Hi all,
I wanted to share the latest omnidirectional project I'm wrapping up for those of you who don’t frequent some of the other forums I have posted it on. Technically, it might be best described as a hybrid omnidirectional speaker system, not a true omni. I have been calling it HOSS (hybrid omnidirectional speaker system). The drivers are all Dayton RS drivers, including the RS265HF 10” sub, two RS150-4’s and an RS28AS for tweeter. I wanted to make it more acceptable aesthetically than my first omni project and at the same time address a couple of possible sources of weakness in that design.
The project includes two RS150-4's in series, one up and one down firing as in my first omni design. The RS150 allows for a narrower baffle and smaller boxes than my first omni which used the 7” RS180-8. They can use sealed boxes about 1/3 the size of the RS180's, can be crossed a few hundred hertz higher, which pushes the RS28AS less at its low end. The baffles are rectangular, rather than square and offset the RS150's toward the front of the baffle. As a result of the forward placement of the RS150-4, the tweeter baffle is also not inset as much as it was on my first omni design, which help reduce diffraction effects from the up and down firing driver baffle edges and increases the vertical dispersion some.
The RSS265HF will be in a separate, tapered enclosure. I wanted to keep it small to improve aesthetics, but have good low end output with an F3 around 30hz and with room gain plays solidly well into the 20’s. I’ve always been interested in TL’s and decided to take that approach with this bass bin. I decided to use the RSS265HF because it has low distortion, works in a relatively small volume and can be crossed higher than some other choices. Paul Kittinger has helped me with the tapered TL design, which ended up at about 71” in length, 20:1 taper, in 31 net liters, with an F3 of around 29hz, F6 around 24hz and F10 of 19hz. The slot output to the TL is in the back. The RSS265 is front rather than down firing, for a little more traditional look, but should be omni for most, maybe all of its effective range, since it is crossed over at about 180hz. The sub driver is placed 14” along the TL, about 20% from the end.
I wanted to keep it as a passive design on the MTM section. For now I’m using 240w PE plate amps, that have a variable crossover of up to 180hz with a 24db/octave low pass slope to low pass the RSS265. The passive crossover for the MTM section includes a high pass to the RS150-4’s (in series) and then a crossover to the front firing RS28 at 1560hz, all with approximately 4th order acoustic slopes.
All drivers are physically positioned to be time aligned on a radius from a 37" ear height, with a 12' distance to the listening position.
So, how does it sound? Let’s say I’m very, very pleased. It is the best speaker I’ve built or owned so far. The low end is very deep and clean and does not seem to excite room modes to the same degree as my previous sealed 12” stereo sub bins. The driver integration is excellent and the presentation is very smooth. I worked very hard to bet the RS-150’s and RS28 to match phase as closely as possible through their pass band. The vertical and horizontal off axis performance is excellent and the soundstage is very wide and deep. I believe that the transition from omnidirectional low and midranges to direct firing tweeter has the benefits of the large omni soundstage, yet provides excellent imaging.
Attached are some photo’s of the almost complete project, some as-built frequency response measurements on and off the horizontal and vertical axis as well as individual TL terminus, driver and reverse null measurements. The finish is Santos Rosewood veneer in high gloss with high gloss black on the horizontal elements. I hope this is not too big a post and images!
I wanted to share the latest omnidirectional project I'm wrapping up for those of you who don’t frequent some of the other forums I have posted it on. Technically, it might be best described as a hybrid omnidirectional speaker system, not a true omni. I have been calling it HOSS (hybrid omnidirectional speaker system). The drivers are all Dayton RS drivers, including the RS265HF 10” sub, two RS150-4’s and an RS28AS for tweeter. I wanted to make it more acceptable aesthetically than my first omni project and at the same time address a couple of possible sources of weakness in that design.
The project includes two RS150-4's in series, one up and one down firing as in my first omni design. The RS150 allows for a narrower baffle and smaller boxes than my first omni which used the 7” RS180-8. They can use sealed boxes about 1/3 the size of the RS180's, can be crossed a few hundred hertz higher, which pushes the RS28AS less at its low end. The baffles are rectangular, rather than square and offset the RS150's toward the front of the baffle. As a result of the forward placement of the RS150-4, the tweeter baffle is also not inset as much as it was on my first omni design, which help reduce diffraction effects from the up and down firing driver baffle edges and increases the vertical dispersion some.
