John Dunlavy Interview

[Deang]

I was doing some digging last night and ran across this interview -- it's good.

http://www.stereophile.com/interviews/163/index.html

Edited December 27, 2013 by DeanG

[Don Richard]

That was an interesting read. I see that Dunlavy designs his speakers using measurements and then does the final adjustments by ear. He ranks step response and impulse response as the most important measurements relative to how the speaker actually sounds. His opinions on the polar response of loudspeakers line up with my own observations. There is a lot of good information in that interview.

[Pete H]

I'm curious after reading the article (being a total novice) what Dean, Al, Bob and others think regarding the type of crossover that should be used to get the most accurate reproduction of sound. Seems like he believed if it couldn't be done with a first order crossover then it really couldn't be done properly, or am I misunderstanding something? I'm not trying to offend anyone, but trying to digest the information, which seems to go against the various extreme slope networks that are being used.

Someone break out the box of crayons so I can grasp this correctly.

[Don Richard]

Dunlavy seems to think active crossovers are the way to go for ultimate sound reproduction. If he is correct in saying step response correlates most accurately with what we hear, then a digital processor is by far the best option for a Khorn. Because of the path length differences between the drivers the impulse response of a Khorn is awful. I don't see any way to correct these sorts of anomalies with any passive crossover without audible ill effects. The same goes for most Klipsch Heritage designs, or for that matter, any speaker that is constructed such that the cones and diaphragms are not properly aligned.

[Quiet_Hollow]

Thank you for posting Dean. Good read.

If he is correct in saying step response correlates most accurately with what we hear...

This has certainly been in-line with my experience, but one also has to keep in mind that an amplifier / driver assemblage intended for audio playback is a band-pass filter.

IOW, we have to keep in mind exactly what a step response input through a good, real-world, band pass filter looks like....it's not a "square wave" and never will be. Speakers don't have infinite bandwidth. For that matter, neither do our senses.

I balk at designers that feel compelled otherwise.

Aligning drivers quickly becomes a packaging issue, which drastically affects cost where horns are concerned.

Suffice to say, anytime I've spec'd out a system, paying particular attention to absolute distortion and system-wide phase response has yielded the best audible results.