Out of your list I have had the ESS Heil AMT, the Dayton AMT4pro and the Beyma TPL150 and believe me, they are all very good tweeters but just like Hans555 said, in the end none of them blend in very well in an open baffle design because of their excessive vertical beaming. Their vertical response is as narrow as 20 degrees or less. Your milage me vary but it is something I just could not live with, it screws up the in room power response and you will always hear your mids and highs as two seperate entities because of their totally different radiation patteren. I do like the smaller AMT's a lot though but those can't be used under 3kHz or so.
Were any of the Mundorf AMT Dipole considered? https://www.mundorf.com/audio/en/shop/Tweeters/hifiAMT_Dipole/
I already have a Mundorf 25D1.1 and I really like it. 🙂
I just don't see for me any possibility to integrate it directly with a Purifi driver, without using an additional driver in between.
The reason has been mentioned before. The Purifi drivers ( even the 4 inch version) will most probably not have a similar radiation pattern on front and backside higher than 1.x kHz due to the big magnets, which work like an HP filter for the backside.
So if you want to achieve an overall similar radiation pattern for front and backside without using an additional (Non-Purifi) driver, you have to crossover lower than the frequency, where front and backside diverge.
There is only one Mundorf Dipole-AMT driver, which can be crossed over as low as 1.8 kHz, but due to its vertical size, it will introduce the same problems with vertical beaming, like the other huge drivers, e.g. Heil or Beyma.
So, as already mentioned by @Sjef and @Hans555, you will get new problems, when trying to solve another one.
So maybe the only working (near perfect) solution at the end will be a pair of compression drivers in a CD horn.
In my current system, a Purifi driver could fill the gap between my Acoustic Elegance Dipole 18 and the Scanspeak 12MU, somewhere between 200Hz (or lower) and 1000 Hz.
Dipole 18 - Purifi PTT6.5 or 8 - Scanspeak 12MU - Mundorf 25D1.1 …. … not to bad as well 😊
You're probably right, but then the Purifi PTT8 seems to be a little like cracking nuts with a sledghammer for me, especially considering its price tag.
Maybe, another nice 8 inch driver would be more suitable (=Value for money, related to the purpose) for this frequency range, e.g. the JA8008 HMQ or the Dayton PM220-8.
Maybe, another nice 8 inch driver would be more suitable (=Value for money, related to the purpose) for this frequency range, e.g. the JA8008 HMQ or the Dayton PM220-8.
@Hans555 Care to share pictures of your speaker with front+rear CD+Horns??
@Matty Sorry that I missed what had already been said before. Have you considered magnet-mounting front+rear driver like Kyron Gaia speakers?
And a noob question. What makes these drivers less equal in front and rear radiation? Small-ish driver with large structure/magnets in rear or something else?
@Matty Sorry that I missed what had already been said before. Have you considered magnet-mounting front+rear driver like Kyron Gaia speakers?
And a noob question. What makes these drivers less equal in front and rear radiation? Small-ish driver with large structure/magnets in rear or something else?
@emilkoz
I have thought about concepts like in the Gaia, but I don't know how this would work in the midrange.
Of cause, you can take 2 equal drivers and use one of them with the back facing forward and you will solve the problem of different frequency responses.
Question is now, how to position both drivers.
I guess the concept in the Kyrons, where the drivers are aligned horizontally works, because the distance of the drivers is still small compared with the wavelengths. Otherwise the addition of both drivers would not work as expected, due to the phase shift. In the midrange this would be a different story. Here, I guess the only reasonable possibility would be to align them vertically. Unfortunately this would influence the vertical directivity, because the overall sound source would be twice as long on the vertical axis, which will make them beam vertically to the higher frequencies. If this fits with your concept, its fine.
Related to your last question:
Its from my point of view e.g. the magnet, the spider, the basket.
Everything can influence the frequency response, as this components impede the sound radiation, and they typically do that more at the higher frequencies.
Probably also the cone shape has influence on radiation pattern differences between front and back.
But may be here are some experts, who have more experience with that than I have.
I have thought about concepts like in the Gaia, but I don't know how this would work in the midrange.
Of cause, you can take 2 equal drivers and use one of them with the back facing forward and you will solve the problem of different frequency responses.
Question is now, how to position both drivers.
I guess the concept in the Kyrons, where the drivers are aligned horizontally works, because the distance of the drivers is still small compared with the wavelengths. Otherwise the addition of both drivers would not work as expected, due to the phase shift. In the midrange this would be a different story. Here, I guess the only reasonable possibility would be to align them vertically. Unfortunately this would influence the vertical directivity, because the overall sound source would be twice as long on the vertical axis, which will make them beam vertically to the higher frequencies. If this fits with your concept, its fine.
Related to your last question:
Its from my point of view e.g. the magnet, the spider, the basket.
Everything can influence the frequency response, as this components impede the sound radiation, and they typically do that more at the higher frequencies.
Probably also the cone shape has influence on radiation pattern differences between front and back.
But may be here are some experts, who have more experience with that than I have.
Out of curiosity, how did you end up using your AMTs (assuming you are)? The vertical radiation mismatch will be there with almost any other solution tpo, so what would you think would make the case in an OB application worse?