I've read quite a bit about Dunlavy's and now Lipinski's being used in studios. I think I understand about the WMTMW design, it being very similiar to the idea of the MTM D'Appolito arrangement. I also understand about Dunlavy using first order crossovers. I understand about using sealed enclosures. What I definitely don't understand is their use of felt and foam around the tweeter in particular. Is it purely a way of trying to reduce diffraction or is there some kind of horn loading of these designs or something else? I'd like to know. I can not recall seeing any discussion about it online. Anyone?
I've read John Dunlavy's white papers on his design principles.
Dunlavy (and Duntech) speakers have stepped baffles, which in conjuction with the first order filters, create a linear phase response. i.e the waveform shape is preserved. Because of the steps in the baffle, severe diffraction would occur, and so heavy felt damping is applied to absorb the sound waves before they hit the edges. It also reduces diffraction with the sides of the baffle.
Take a look at www.duntech.com.au , the company John Dunlavy founded a few decades ago in Australia, and is still operating.
Also see http://www.vaf.com.au who use a similar principle.
Dunlavy (and Duntech) speakers have stepped baffles, which in conjuction with the first order filters, create a linear phase response. i.e the waveform shape is preserved. Because of the steps in the baffle, severe diffraction would occur, and so heavy felt damping is applied to absorb the sound waves before they hit the edges. It also reduces diffraction with the sides of the baffle.
Take a look at www.duntech.com.au , the company John Dunlavy founded a few decades ago in Australia, and is still operating.
Also see http://www.vaf.com.au who use a similar principle.
Landroval, John Kreskovsky has papers in his website regarding the subject.
Shame that Dunlavy Audio Labs went belly-up. I wonder if Dunlavy is still active (in)directly at Duntech. Those Sovereigns must really sound good. Boy I'd like to hear Telarc's Stokowski Sound CD-80129 through a pair of those
Shame that Dunlavy Audio Labs went belly-up. I wonder if Dunlavy is still active (in)directly at Duntech. Those Sovereigns must really sound good. Boy I'd like to hear Telarc's Stokowski Sound CD-80129 through a pair of those
Landroval said:
You can analyse it in the time domain using COOLEDIT 2000 or similar with added functions. Its the acoustical response we are after. But what happens once we are in room?
An externally hosted image should be here but it was not working when we last tested it.
http://db.audioasylum.com/cgi/m.mpl?forum=hug&n=82145
http://www.triodefestival.net/members/cleach/jml_crossovers_etf04.zip
Thanks everyone. I am aware about the stepped baffles. I remember reading awhile back, before I know much about speakers, where someone implied that there was more to the felt than just diffraction reduction. That's why I asked.
Have any of you messed with felt like this? Given the inset drivers and the shape of the baffle, it seems to me that reducing diffraction would be a monumental task, with felt or without.
f4ier, Dunlavy didn't exactly go "belly up". They were sold to a larger company that was hoping to make more money with speakers by increasing production and/or productivity or something like that. The buyer got Dunlavy and then realized it was cheaper for them to just close down the factory then to put any more money into changing Dunlavy. At least that's what I heard. Dunlavy was a profitable company and didn't go out of busines by going bankrupt, as far as I know.
I found an archive of Dunlavy's newsgroup postings awhile back, here I think: http://home.austin.rr.com/tnulla/ , but it seems to be gone. Does anyone know if it's still up somewhere? Google doesn't bring up any working links for me.
Have any of you messed with felt like this? Given the inset drivers and the shape of the baffle, it seems to me that reducing diffraction would be a monumental task, with felt or without.
f4ier, Dunlavy didn't exactly go "belly up". They were sold to a larger company that was hoping to make more money with speakers by increasing production and/or productivity or something like that. The buyer got Dunlavy and then realized it was cheaper for them to just close down the factory then to put any more money into changing Dunlavy. At least that's what I heard. Dunlavy was a profitable company and didn't go out of busines by going bankrupt, as far as I know.
I found an archive of Dunlavy's newsgroup postings awhile back, here I think: http://home.austin.rr.com/tnulla/ , but it seems to be gone. Does anyone know if it's still up somewhere? Google doesn't bring up any working links for me.
just to share to may experience...
I audition a system with big Dunlavy speaker with WMTMW...
and at the same time there is a big Tannoy coaxial speaker (wider than the Dunlavy, which is about 2meters tall, and 50cm wide)
and to my amazement... Dunlavy sound a lot better than the Tannoy...
Dunlavy with multiple drivers simply disappear... (just like a single source speaker) and throwing a very spacious soundstage...
---
David
I audition a system with big Dunlavy speaker with WMTMW...
and at the same time there is a big Tannoy coaxial speaker (wider than the Dunlavy, which is about 2meters tall, and 50cm wide)
and to my amazement... Dunlavy sound a lot better than the Tannoy...
Dunlavy with multiple drivers simply disappear... (just like a single source speaker) and throwing a very spacious soundstage...
---
David
Seems to ba apopular design in Australia;
http://www.equinoxaudio.com.au/products/testimonials/i-93vsjupiter.php
I really like those Jupiters....
http://www.equinoxaudio.com.au/products/testimonials/i-93vsjupiter.php
I really like those Jupiters....
Actually, it *does* horn load it. The acoustic foam won't be very 'lossy' at high frequencies, and it *will* act like a conical horn.
An externally hosted image should be here but it was not working when we last tested it.
more measurements here: Dunlavy Audio Labs SC-I loudspeaker | Stereophile.com
You can see this in the measurements at Stereophile in two ways:
1) the tweeter response has the classic high frequency rolloff of a conical horn
2) the speakers overall sensitivity is much higher than you'd expect with such low efficiency drivers. In fact, many reviewers are surprised by how sensitive the Dunlavy speakers are.
If I had to take a guess, I'll bet that Dunlavy didn't use a 'real' waveguide because his customers would be aghast if he did. Waveguides and horns were completely out-of-style in the 90s.
While I haven't heard the Dunlavy speakers, I've found that speakers with low order crossovers, like Dunlavy, Dynaudio, Duntech, etc tend to get the midrange correct in a way that other speakers do not.
They also tend to image like a champ.
Really! never ? ?
Dunlavy, Thiele, Vandersteen, Tannoy used 1st order
..ALL image very well indeed.
Best of Breed actually.
Sadly these are all exceptionaly pricey and Audio (music as opposed To HT for Terminator Movies) is essentially on life support, except for a few diehards spending their sunset years DIYing their time away.
Nope, never heard a Dunlavy.
For the past decade or so I've been fairly obsessed with Unity horns, which are also phase-coherent. The Unity horns use a much tighter spacing than the Dunlavy, Duntech, or Dynaudio speakers. The former uses one-quarter-wavelength spacing, while the latter uses one-wavelength spacing.
For the longest time, I'd assumed that the Danley solution was the way to go, but at CES in January 2012 I heard a speaker which used a Dunlavy-type recipe. (Basically one-wavelength spacing, and first order xovers.)
I was surprised how good that could sound. While the 'sweet spot' was clearly a lot smaller, it sounded a lot like my beloved Unity projects as long as you were sitting in the sweet spot.
So that got me thinking about doing Dunlavy-type speakers, since they're much easier to implement.
I think you're jumping to an unsupportable conclusion. Felt in this configuration isn't likely to have any significant horn effects. The high frequency rolloff is essentially the response of a 1" dome tweeter when averaged across a window, roughly indicating the power response. The description of the graph at Stereophile clearly states
Actually, it *does* horn load it. The acoustic foam won't be very 'lossy' at high frequencies, and it *will* act like a conical horn.
An externally hosted image should be here but it was not working when we last tested it.
more measurements here: Dunlavy Audio Labs SC-I loudspeaker | Stereophile.com
You can see this in the measurements at Stereophile in two ways:
1) the tweeter response has the classic high frequency rolloff of a conical horn
2) the speakers overall sensitivity is much higher than you'd expect with such low efficiency drivers. In fact, many reviewers are surprised by how sensitive the Dunlavy speakers are.
The sensitivity is also not surprising. Vifa has made tweeters in that range for a long time. Dunlavy could easily also have had one made to his specs. I can also say from experience that felt configured this way will not do much for sensitivity. One could say horn loading, of a sort I suppose, it's not a flat surface. But the actual effects are primarily what they say in the article. It's for diffraction control. I've tested configurations nearly identical to that. That's all it can do effectively configured the way it is and given it's proximity to the dome.
dlr
Last edited:
An externally hosted image should be here but it was not working when we last tested it.
Yeah, the more I thought about it, the more I think my conclusion was incorrect. As I understand it, the high frequency rolloff in a horn is due to pathlength differences between the wavefront that's transmitted axially down the horn or waveguide, and the pathlength of sound that's *doesn't* travel axially down the horn.
For instance, energy off of the radiator which is 45 degrees off axis will reflect off the waveguide, and when it crosses over the central axis of the horn or waveguide, it will be delayed in time. And at high frequencies, it doesn't take much of a distance to wind up out of phase. For instance, at 13500khz a pathlength difference of just half an inch will make the radiation 180 degrees out of phase.
As frequencies get lower and lower, the energy that's reflected off the horn/waveguide sides is in-phase, and it raises the on-axis response. But at high frequencies, the effect is destructive.
But the felt complicates this, as the felt will be absorptive at high frequencies, but reflective at low frequencies. Or perhaps it's simply invisible at low frequencies! If it's absorptive at high frequencies, it would actually *raise* efficiency at high frequency, as it would reduce the destructive comb filtering.
Only trial and error could tell us what it's effect would be unfortunately.
Depends on what you call low. It's useful down to maybe 1-2K (tweeter low end) and with the right kind is most effective roughly where typical baffle diffraction is most significant.
I've done quite a lot of that.
But I never tried to create a horn with it. Dave Dal Fara (DDF on other boards) used to suggest trying to make a horn with it. The response may end up looking good on some axes, but it may be ragged on others. That will most certainly be very application specific.Dave
Afaik, John Dunlavy is RIP.
That stepped affair in the images above, imo will be more like multiple diffraction steps than absorption, at the high freqs that they likely say it is intended for...
There were some articles and threads here as well that years back discussed the ring around the tweeter. Iirc, the conclusion was that they failed to work as expected and caused some problems - as typically employed.
There was at least one speaker that used an extremely thick absorptive layer on the face of the speaker... my misfiring synapses have retrieved this information, it's veracity can not be confirmed.
The Dunlay design works *on axis* to reproduce a square wave at moderate, mid band frequencies. If you look up the B&O design in JAES VolI Speaker Anthology you'll see a similar design that does the same trick.
The weakness of those speakers is in part that 1st order xover, it puts a lot of lower freq energy into the tweeter. The tweeter will jump in IM as the power goes up...
...also may folks reported that they did not like the way the LF section performed. I did not hear them to good advantage any of the times I heard them. When I heard them at CES in the Dunlavy room, the sound was a bit un-dynamic and the bass was reticent. Others that I spoke to had similar impressions.
The idea has merit, regardless...
_-_-bear
That stepped affair in the images above, imo will be more like multiple diffraction steps than absorption, at the high freqs that they likely say it is intended for...
There were some articles and threads here as well that years back discussed the ring around the tweeter. Iirc, the conclusion was that they failed to work as expected and caused some problems - as typically employed.
There was at least one speaker that used an extremely thick absorptive layer on the face of the speaker... my misfiring synapses have retrieved this information, it's veracity can not be confirmed.
The Dunlay design works *on axis* to reproduce a square wave at moderate, mid band frequencies. If you look up the B&O design in JAES VolI Speaker Anthology you'll see a similar design that does the same trick.
The weakness of those speakers is in part that 1st order xover, it puts a lot of lower freq energy into the tweeter. The tweeter will jump in IM as the power goes up...
...also may folks reported that they did not like the way the LF section performed. I did not hear them to good advantage any of the times I heard them. When I heard them at CES in the Dunlavy room, the sound was a bit un-dynamic and the bass was reticent. Others that I spoke to had similar impressions.
The idea has merit, regardless...
_-_-bear
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.