John Larsen, the designer of these intriguing and unusual speakers, worked with the late Stig Carlsson for sixteen years until Carlsson’s death, and the Larsen Model 8s are in a sense a continuation of the ideas that arose from their work together. But there are new features in the design compared to the Sonabs of yesterday, the speakers which brought Carlsson’s work to the USA in the 1970s and made him well known as a theoretician and designer.
Words like “unforgettable” can sometimes be flung around a bit, but in the case of the Sonabs this is the literal truth. I not only recall the specific occasion when I first heard them (at someone’s house), but also the recording I heard (namely Handel’s Rinaldo on Columbia Masterworks M334592)—the impression they made was that vivid. To say that the occasion was memorable is an understatement!
Naturally I was interested when it came to my attention that John Larsen had continued the development of this type of speaker. After a convincing demonstration at T.H.E. Show Newport 2014, a review pair was on the way.
Not to maintain artificial suspense, the Larsen 8s are, like their ancestors, speakers that produce an unusually natural sound, a sound that bears a surprising resemblance to concert music. But, again like their ancestors, they are quite different from ordinary speakers—by intention. So to appreciate them, one needs to think for a moment about certain fundamental matters of speakers in rooms.
What Gives With Speakers
It is, of course, familiar stuff in audio that speakers can have various basic radiation patterns: omni, dipoles, or that intrinsically strange if most common hybrid, the “forward-radiating” box speaker designed to be out in the room, which is typically omni in the bass, shifts to radiating primarily forward (half-space radiation) somewhere around 300 to 700Hz, depending on the width of the front baffle, and then narrows its pattern yet more in the high frequencies. On the face of it, this most common kind of speaker seems, like the chimera, to be led from ill-matching parts, as far as radiation pattern goes.
Designers have, of course, become adept at making these transitions of pattern occur smoothly, and thus sound less troublesome to the ear. But the forward-radiating box out in the room remains an odd creature on the face of it.
And then there is the question of “free space mounting,” the “out in the room” part. Since bass works better if the speaker is up against a room boundary (or two or three), how did it get to be a habit to put the speaker out in the room—and to have its woofer off the floor, usually? The answer is historically complex. But roughly speaking, you can blame it on the British. Free space mounting makes it easiest to avoid boundary-induced coloration of the midrange, and the midrange was most of what the British audio establishment of the 1960s and 1970s was interested in. Midrange and precision stereo imaging is also easiest with free space mounting, which delays reflections for a long time.
What To Do About It?
In other places, though—back in the USA and also in continental Europe—people were thinking hard about how advantageous it was for bass to have the speaker up against a wall or two, and to have the bass driver on the floor, as well. Allison in the USA promoted this. And so did Stig Carlsson in Europe.
Carlsson in fact began a systematic rethinking of how speakers interacted with rooms and how sound ought to be radiated into the room. The placement against the wall was advantageous not only for the bass but also for ideal stereo. One needs room sound in stereo—anechoic stereo sounds odd and unconvincing. But the sound off the back wall, from that part of the speaker’s sound that came around behind (a lot in the bass and less as frequency rose), arrived back at the listening position very early, if one had the speakers anything but a long distance from the back wall. And that secondary arrival was not really useful in conveying information about the original recording venue. So went the thinking, as I understand it.
So the natural thing to do was to eliminate the back-wall reflection by placing the speaker against the wall so that the sound off the wall formed in effect a unit with the forward sound. Room sound then arrives a long time after, especially if one puts the speakers on the long wall so that they are far from the sidewalls.
Another aspect of the situation is that the directivity of the speaker ought, in one theory anyway, to remain essentially constant. The up-against-the-wall placement eliminates the baffle step (omni-to-forward) shift in directivity since the speaker is forward radiating all the way down. (The shift to the involvement of the floor is at a quite low frequency and is less consequential in effect than the usual baffle step in the midrange.) In an attempt to achieve constant directivity, the Larsen speaker has a much wider treble dispersion, the result of the upward-firing tweeter. This aim at constant directivity includes in the Larsen speakers adding extra tweeters, not firing directly at the listener, to fill in the power response in the high frequencies, where most speakers become quite directional.
What Is In The Larsen Speaker?
The Larsen Model 8 is a relatively small speaker, which sits on the floor against the wall. Domestic compatibility is certain: It is elegant looking in an understated way and it is not underfoot. But, of course, appearance is not the point of the wall placement, as noted. The speaker contains a side-firing driver near the floor, a midrange/bass driver which fires forward, up a little and inward pointing (the speakers are mirror imaged). There is a main tweeter mounted above the midrange in the angle formed by two surfaces. And two auxiliary tweeters, each attenuated by 10dB, fire upwards from the top surface in front of the midrange driver. (The picture is worth quite a few words here.)