer attention to is the 3,000Hz x-over point which looks on paper a bit high and might be contributing to lobing and cancellation issues between the two small cones.

Indeed, there is a little, but I prefer some limited HF dispersion to the floor and ceiling anyway. This isn't meant to be a wide-dispersion design. I worked in audio for broadcast TV control rooms for several decades using fairly directional speakers, Tannoy, etc. and it's my preference. I tried crossing over to the pair of 6" mids at 1.5, 2, and 2.5kHz and prefer 3kHz. In principle, I like using a single pair of identical drivers to cover most of the voice region from 300Hz to 3kHz.

 

[Arindal]

there is a little, but I prefer some limited HF dispersion to the floor and ceiling anyway.

Both attenuating floor/ceiling reflections and keeping directivity index high are in general excellent ideas for home listening. But I found linearly attenuated reflections and reverb to be the most helpful. D´Apollito-style arrangements with a rather high x-over freq tend to show a medium-width dip caused by interference/lobing.

I tried crossing over to the pair of 6" mids at 1.5, 2, and 2.5kHz and prefer 3kHz.

You mean the AMT allows 1.5k of lowest x-over point? May I ask what was the issue with this one choice?

In terms of vertical lobing, this from a theoretical point is surely the best way to go, and with two midrange drivers I would not overly worry regarding localization issues.

It often leads to a problem with directivity index, though, as most of broad-diaphragm, medium-height AMTs tend to have a steeply decreasing directivity index below 2.5k towards lower frequencies (as width and height get increasingly small compared to wavelengths). So crossing around 1.5k brings a lot of energy into the room between 1.5k and 2.5k making it sound very direct, present, bringing overly bright formants and shiny vowels, if that makes sense. This has to be addressed somehow, either with a waveguide, some directivity control via midrange driver, a slight additional on-axis dip, or whatever measure.