Horns and Time Delays

[Bill Fitzpatrick]

Even though I can appreciate their positive attributes, I am not a big fan of horns nor do I know anything of the theory beyond the basics.

What I would like to discuss here is what I consider a fundamental flaw in the way that most people use horns particularly in 3 way systems where the woofer may be in a simple vented box with a flush mounted horn mid and a flush mounted horn tweeter.

The basic problem is that the voice coils are not in the same plane. As a result, there must be some fairly audible time delays involved regarding the arrival time at the listeners ears.

Particularly problematical I would think, at least on transient information such as a castinet for example, is the difference in arrival time between the mid-range and tweeter which might be on the order of 2ms.

At the minimum, this has got to smear the sound. At the worst, I would think you might hear an echo - two castinets rather than one.

Somewhat less delay would exist between the woofer and mid.

How can someone listen to and be happy with this terrible state of driver mis-alignment?

 

[Elso Kwak]

Time delay/difference

Hi Bill,
25 years ago I unscrewed the tweeters from the frontboard of my Klipschorns and placed them all the way way back in the corner. Also modified the crossover filter .
The backwards placing of the tweeters is not a day and night difference but I prefer it that way.
The crossover mods are, especially at high listening levels.

 

[roddyama]

Hi Bill,

Sorry, weren't you expecting me?

How can someone listen to and be happy with this terrible state of driver mis-alignment?

First, an answer to your question. Easily; I do it every chance I get, and I am, for the most part, happy with what I hear. It has to do with the positive attributes of horns you speak of. The dynamics and the lack of IM in good horns produce a sound that the vast majority of cone speakers could never hope to achieve. Only the best of the cone speakers can close to matching the dynamics or IM levels of horns, but only at modest volumes. Once you raise the volume, the cone's sound becomes strained and compressed. It's a trade-off I've been happy with for 25 years.

Now to address your discussion. I agree that this fundamental flaw exists. I can tell you it is in fact audible, but you don't get an echo. What you get is a loss of coherence. The details are there, but slightly altered from what we know to be correct. For example, the "tick-splash" of a drumstick striking a cymbal face or the "tick-ring" of hitting the rim of the cymbal is reproduced out of sequence. It is audible, and I have heard this in my system. Although in the context of the music, it is not "terrible" or even annoying. It does, however, take away from the impact in the music. The initial attack that is discussed in this thread:
http://www.diyaudio.com/forums/showthread.php?postid=56712#post56712

There was a time when my system configuration was such that could align the voice coils of the bass, mid bass, and mid horn in relation to my listening position. I did this again, temporarily in response to this thread to verify what I heard. It is readily apparent that the coherence and the impact are available by doing this. However, it is not profound enough over come the hassle of the SAF. I will eventually implement a crossover with a time delay.

Lets see... analog or digital. It's only time and money.
Rodd Yamas***a