I have been interested for some time in the the Earl Geddes loudspeaker design philosophy of using compression drivers with oblate spheroidal waveguides. Here is the commercial product using Earl's design that I will refer to for comparison:
ESP15 Loudspeaker
I have been looking a long time for a driver & waveguide combination that can span 400 to 20,000 Hz to work as the top cabinet for a bass horn.
Recently, I found a waveguide with a 1.4 inch diameter throat and an outside diameter of 24 inches. It is model DDS SW 1.4-90. It is manufactured in fiberglass by Design Direct Sound (DDS) now located at Sound Bridge Acoustic Labs in Waxahachie, Texas. I purchased three of these waveguides from Assistance Audio for my home theater:
Assistance Audio
I also purchased three patent pending foam plugs from Earl Geddes:
Gedlee Website
Foam Plug Patent Application
I selected the BMS 4594ND compression driver for my system, also available from Assistance Audio. This is the latest coaxial compression driver design from BMS.
BMS 4594nd
A friend brought over a Radian Audio 745PB-8 driver to try as well.
I performed some frequency response measurements to study the acoustic performance. I measured only the compression driver & waveguide mounted on a rectangular baffle 24.25” x 30”. The baffle dimensions were chosen after reading the following information and performing an analysis with "Edge":
Linkwitz on baffle diffraction
Edge Baffle analysis
The attached data was measured with a calibrated Behringer ECM8000 microphone and FuzzMeasure software:
Calibrated mic supplier
FuzzMeasure
An ART USB Dual preamp was used with loopback correction to compensate errors in the preamp response. The frequency response was measured at a distance of 2 feet from the baffle on axis and at 7.5 degree increments out to 90 degrees off axis. The time window was set to preclude room reflections from the measurements.
A Rane RPM88 processor was used for electronic x-over filters. Some conventional parametric EQ was applied manually to flatten the response without the foam plug. You could easily duplicate this performance with an inexpensive Behringer DCX2496. The foam plug response was not re-EQ'd so the effect of the plug can be seen. This could easily be EQ'd flat if desired.
I can not describe the sound of loudspeakers very well, I prefer to provide measured data. The dynamic range is completely effortless while playing uncompressed percussion recordings at levels equal to a live drum kit. At peak rated power, 140 dB at 1 meter is possible, so the BMS driver is quite at ease in the home environment. The sound from this driver/waveguide combination is extremely articulate, precise, and accurate.
The Radian driver was good also, enough for my friend to choose for his home theater. It is available from PartsExpress.
Here are some photos. Data plots in following posts.
The DDS waveguides
A waveguide in the test baffle with foam plug
Center speaker listening test stack (IB Sub Speaker Stand)
ESP15 Loudspeaker
I have been looking a long time for a driver & waveguide combination that can span 400 to 20,000 Hz to work as the top cabinet for a bass horn.
Recently, I found a waveguide with a 1.4 inch diameter throat and an outside diameter of 24 inches. It is model DDS SW 1.4-90. It is manufactured in fiberglass by Design Direct Sound (DDS) now located at Sound Bridge Acoustic Labs in Waxahachie, Texas. I purchased three of these waveguides from Assistance Audio for my home theater:
Assistance Audio
I also purchased three patent pending foam plugs from Earl Geddes:
Gedlee Website
Foam Plug Patent Application
I selected the BMS 4594ND compression driver for my system, also available from Assistance Audio. This is the latest coaxial compression driver design from BMS.
BMS 4594nd
A friend brought over a Radian Audio 745PB-8 driver to try as well.
I performed some frequency response measurements to study the acoustic performance. I measured only the compression driver & waveguide mounted on a rectangular baffle 24.25” x 30”. The baffle dimensions were chosen after reading the following information and performing an analysis with "Edge":
Linkwitz on baffle diffraction
Edge Baffle analysis
The attached data was measured with a calibrated Behringer ECM8000 microphone and FuzzMeasure software:
Calibrated mic supplier
FuzzMeasure
An ART USB Dual preamp was used with loopback correction to compensate errors in the preamp response. The frequency response was measured at a distance of 2 feet from the baffle on axis and at 7.5 degree increments out to 90 degrees off axis. The time window was set to preclude room reflections from the measurements.
A Rane RPM88 processor was used for electronic x-over filters. Some conventional parametric EQ was applied manually to flatten the response without the foam plug. You could easily duplicate this performance with an inexpensive Behringer DCX2496. The foam plug response was not re-EQ'd so the effect of the plug can be seen. This could easily be EQ'd flat if desired.
I can not describe the sound of loudspeakers very well, I prefer to provide measured data. The dynamic range is completely effortless while playing uncompressed percussion recordings at levels equal to a live drum kit. At peak rated power, 140 dB at 1 meter is possible, so the BMS driver is quite at ease in the home environment. The sound from this driver/waveguide combination is extremely articulate, precise, and accurate.
