can we see a frontal picture, @CinnamonRolls? I still do not understand how the compression driver terminates into the waveguide ...
I would agree with that statement.
I looked into putting the knee at 500 Hz. But the waveguide and the woofer (24" and 18" respectively) was just too much to handle. If it had been easier, I would have done it. After nearly some two decades with them now, I am happy with the compromise. Several people have noted that the 15" is just a bit better (subjectively) than the 12" which, of course, was better than the 10", and from this I took that going to 18" wasn't going to buy me much.
The green part is where the bolts fasten the waveguide into the surface of the compression driver. Below that is the conical portion of the compression driver with the 22.5mm cylindrical throat. It ends just above the phase plug about 25 mm down into the compression driver. The DE250 has a conical exit that is empty so it's easy to fit the 3D printed "internal ring" in there. The phase plug ends where the orange waveguide starts. The DE111 is similar but not as deep. I tried to make the cylindrical throat diameter the same diameter as the phase plug exit. Does this help?
I attached a 3mf file so you can see what it looks like in 3D if you have the free Microsoft 3D Builder program (or any 3mf viewer).
Last edited:
Very much so. What is the acoustical effect of the parallel walla duct section?
Even if there are resonances induced by the ring extension, they are well-behaved as you showed already. It should be easy to correct the FR even with a simple minimum-phase EQ (including any time domain effects). It probably illustrates what Earl has been saying all along, that the throat acoustic impedance has only a mild effect on the frequency response.
As for the DI knee, this is what I'm getting for my large rectangular horn, 64 x 48 cm ~ 25" x 19" (https://www.diyaudio.com/community/...-design-the-easy-way-ath4.338806/post-7366204):
Horizontal, vertical (simulated data):
It's not nearly as clean as a R-OSSE, obviously, there's no rollback. Purely aesthetic reasons. Hopefully it's good enough
Horizontal, vertical (simulated data):
It's not nearly as clean as a R-OSSE, obviously, there's no rollback. Purely aesthetic reasons. Hopefully it's good enough
Last edited:
Just some anecdotical evidence: I used a screw-on adapter to fit a flange-mounted compression driver to a waveguide that doesn’t. This introduces a parallel duct section, probably much longer than the short section in your printed waveguide. It did not appear to be desirable:
https://www.diyaudio.com/community/threads/digital-bsc-question.370349/post-6696468
Let's not forget that even without a ring plug, such reflections are already there, simply due to the existing conical exit section. With the ring plug it's just a bit different shape. The advantage being the smaller throat diameter within which all this happens. With the ring plug it's actually a bit shorter. It really improves the system in the end, as the presented data suggest.
Adding a cylindrical extension to an existing exit of a CD is a different and lot worse thing to do..
Adding a cylindrical extension to an existing exit of a CD is a different and lot worse thing to do..
Last edited:
- I would try to insert a piece of an open-cell foam (cotton wool?) into the cylinder, just to see the effect. It could be relatively dense for this purpose so it would better show what are the frequency regions where these reflections/resonances are actually in effect (if at all).
In this context I would like to draw attention again to the FANE CD131, which is quite nice (and cheap) 1" driver - it has a 3-slot annular phase plug ending at ⌀16 mm and 30° conical exit. Just perfect for these experiments. I hesitated to use the cylindrical extensions but as they seem not to cause any real trouble, this opens up some new possibilities.
In this context I would like to draw attention again to the FANE CD131, which is quite nice (and cheap) 1" driver - it has a 3-slot annular phase plug ending at ⌀16 mm and 30° conical exit. Just perfect for these experiments. I hesitated to use the cylindrical extensions but as they seem not to cause any real trouble, this opens up some new possibilities.
Last edited:
You posted some simulations a while ago:
https://www.diyaudio.com/community/...he-easy-way-ath4.338806/page-194#post-6403180
and
https://www.diyaudio.com/community/...he-easy-way-ath4.338806/page-192#post-6390401
Wouldn't it be more convenient to find a driver with a Neodymium magnet to reduce the depth of the conical section?
https://www.diyaudio.com/community/...he-easy-way-ath4.338806/page-194#post-6403180
and
https://www.diyaudio.com/community/...he-easy-way-ath4.338806/page-192#post-6390401
Wouldn't it be more convenient to find a driver with a Neodymium magnet to reduce the depth of the conical section?
In the light of the above data I would now tend to think that a smaller exit diameter of a phase plug is favorable over a reduced depth.
- Do you have any tips?
- Doesn't anyone know how does the SICA CD 105.65/N220 look like under the front cover?
- Do you have any tips?
- Doesn't anyone know how does the SICA CD 105.65/N220 look like under the front cover?
Last edited:
That's another Rubicon. Absolute, non-normalized measured data is all that matters here I guess. Go for it
I would be very interested to see it in action, in a direct comparison with the previous results.