I was wondering which parameters determine the xover point with a waveguide in terms of directivity. I am using a 15-inch OB design with a B&C DE250 in a 10.6-inch 90 degree DDS waveguide.
From everything I have read the 15-inch narrows significantly around 800-900 Hz and the DDS loses directivity around 1K - 1.2 Khz. Given that I am using extremely steep slopes (48 dB, LR active) what might be the optimum compromise for a xover point? I am currently set at 957 Hz.
From everything I have read the 15-inch narrows significantly around 800-900 Hz and the DDS loses directivity around 1K - 1.2 Khz. Given that I am using extremely steep slopes (48 dB, LR active) what might be the optimum compromise for a xover point? I am currently set at 957 Hz.
I've been fooling around with the same issue as of late, though with a different set of drivers/ horn...
My conclusion was that there's no textbook answer, and that the only way to figure out the best crossover point in terms of matching directivity is to get out the measuring equipment and knock out a series of on- & off-axis measurements. For me, I found what was actually going on in terms of directivity to be a bit suprising, and not what I was expecting at all (of course, that could be just because of my lack of experience).
Regards,
John
My conclusion was that there's no textbook answer, and that the only way to figure out the best crossover point in terms of matching directivity is to get out the measuring equipment and knock out a series of on- & off-axis measurements. For me, I found what was actually going on in terms of directivity to be a bit suprising, and not what I was expecting at all (of course, that could be just because of my lack of experience).
Regards,
John
That DDS waveguide would most likely be best used with an 8" mid driver, 10" at the most.. crossed somewhere between 1.5 - 2 KHz. I have three pro 12's that I bought with the intention of using them with this DDS waveguide (never bought them tho), but put the project off for a number of reasons, including this. It just seems to me like anything near or above about 1 KHz from larger cone drivers (8" or 10" and above?) quickly turns into screaming. 
I get pretty good pattern matching in the horizontal plane by overlapping my 12" woofer with the DDS waveguide somewhat, with both of them shelved down a bit. The tradeoff is lobing in the vertical plane, but in my opinion that's not as bad as an anomaly in the horizontal plane, and nobody has ever noticed it. Accomplishing the shelved overlap calls for some tweaking of the crossover, and isn't something that's readily do-able with an off-the-shelf steep-slope active crossover. That's why I prefer a passive crossover - I can usually make it do what I want.
Also, the 12" woofer I'm using has better off-axis characteristics than most (AE Speakers Lambda series TD-12M). I wouldn't try to match up the DDS with the Alpha 15.
My suggestion: Get the big Geddes waveguides if you want to use the Alpha 15.
Duke
edit - Horizons, if you didn't get an e-mail from me, shoot me one at: audiokinesis at yahoo dot com.
Also, the 12" woofer I'm using has better off-axis characteristics than most (AE Speakers Lambda series TD-12M). I wouldn't try to match up the DDS with the Alpha 15.
My suggestion: Get the big Geddes waveguides if you want to use the Alpha 15.
Duke
edit - Horizons, if you didn't get an e-mail from me, shoot me one at: audiokinesis at yahoo dot com.
>>overlapping my 12" woofer with the DDS waveguide somewhat, with both of them shelved down a bit.
I think I could kludge that with the DCX2496 by overlapping a bit and adding a small Q parametric dip near the overlap point on both sides of the xover.
I still don't know where to do this though... Does anyone know where the DDS starts to lose directivity? 1.2 K or thereabouts?
I think I could kludge that with the DCX2496 by overlapping a bit and adding a small Q parametric dip near the overlap point on both sides of the xover.
I still don't know where to do this though... Does anyone know where the DDS starts to lose directivity? 1.2 K or thereabouts?
Now THIS is interesting to me. This is a direct quote from Clayton Shaw and EP regarding his new CS1. I am surprised that he is crossing over the BMS unit in the DDS WG at 750 Hz. I am fairly certain that the B&C DE250 can be crossed as low as 800. Now I am really confused. I guess more trial and error/RTA will be needed.
>>For the record, the CS1 incorporates the DDS waveguide and a German made BMS compression unit with neodymium motor assembly. This combination not only sounds very good and smooth, but is more stable at low frequencies (CS1 xover is 750Hz) than most other 1 inch format compression drivers. Its annular diaphram design has a ring radiator layout with twin, concentric voice coils and a mylar diaphram material.
>>For the record, the CS1 incorporates the DDS waveguide and a German made BMS compression unit with neodymium motor assembly. This combination not only sounds very good and smooth, but is more stable at low frequencies (CS1 xover is 750Hz) than most other 1 inch format compression drivers. Its annular diaphram design has a ring radiator layout with twin, concentric voice coils and a mylar diaphram material.
nullspace said:
Absolutely true. It takes me more that a full day of measurements and analysis (and thats with software that I have evolved over the last ten years) to even find where the crossover points should be. And then the odds are that the upper and lower points are not the same. Matching a waveguide with a woofer is not an easy thing, which is why I don't recommend people just buy a waveguide and woofer and throw them together. High slope rates don't help at all and in fact can often make things worse. I now use lower slopes than I used to because I get better response.
The only thing that works is careful design. I have analysis software that will yield the full polar response with a simulated crossover which cuts out days or weeks of trial and error. Text book approaches always seem to fail because there is so much going on at the low end of a waveguide and just as much at the high end of the woofer.
Text book crossovers never work that well with pistons and they don't work at all with waveguides.
Looking at this device the mouth radius is too small for optimum low end extension.
Keele showed that if the horn consists of two conical sections with the diameter ratios of around .65-.7:1, the outer section around twice the included angle of the inner, then the low end would have constant directivity down to a frequency around 1.2 times lower than the frequency where waisting occurs with a plain conical horn.
A plain conical horn with a mouth .267m. dia. gives..
25306/(90*.267) = 1053Hz.
If we terminate the inner 90degree section at .174m. dia. and provide a 160degree outer section, (this can be a circular arc), then the 90degree directivity will be extended down to 878Hz.
The plain conical horn will have a directivity of around 75degrees at this frequency.
rcw
Keele showed that if the horn consists of two conical sections with the diameter ratios of around .65-.7:1, the outer section around twice the included angle of the inner, then the low end would have constant directivity down to a frequency around 1.2 times lower than the frequency where waisting occurs with a plain conical horn.
A plain conical horn with a mouth .267m. dia. gives..
25306/(90*.267) = 1053Hz.
If we terminate the inner 90degree section at .174m. dia. and provide a 160degree outer section, (this can be a circular arc), then the 90degree directivity will be extended down to 878Hz.
The plain conical horn will have a directivity of around 75degrees at this frequency.
rcw
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.