The RSS265HF will be in a separate, tapered enclosure. I wanted to keep it small to improve aesthetics, but have good low end output with an F3 around 30hz and with room gain plays solidly well into the 20’s. I’ve always been interested in TL’s and decided to take that approach with this bass bin. I decided to use the RSS265HF because it has low distortion, works in a relatively small volume and can be crossed higher than some other choices. Paul Kittinger has helped me with the tapered TL design, which ended up at about 71” in length, 20:1 taper, in 31 net liters, with an F3 of around 29hz, F6 around 24hz and F10 of 19hz. The slot output to the TL is in the back. The RSS265 is front rather than down firing, for a little more traditional look, but should be omni for most, maybe all of its effective range, since it is crossed over at about 180hz. The sub driver is placed 14” along the TL, about 20% from the end.
I wanted to keep it as a passive design on the MTM section. For now I’m using 240w PE plate amps, that have a variable crossover of up to 180hz with a 24db/octave low pass slope to low pass the RSS265. The passive crossover for the MTM section includes a high pass to the RS150-4’s (in series) and then a crossover to the front firing RS28 at 1560hz, all with approximately 4th order acoustic slopes.
All drivers are physically positioned to be time aligned on a radius from a 37" ear height, with a 12' distance to the listening position.
So, how does it sound? Let’s say I’m very, very pleased. It is the best speaker I’ve built or owned so far. The low end is very deep and clean and does not seem to excite room modes to the same degree as my previous sealed 12” stereo sub bins. The driver integration is excellent and the presentation is very smooth. I worked very hard to bet the RS-150’s and RS28 to match phase as closely as possible through their pass band. The vertical and horizontal off axis performance is excellent and the soundstage is very wide and deep. I believe that the transition from omnidirectional low and midranges to direct firing tweeter has the benefits of the large omni soundstage, yet provides excellent imaging.
Attached are some photo’s of the almost complete project, some as-built frequency response measurements on and off the horizontal and vertical axis as well as individual TL terminus, driver and reverse null measurements. The finish is Santos Rosewood veneer in high gloss with high gloss black on the horizontal elements. I hope this is not too big a post and images!
The last image was cut off, so here it is, the plus vertical off axis measurement.
An externally hosted image should be here but it was not working when we last tested it.
My room is about 15'w x 20' deep, with steeply vaulted ceilings from the end of the room the speakers are in over about 12' of that 20' depth. The speakers are about 3' out from the front wall and 3-4' from the side walls. It is a somewhat open floor plan and this room opens up into other rooms as well. This is our major living area and and they are used for HT and music.
Beautiful build Dan! I'm familiar with the Daytons so I know the speakers sound great. I love to see "unconventional" designs, but more than that your speakers look like something one would find in a boutique - the finish and detail are spectacular - bravo!!
Thanks!
An intereting aspect of the omni designs where you are listening 90º off axis to the drivers, as you are here with the RS150-4's, is that their break up modes and distortion profiles also measure well down in comparison to on axis applications of the same driver since the driver naturally rolls off as you move off axis. In other words, I believe they sound better and are easier to handle in the crossover in this applciation than they are in a typical monopole application.
An intereting aspect of the omni designs where you are listening 90º off axis to the drivers, as you are here with the RS150-4's, is that their break up modes and distortion profiles also measure well down in comparison to on axis applications of the same driver since the driver naturally rolls off as you move off axis. In other words, I believe they sound better and are easier to handle in the crossover in this applciation than they are in a typical monopole application.
Yes that makes sense - break-ups at the higher frequencies will not be as noticable off axis - makes for some interesting experimental potential for metal drivers.
Couple of questions - do you have a schematic of the crossover? I understand it's 4th order but did you notch the woofers too? Or with the natural off-axis attenuation did you find this unnecessary?
Also, how did you arrive at using two midwoofers, and have them facing eachother? I'm wondering if this arrangement might provide some kind of complimentary lensing effect, although they may be too far away from eachother for this to work. Or was it simply to help compensate for the loss of efficiency inherent in omni designs?
Couple of questions - do you have a schematic of the crossover? I understand it's 4th order but did you notch the woofers too? Or with the natural off-axis attenuation did you find this unnecessary?
Also, how did you arrive at using two midwoofers, and have them facing eachother? I'm wondering if this arrangement might provide some kind of complimentary lensing effect, although they may be too far away from eachother for this to work. Or was it simply to help compensate for the loss of efficiency inherent in omni designs?
Here are a few more images. The crossover, which also includes a high pass to the RS150-4's, the measured impedance and impedance phase and the measured frquency response with the phase shown. The phase tracks very well from around 400hz through 2.7khz (see the red, black and blue phase lines). The purple phase line represents the phase with the reverse null at the xover point.
