I have had a request to make a slightly nicer version of this horn. I have a number of questions that I've tried to get answers to on FB, but I haven't gotten what I'm specifically looking for. I know there are many issues around time arrival, so you don't need to comment on that. My very specific question here is.
What happens to the high frequencies in a horn that is twisted, i.e. a Sato/WE horn?
My own thought is that the curvature will act as a low pass filter (HF wants to go in a straight line) so that the very highest frequencies do not come out of this midrange horn.
The system
Specifically, it is a Vitavox GP1 driver that sits in the horn small Sato horn. The horn is 1,6 m long and the mouth is 58x58 cm. The driver should naturally roll off at the top around 12-14K Hz. Then a super tweeter takes over. But will it play up in that frequency range (out off the horn) at all because of the curvature, that's what I want to hear about. And I've never heard the horn myself or the system must simply make the horn but can't help but try to understand the horn.
What happens to the high frequencies in a horn that is twisted, i.e. a Sato/WE horn?
My own thought is that the curvature will act as a low pass filter (HF wants to go in a straight line) so that the very highest frequencies do not come out of this midrange horn.
The system
Specifically, it is a Vitavox GP1 driver that sits in the horn small Sato horn. The horn is 1,6 m long and the mouth is 58x58 cm. The driver should naturally roll off at the top around 12-14K Hz. Then a super tweeter takes over. But will it play up in that frequency range (out off the horn) at all because of the curvature, that's what I want to hear about. And I've never heard the horn myself or the system must simply make the horn but can't help but try to understand the horn.
Attachments
I dont think you get a high pass affect, but rather an odd vertical polar pattern if you were to use the horn image above. the high frequencies sort of follow the wall which obscures the straight path, but the larger wavelengths are less affected as they can diffract around the wall the bends away from the straight path effectively so you get a kind of frequency separation, with the high frequencies firing more downward than the lows. Since a majority of the power response at high frequencies is projected not directly at the listener, you are relying on room reflections off the floor, which may or may not have significant absorption mechanisms.
Last edited:
I wish my English was better so I could understand it better. In addition, I must admit that I do not know what BW is
Sorry, my 'English' (technical or otherwise) is often more local 'slang' than what is in a textbook. 🙁 BW = (frequency) bandwidth (Hz).
flare frequency = horn expansion frequency (Hz)
flare factor = hyperbolic, etc.
'throat' in this case is where the horn's expansion begins
If there is no design tutorial in a language you understand well enough for curving a spiral horn, then reverse engineer the W.E. 15/16A horn
For non technically related abbreviations: NetLingo
flare frequency = horn expansion frequency (Hz)
flare factor = hyperbolic, etc.
'throat' in this case is where the horn's expansion begins
If there is no design tutorial in a language you understand well enough for curving a spiral horn, then reverse engineer the W.E. 15/16A horn
For non technically related abbreviations: NetLingo
The effect would vary at different places and frequencies. It is a challenge to maintain a coherent wavefront when the horn curves, so it is kept slow and gradual. At some point higher frequencies may begin to beam inside the horn and reflect their way out.
thanks for your inputs. A Sato horn is always curved, it's probably the whole idea of the horn that you have a long horn that you turn around so that it doesn't take up so much space at the back. However, no one is saying that it has to look exactly as it does today. The point of me making a new one for him is that it shouldn't look like that, since it's a bit too DIY and made from different parts. I might also want to be able to deviate slightly from the curvature of the horn and have it turn less times. Have seen Sato where it seems to only be one 180 arc, where the driver sits just above the mouth and then just turns it 180 degrees. Is that a better solution if I can still keep the length. The shape the horn has now may resemble a fobonacci spiral, but it has a bit too much curvature at the start.
My take on this is that I will not even attempt to get meaningful treble output from a bent horn again. I built a Iwata 220 with a slow 90 degree bend in the middle. And it reduced treble above 2000Hz. My current Iwata 300 horn has output, without any reduction, up to 16kHz, which is as high as the driver can go. Of course, 1500-2000Hz is a great place to cross to a smaller horn.
Everything above 450Hz is straight, and that does not require a very deep horn. Below 450Hz can hopefully be bent and twisted without much trouble, which is plan for my midbass horn.
Everything above 450Hz is straight, and that does not require a very deep horn. Below 450Hz can hopefully be bent and twisted without much trouble, which is plan for my midbass horn.
Yes. One wants to look at what are the wanted properties. How important is loading, avoiding a crossover, where and how the sound flows etc.
The horn is cross at the low at 200-300 Hz. At the top natual rolle off. Super tweeter XO at the button on 13K Hz 1 order.
I will answer in the other thread.
https://www.diyaudio.com/community/threads/sato-we-long-tractrix-vs-expontial.394998/post-7248713
https://www.diyaudio.com/community/threads/sato-we-long-tractrix-vs-expontial.394998/post-7248713