The Radian driver was good also, enough for my friend to choose for his home theater. It is available from PartsExpress.
Here are some photos. Data plots in following posts.
The DDS waveguides
An externally hosted image should be here but it was not working when we last tested it.
A waveguide in the test baffle with foam plug
An externally hosted image should be here but it was not working when we last tested it.
Center speaker listening test stack (IB Sub Speaker Stand)
An externally hosted image should be here but it was not working when we last tested it.
I can only post 4 pics per post.... (see first post in this thread for more info)
Yes, the filter is -6dB at 400 Hz.
The frequency response was measured at a distance of 2 feet from the baffle on axis and at 7.5 degree increments out to 90 degrees off axis. The time window was set to preclude room reflections from the measurements.
I tried to mimic the measurement plot format of the ESP15 (Summa) loudspeaker which was obtained from the referenced web site. The plots are the same scale, size, and proportion for easy visual comparison using the Summa as a benchmark. All plots use 1/3 octave smoothing.
Note the directivity in plot one, the effect of the foam pug in plot two, harmonic distortion in plot three, and the Summa for comparison in plot 4.
Keep in mind the Summa plots include the woofer/midrange. I plotted only the waveguide.
The DDS with BMS and -6dB @ 400 HZ crossover
An externally hosted image should be here but it was not working when we last tested it.
The DDS with BMS and foam plug
An externally hosted image should be here but it was not working when we last tested it.
The DDS with BMS and foam plug @ 15 degrees off axis, 2nd & 3rd order harmonic distortion
An externally hosted image should be here but it was not working when we last tested it.
The reference standard ESP 15 (Summa including woofer)
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Whoah, I thought my Summas were big :O
A few questions:
Anyways, cool stuff! Would love to hear it!
A few questions:
- are you using the stock crossover from the midrange to the hi frequency compression driver?
- Would you be willing to post the price you paid for the compression driver? Assistance Audio never answers my phone calls or emails, and US Speaker doesn't list that driver. I checked on the 'bay too.
- Didn't you use to have Unity horns? IIRC you had lab subs too.
Anyways, cool stuff! Would love to hear it!
I bi-amped the BMS with a Rane RPM88 using a 4th order x-over.
Here's a direct link to the waveguide price (at bottom of page):
Waveguide $$$
The BMS 4594 ND was $610 back in April 2009.
Answer to last question is yes
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Here's the response with a Radian 745PB-8 driver. This is a single diaphragm driver not a coaxial like the BMS.
Again, like with the BMS, I applied some parametric EQ to flatten the response.
The DDS with Radian and -6dB @ 500 HZ crossover
Again, like with the BMS, I applied some parametric EQ to flatten the response.
The DDS with Radian and -6dB @ 500 HZ crossover
An externally hosted image should be here but it was not working when we last tested it.
The frequency response looks lame and would be un-useable without EQ or passive filtering like Earl Geddes uses. This is a constant directivity waveguide and as such the driver has to be EQ'd to flat power response. Sorry, I had best guess filters in place before I even powered it up.
Update: I found a measurement I took with the Radian driver and no filters.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Hi Brad,
Very cool stuff. What bass driver are you going to use?
I've look at your Labsub 3.0 plans a number of times and would love to get the the point of being able to build some.
Any tips for setting up Sigmaplot to do graphs like Earl's?
Earl had mentioned not being a big fan of the BMS compression drivers due to some polar response anomalies at the crossover point (~6K) in the coax compression driver he measured. I don't see anything liek that here though.
Very cool stuff. What bass driver are you going to use?
I've look at your Labsub 3.0 plans a number of times and would love to get the the point of being able to build some.
Any tips for setting up Sigmaplot to do graphs like Earl's?
Earl had mentioned not being a big fan of the BMS compression drivers due to some polar response anomalies at the crossover point (~6K) in the coax compression driver he measured. I don't see anything liek that here though.
The "stock" crossover on the BMS 2" coaxial compression driver is notorious for not working very well. Tom Danley discussed it over on audioasylum back when he was designing the synergy horn. (If you look at the synergy horn patent, it appearts to be using a BMS coax.)
As Brett noted, he's using his own crossover here, and this is also a newer, more expensive, and smaller coax.
IMHO, when it comes to doing coaxial speakers, the closer you can get the drivers together, the better.
It's a very interesting project, can't wait to see how it turns out. I've messed around a lot with coaxial horns and BMS compression drivers. Here are a few of the threads:
http://www.audiogroupforum.com/csforum/showthread.php?t=62789
http://www.diyaudio.com/forums/showthread.php?postid=1495477
http://www.diymobileaudio.com/forum/diy-mobile-audio/60146-creating-perfect-soundstage.html
It's quite interesting to see the same dip in these waveguides. I wonder what they have in common. Up to now, I have not seen such on-axis dips at such low frequency in any test of my own waveguides.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.