Someone built some OSWGs
Building Oblate Spheroid Waveguides Far and away the most difficult project I've done to date. Article By Jeff Poth
Heaven knows who it was
Building Oblate Spheroid Waveguides Far and away the most difficult project I've done to date. Article By Jeff Poth
Heaven knows who it was
This conclusion:
"..Only Dr. Geddes offers an OSWG design with a waveguide as large as this one, the imposing "Summa"."
..may be correct, but there is at least one other similar profile that is large:
http://www.diyaudio.com/forums/multi-way/149085-testing-big-waveguide.html
I think the small horn of this classic piezo tweeter would fit the 075" version:
Goldwood GT-1016 2" x 5" Wide Dispersion Piezo Horn Tweeter | Parts-Express.com
Goldwood GT-1016 2" x 5" Wide Dispersion Piezo Horn Tweeter | Parts-Express.com
There is also a great waveguide group buy going that includes OSWGs, even an incredible looking speaker enclosure with a 12" OSWG. I think its about $800 for a pair if we can get a pallet of 4 into the US.
How do the JBL screen array designs do so well with their great then > 1" CDs??
Also how did Paul W. Raptor build with a 1.5 CD win a ton of fans at the DIY Atlanta event?
I guess my real question here, aren't there some > 1" CD designs that can do very well above 8KHz?
None that I have seen. I have built several 2" OS waveguides and tested them. They all died at about 8 kHz., just like the 1" tend to die at about 16 kHz and the 1.5" at about 11 kHz. I can live with 16 kHz, and maybe even 11 kHz, but certainly not 8 kHz.
Please, don't quote me subjective data, I'm talking about quantified facts here not impressions.
It's loss of directivity above 10K. The older JBL Biradials used a diffration slot to narrow the throat to control the directivity up to say 15-16k. I use 1.5" PT Waveguides and they do essentially the same thing as the older biradials but with much smoother transitions. Thet are all going to have a increasing DI above 10K just like a 1" dome tweeter does.
Rob
I try not to quote subjective anything because Im not subjective. The Raptor build is easy to find, all meaurements show that it did very well above 8KHz with the 18sound waveguide/CD. JBL arrays are famous for high end sound and they have measured response way past 8KHz.
Right now the only subjective POV in our discussion was your posting stating "do not have a satisafcatory response above 8-10 kHz". From the measurements I have seen this is a false statement so Im looking for clarification what you mean by satisfactory??
here is the Raptor with a 1.5" CD
HTGuide Forum - The Raptor ... a 10" MTM
What is unsatisfactory about the response in that build?
JBL :: Product Family
Those use 4" CDs....and go to 19KHz (+/3dB) so what is bad about that response?
I have found that pushing the tweeter lower is far better then running the woofers higher so Im very curious about 1.5" CDs and larger waveguides going down to 600Hz or so. Some of the best HT speakers I have heard are the JBL screen arrays
Last edited:
Doug - Earl has been describing an OS profile. The 18 Sound XT1464 is not an OS profile.
In other words, if you want an upper freq. response that is extended BEYOND 8 kHZ on and off axis in an OS waveguide, THEN you generally need something smaller than a 1.4" exit driver.
There isn't anything wrong with the XT1464, but it's directivity is narrower than an OS profile, and past 40 degrees it essentially will NOT match the directivity of the pairing woofer (..and doesn't in the Raptor.) It is of course still a very good design, but it is different.
Attachments
Last edited:
What Rob says. 
1.5":
1.0":
Larger throat diameter does not support constant directivity as far into the top octave; Earl told us this years ago.
1" BMS 4552nd on QSC 152i does not quite make it to 20 kHz.
1.5" BMS 4555 on JBL PT-F95HF begins losing it down at 8 kHz.
[These are normalized polar maps from 5° increment data.... ]
1.5":
1.0":
Larger throat diameter does not support constant directivity as far into the top octave; Earl told us this years ago.
1" BMS 4552nd on QSC 152i does not quite make it to 20 kHz.
1.5" BMS 4555 on JBL PT-F95HF begins losing it down at 8 kHz.
[These are normalized polar maps from 5° increment data.... ]
Attachments
Last edited:
http://techtalk.parts-express.com/showthread.php?p=1683904#post1683904
Here are the non-normalized maps, from which Earl might be willing to analyze these designs.
1.5":
1.0"
1/6 octave smoothed also provided below:
Here are the non-normalized maps, from which Earl might be willing to analyze these designs.
1.5":
1.0"
1/6 octave smoothed also provided below:
Attachments
Last edited:
No thanks. The last time you asked me to do that, you crucified me because you didn't like what I had to say.
They all look heavily smoothed. Send me the raw data and I'll post them plotted the same way as I do my own.
Yes, smoothed and normalized to illustrate the difference we are presently discussing.
Elias was able to read the CLIO MLS measurement files. Can you do the same?
Otherwise, I will reprocess them into ASCII data files, 2048 points max. My options are:
1) Display Frequency Data
2) FFT Frequency Data
3) Time Data
512 points sufficient?
Elias was able to read the CLIO MLS measurement files. Can you do the same?
Otherwise, I will reprocess them into ASCII data files, 2048 points max. My options are:
1) Display Frequency Data
2) FFT Frequency Data
3) Time Data
512 points sufficient?
Last edited: