The contour is a hyperbola (i.e. exact oblate spheroid shape) connected to clothoid at about 1/3 of the depth from the throat. I don't have the actual contour at hand.
Is it really 1/3 from the throat?
So it's only about 1/3 hyperbola, 2/3 clothoid?
Seems not much hyperbola, is it actually 1/3 of the depth from the mouth?
Best wishes
David
That's right. 1/3 to 1/2 of hyperbola give pretty good results. Make it less, it will beam more, make it more, all the mouth related issues will come up.
Is this based on an arbitrary size for a given frequency or has one been chosen to suit the other, if so by what criteria?
In the following drawing it's only about 1/5 of hyperbola, the rest is clothoid. In the rendering in will be a bit more.
AllenB: Total depth of the horn/WG, coverage angle and this hyperbola ratio is given. From that the shape is calculated. The clothoid is just scaled to be tangent to a baffle.
AllenB: Total depth of the horn/WG, coverage angle and this hyperbola ratio is given. From that the shape is calculated. The clothoid is just scaled to be tangent to a baffle.
Attachments
Last edited:
Ok, that's less than I expected, my recollection is that D.B. Keele recommended around 1/3 for the mouth termination and this seems to have become traditional.
The optimisation of the available space between "horn" and "transition" is a problem that I recently started to think about.
It did seem that if the space constraints are not severe then extra transition would be beneficial.
Your simulation would seem to support that idea, thank you for the data.
For a 2 way speaker there is a complication that the woofer separation must not increase too much.
I think I need to blend the woofer into the wave-guide, so it won't be completely axisymmetric
Best wishes
David
In fact clothoid is not curved much at the beginning so the WG continues in almost conical shape for a while. But it slowly diverges from that all the time, which is very good for the overall smoothness of the countour.
Nice picture, even when the hyperbola is only 1/5 of the depth the curve is still not too far from a hyperbola for quite a bit further, explains why such a small ratio is still useful.
Best wishes
David
Is this the contour of the horn in post #2890?
In the end, you prefer round over rectangular/elliplical?
Yes, I had essentially the same observation but cross posted after you and didn't notice until after the edit window.
I had planned to be lazy and just have a simple radius but your work makes me rethink, the clothoid is not only theoretically better but probably makes a practical difference too.
Do you have a comparison of a hyperbola terminated with a radius versus the clothoid?
Best wishes
David
😉 Finishing my impressions as we speak.
How about this as the complete opposite of a smooth (OS) throat?
How about this as the complete opposite of a smooth (OS) throat?
Last edited:
These are all just random examples I gathered and I don't remember but I don't think so.
It's hard to decide. If both are comparably good in the end, it's a matter of aesthetics I guess.
Not yet but I'm going to add radius as an option into the program. Then it will be easily possibe. I'm curious as well.
Is this a picture of "the merging chamber" output/ high way of J-1?
Hope they were off when you took pictures!!!
Hope they were off when you took pictures!!!
This is the combiner indeed.
Picture isn't mine. I hardly ever take a phone to parties, or other social events.
This is what the rear looks like:
Picture isn't mine. I hardly ever take a phone to parties, or other social events.
This is what the rear looks like:
Woaw... another low efficiency design!
You lost my interest... 😀
No phones at social events? (I'm assuming your way younger than me) Man you made my day! I'm deseparate to see youth not communicate old way this days.
You just brought hope to me. Thank you.
You lost my interest... 😀
No phones at social events? (I'm assuming your way younger than me) Man you made my day! I'm deseparate to see youth not communicate old way this days.
You just brought hope to me. Thank you.
Last edited: