Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

[romanbednarek]

I had saved these posts after reading them several years ago and recently uploaded them to my web server. I first posted these links at the end of a different thread but thought that I would start a new thread so that more people might see it and get a chance to read some of John Dunlavy's comments on loudspeaker design and speaker theory in general. It has been a while since I've read them myself so I don't recall everything that was discussed but I do remember that I learned quite a bit from the discussions at the time (probably around 7 years ago or so). You may or may not agree with all of the "opinions" or theories presented but I still think that there may be some topics of interests within the posts.

http://www.rjbaudio.com/Dunlavy/John...sts%201997.txt

http://www.rjbaudio.com/Dunlavy/John...sts%201998.txt

http://www.rjbaudio.com/Dunlavy/John...sts%201999.txt

http://www.rjbaudio.com/Dunlavy/John...sts%202000.txt

http://www.rjbaudio.com/Dunlavy/John...sts%202001.txt

I actually haven't heard any of Dunlavy's speakers which were released around the time of these posts but I've always wanted to. I recall that he used rather "average" Vifa drivers along with rather complex crossover topologies to achieve a transient perfect speaker design. I'm curious if any of you have auditioned or own a pair of his speakers and could comment on your impressions. I would also encourage comments regarding his theories if you have any.


Dunlavy SC-V

[Jeff B.]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

I had saved these posts after reading them several years ago and recently uploaded them to my web server. I first posted these links at the end of a different thread but thought that I would start a new thread so that more people might see it and get a chance to read some of John Dunlavy's comments on loudspeaker design and speaker theory in general. It has been a while since I've read them myself so I don't recall everything that was discussed but I do remember that I learned quite a bit from the discussions at the time (probably around 7 years ago or so). You may or may not agree with all of the "opinions" or theories presented but I still think that there may be some topics of interests within the posts.

http://www.rjbaudio.com/Dunlavy/John...sts%201997.txt

http://www.rjbaudio.com/Dunlavy/John...sts%201998.txt

http://www.rjbaudio.com/Dunlavy/John...sts%201999.txt

http://www.rjbaudio.com/Dunlavy/John...sts%202000.txt

http://www.rjbaudio.com/Dunlavy/John...sts%202001.txt

I actually haven't heard any of Dunlavy's speakers which were released around the time of these posts but I've always wanted to. I recall that he used rather "average" Vifa drivers along with rather complex crossover topologies to achieve a transient perfect speaker design. I'm curious if any of you have auditioned or own a pair of his speakers and could comment on your impressions. I would also encourage comments regarding his theories if you have any.


Dunlavy SC-V

I love reading these kind of things. However, it looks like these need to be reformatted to improve readability. If I get a chance I might pull them into a Word doc or something.
Jeff B.

[mzisserson]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

Wow. There is lots of useful info there! ALso just to listen to a "legend" speak so freely is nice. Thank you very much!

Its stuff like this that neds a home vs. getting lost in the mix...

[dlr]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

I actually haven't heard any of Dunlavy's speakers which were released around the time of these posts but I've always wanted to. I recall that he used rather "average" Vifa drivers along with rather complex crossover topologies to achieve a transient perfect speaker design. I'm curious if any of you have auditioned or own a pair of his speakers and could comment on your impressions. I would also encourage comments regarding his theories if you have any.


Dunlavy SC-V

I heard a set at Son et Vision in Montreal many years ago, either the SC-IV or SC-V. It was a small room and you had to stand way too close so the listening position was not at all where the design position was intended to be. They need a very large room. The level was also quickly irritating (to impress, I'm sure), so the top end was to strong and with poor driver integration when close and above the tweeter, it was not good. I never had the chance to hear it in a target design setting. He used 10' on the tweeter axis.

Jeff, that would be helpful. I hate trying to read text that runs on like that.

dlr

[johnnyrichards]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

Dunlavy was the man. I converted the .txt to .doc, Word 2003.

[jkrutke]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

I heard a set at Son et Vision in Montreal many years ago, either the SC-IV or SC-V. It was a small room and you had to stand way too close so the listening position was not at all where the design position was intended to be. They need a very large room. The level was also quickly irritating (to impress, I'm sure), so the top end was to strong and with poor driver integration when close and above the tweeter, it was not good. I never had the chance to hear it in a target design setting. He used 10' on the tweeter axis.

I heard a pair of these in what was likely a larger and better configured room. Listening distance must have been about 15 feet. Indeed when standing up and walking towards the front of the room, everything went to hell. But from the listening chairs, it was pretty impressive. Now, that was maybe 12-13 years ago and what I thought was impressive then might give me a headache today, just so the proper disclaimer is in place.

Dunlavy was the man. I converted the .txt to .doc, Word 2003.

Thanks for the cleanup work. I'll have to go through these and compare to what I have saved on my hard drive. Dunlavy was a truly intelligent and objective scientist rising above an audiophile world of BS and incompetent designers. He didn't mind pointing out, in a friendly way, where people were going wrong. If I recall, he passed away a few years ago. I've read a lot of his writings also and learned a lot from them.

[mzisserson]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

Thanks for the cleanup work. I'll have to go through these and compare to what I have saved on my hard drive. Dunlavy was a truly intelligent and objective scientist rising above an audiophile world of BS and incompetent designers. He didn't mind pointing out, in a friendly way, where people were going wrong. If I recall, he passed away a few years ago. I've read a lot of his writings also and learned a lot from them.

Dunlavy was extremely well respected and it is not very hard to understand why. In one of his messages he talks about a scientifically perfect speaker vs. one that someone may want to listen too. He hit it right on the head. "Audiophiles", self-proclaimed or otherwise, are artists to a degree. They are painting a subjective picture of what they think sounds good with their equipment and the like. Weather its a SET setup that is full of snake oil, or micro monitors suspended from the ceiling driven by 1KW monoblocks, what they enjoy, they enjoy.

It was obvious what he enjoyed. Loudspeaker design and audio. He has no issues at all with a speaker that was more "artsy", he just preferred to try and scientifically perfect the loudspeaker. Notice how many different complexities he talks about and admits it is nearly impossible to quantify them all?

NOT ONCE did he call anyone ignorant, foolish, idiotic, or otherwise for liking something, or believing something different. Imagine that? :rolleyes:

[nick29498141]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

I love reading these kind of things. However, it looks like these need to be reformatted to improve readability. If I get a chance I might pull them into a Word doc or something.
Jeff B.

I may run them into a PDF then load them onto my .Me server for everyone.

I had a chance to hear a pair of SC-IVs recently, driven by a pair of Levinson 33 amps and an all-digital, DSD(!), front end. They sound great, to say the least.

NK

[nick29498141]

*PDFs Loaded To iDisk For Anyone To Download*

public.me.com/pkicekat

They're justified PDFs for anyone to download.

NK

[Dennis H]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

Thanks Roman!

I've always liked the idea of the Dunlavy speakers except it's so hard to get 1st order acoustic slopes. You inspired me to go ahead and simulate something I'd been thinking about -- combine the ideas of Dunlavy and Duelund. Duelund stacks LR2 filters to get his acoustic transfer functions. Each driver has LR2 filters, either highpass or lowpass, at the same frequencies. The result is most drivers have a roll-off steeper than LR2 and all the drivers are in phase at all frequencies. So, if you did that with 1st order filters, could you get steeper roll-off in the stopband and still retain the 'transient perfect' behavior? The answer is YES! I did a small writeup over at HTG for those who are interested.

http://www.htguide.com/forum/showthr...438#post474438

[nick29498141]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

Thanks Roman!

I've always liked the idea of the Dunlavy speakers except it's so hard to get 1st order acoustic slopes. You inspired me to go ahead and simulate something I'd been thinking about -- combine the ideas of Dunlavy and Duelund. Duelund stacks LR2 filters to get his acoustic transfer functions. Each driver has LR2 filters, either highpass or lowpass, at the same frequencies. The result is most drivers have a roll-off steeper than LR2 and all the drivers are in phase at all frequencies. So, if you did that with 1st order filters, could you get steeper roll-off in the stopband and still retain the 'transient perfect' behavior? The answer is YES! I did a small writeup over at HTG for those who are interested.

http://www.htguide.com/forum/showthr...438#post474438

Good stuff!

The "Duelund Epic" was a great read for me, as well.

NK

[romanbednarek]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

Thanks Roman!

I've always liked the idea of the Dunlavy speakers except it's so hard to get 1st order acoustic slopes. You inspired me to go ahead and simulate something I'd been thinking about -- combine the ideas of Dunlavy and Duelund. Duelund stacks LR2 filters to get his acoustic transfer functions. Each driver has LR2 filters, either highpass or lowpass, at the same frequencies. The result is most drivers have a roll-off steeper than LR2 and all the drivers are in phase at all frequencies. So, if you did that with 1st order filters, could you get steeper roll-off in the stopband and still retain the 'transient perfect' behavior? The answer is YES! I did a small writeup over at HTG for those who are interested.

http://www.htguide.com/forum/showthr...438#post474438

You're very welcome Dennis (you should really thank the fellow who put them together the first time when I decided to save a copy for myself), and thanks guys for converting those files into a format that is a bit easier to read. I'll have to take some time and re-read them myself (at quick glance I noticed a few topics that have led to very long and heated arguments on this forum in the recent past, yet support my views which I chose to keep to myself).

I've also been very interested in transient perfect designs and I guess that the reason why I never built one is based on exactly what you said about the requirement of drivers that are well behaved across a wide frequency range so that you don't run into phase issues (or linear/non-linear distortion issues) at the extreme ends of the response of each driver. I would have to say that a 5 driver, WMTMW configuration seems like the best way to go even if a subwoofer has to be added to cover the lowest octave.

I recently wondered about what Dunlavy could have created if he used Scan-Speak drivers instead of the Vifas along with his cabinet and crossover design strategy. I understand that this speaker would be very difficult to market due to the increased cost of using so many expensive drivers, but I stil wonder how the combination of lower distortion drivers with this type of design would sound?

I recall that John k... had worked on a similar 5 driver WMTMW design in the past but I'm not sure what happened to it. Also there was somebody in the past 5 years (Bob C?) who made a rather small transient perfect speaker using three drivers (TMW - Morel?) in a sloped baffle, bookshelf sized cabinet.

In general I think that there are two reason why transient perfect speakers are pretty rare in the DIY community which include the difficulties in designing the crossover for such a speaker as well as the sacrifices that need to be made in terms of being limited to a certain subset of drivers that have a good chance of working in this type of configuration.



If any of you would like me to host either of the modified Dunlavy files under a link in the "Audiofiles" section of my website just let me know (I have plenty of server disk space and bandwidth) but I guess if I was really ambitious I could just put them into HTML format so they could be read from my site without having to be downloaded.

[nick29498141]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

If any of you would like me to host either of the modified Dunlavy files under a link in the "Audiofiles" section of my website just let me know (I have plenty of server disk space and bandwidth) but I guess if I was really ambitious I could just put them into HTML format so they could be read from my site without having to be downloaded.

Hey Roman, my iDisk should be up for another year, or two.

NK

[johnnyrichards]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

Here you go, for those without Word or who can handle large amounts of text on a website.

[Dennis H]

Re: Archives: John Dunlavy (rec.audio.high-end forum posts 1997-2001)

I would have to say that a 5 driver, WMTMW configuration seems like the best way to go even if a subwoofer has to be added to cover the lowest octave.

......

Also there was somebody in the past 5 years (Bob C?) who made a rather small transient perfect speaker using three drivers (TMW - Morel?) in a sloped baffle, bookshelf sized cabinet.

Yeah, I remember Bob Cordell's speaker where he showed that rolling off steeper than 1st order in the stopband wasn't necessarily bad. What the Duelund/Dunlavy concept does is quantify that a bit so you can generate some target curves. If you were going to add a sub to get down to 20Hz, you'd ideally want to use the same concept on the sub so you'd end up with a Duelund/Dunlavy 4-way. 4 drivers strikes me as the fewest that would be practical to cover 20-20K with such shallow initial slopes.