It's a pretty standard LR crossover, 4th order acoustic, not electrical, with a zobel on the RS150-4's, but it attenuates the woofer breakups by 40-45db .
The reason for two drivers is primarily due to losses from your listening axis being 90º off the drivers axis and alos due, I believe, to the fact that they are really playing in 4pi space throughout their range rather than 2pi for part of the range. The woofer firing at each other in order to better integrate them with the tweeter. The output is much like a single driver in many ways, IMO. The distance between them is quite important and was subject to a lot of testing in this and my prior omni project. Get them to close together and you start getting peaks and valleys created by interactions between the baffles and drivers. Get them too far apart and they won't sound as integrated. They are 9.5" apart, which is actually about 1" closer than would be practical with these drivers on a traditional flat baffle, so they should integrate better for that reason alone. I have not found any lobing to occurr vertically or horizontally off axis in my measurements.
It's a pretty standard LR crossover, 4th order acoustic, not electrical, with a zobel on the RS150-4's, but it attenuates the woofer breakups by 40-45db .
The reason for two drivers is primarily due to losses from your listening axis being 90º off the drivers axis and alos due, I believe, to the fact that they are really playing in 4pi space throughout their range rather than 2pi for part of the range. The woofer firing at each other in order to better integrate them with the tweeter. The output is much like a single driver in many ways, IMO. The distance between them is quite important and was subject to a lot of testing in this and my prior omni project. Get them to close together and you start getting peaks and valleys created by interactions between the baffles and drivers. Get them too far apart and they won't sound as integrated. They are 9.5" apart, which is actually about 1" closer than would be practical with these drivers on a traditional flat baffle, so they should integrate better for that reason alone. I have not found any lobing to occurr vertically or horizontally off axis in my measurements.
Hope this is more of a sidebar than a thread hijack, because it's on the same topic - but given your experience with this type of design, perhaps you could give me some feedback on an idea of mine -
I have envisioned a dual woofer "omni" also, but with each mounted at 45º instead of horizontal, and with one front firing and the other rear firing. They would be mounted directly opposite eachother. The tweeter could be mounted in between the woofers, in its own little baffle similar to yours, on the top of the cabinet, or perhaps in the cab below the front woofer.
Perhaps the effect would be something of an omni/bipole hybrid?
I have envisioned a dual woofer "omni" also, but with each mounted at 45º instead of horizontal, and with one front firing and the other rear firing. They would be mounted directly opposite eachother. The tweeter could be mounted in between the woofers, in its own little baffle similar to yours, on the top of the cabinet, or perhaps in the cab below the front woofer.
Perhaps the effect would be something of an omni/bipole hybrid?
If I follow your description correctly the potential problem I see with any front to back omni woofer design is how do you get the phase to match? If wired out of phase, will they cancel each other out if they are too close together? If wired in phase, How will they be able to integrate with each other and the tweeter, since all three drivers are at a significantly different distance from the listening position?
I'm not saying it won't work, but I'm skeptical that the phase match between the two woofers and the woofers and the tweeter could be overcome without either a digital delay element added or a complex passive ladder delay network. You might want to experiment with that concept with some inexpensive drivers before sinking a lot into it.
My omni designs so far have been very careful to have the drivers VC's time aligned physically to the listening position to ensure that this is not an issue.
I'm not saying it won't work, but I'm skeptical that the phase match between the two woofers and the woofers and the tweeter could be overcome without either a digital delay element added or a complex passive ladder delay network. You might want to experiment with that concept with some inexpensive drivers before sinking a lot into it.
My omni designs so far have been very careful to have the drivers VC's time aligned physically to the listening position to ensure that this is not an issue.
Gotcha - thanks.
Now back to your speakers - did you do all the woodwork yourself? What kind of veneer is it and how did you finish it?
Have you considered a website for your designs?
Now back to your speakers - did you do all the woodwork yourself? What kind of veneer is it and how did you finish it?
Have you considered a website for your designs?
Truly, you did a great job with the finish. I forget where you're from- do you ever plan on bringing them to a DIY meet?
sdclc126 said:
Yes I did all the woodwork. They are all mdf with veneer and/or paint. The veneer is Santos Rosewood with a high gloss poly finish (which was a pain to do since I found out that Rosewood is an oiley wood like teak). The paint is zinser sanding sealer over mdf (about 3 coats with sanding in between) the spray primer (few coats), then black enamel paint, then an enamel clear coat, all done with spray cans and sadingf between with various grits up to 1500.
Lou Corragio has generously offered to donate his website and skills to create some pages and host my designs. I'm in the process of writing this one up to send to him to get that started.
- Status
- Not open for further replies.