2-way loudspeakers

August 17, 2016
 by Paul McGowan

The 2-way loudspeaker is the cornerstone of audio. A tweeter and a woofer cover the full spectrum of sound.

From the beginning, 2-way speakers had crossovers separating highs from lows.

I am no history buff, as Copper Magazine's editor Bill Leebens is, but my best recollection of an original multi-driver loudspeaker would include the Altec Lansing A-7 Voice of the the Theater.


Horns, yes, but a tweeter and woofer, all the same. I am sure history buffs will point out others I have missed, but an accurate history misses the point of this post. My aim is to point out the how and why of these developments.

The first 2-way loudspeaker I ever heard was one shared by many of my readers.



The 1953 AR-1 by Edgar Villchur and his student, Henry Kloss. My father, ever the hi fi buff, had a pair in the garage that I eventually acquired. Much to his horror, I pulled out the drivers and made them into my own loudspeaker creation, but that's another story.

Though the two drivers both look like woofers, this was Villchur and Kloss' two-way loudspeaker that hasn't much changed since these early days.

Tomorrow I'll show you the schematic for this speaker and explain its operation.

Subscribe to Paul's Posts

19 comments on “2-way loudspeakers”

  1. The crucial points with multi-way speakers are the crossover regions and the phase shift between the drivers. What was the crossover point for the AR-1? The problems become most apparent when a single instrument or a voice is reproduced. BBC monitor speakers were said to be designed covering the frequency range of the human voice with a single driver. Is that true?

    1. @paulsquirrel

      I'm not sure the cutoff frequency of the BBC Monitor , there are so many models of it.

      But the great R. Sequerra in their latest versions of the famous small monitor: MET 7 uses as a cutoff : 3.6 KHz.

      Perhaps this is the reason for the voice , particularly the sopranos are heard more "human" .

      I hope this will say something to you.

  2. The AR-1 is considered the Holy Grail for avid seekers and lovers of vintage speakers. Much more than the AR -3a and Altec A7 .

    They are very scarce and its price is high when you can find them in good condition.

    1. Among collectors of AR speakers, their ultimate quest is for AR9. It had by far the best bass of any speaker system AR produced. It had a different manufacturing variant as the same 12" driver all AR speakers used and are considered interchangeable. AR9 uses two 12" woofers per channel in a double sized cabinet and a very tricky crossover design. It also uses a tricky method of stuffing having two resonant frequencies, one well above its crossover point. it also has a clever method of integrating the subwoofers with the lower midrange and coupling the subwoofers to the room. It does not have the high frequency dispersion of LST or even of AR3a.

      1. The appreciation of audio playback , is subjective and I respect how the most, your penchant for AR -9 , I remember that in the golden age of a well-known magazine , referring to the woofers of AR-- 9 reads: "have the inertia of a Mack truck."

        I was able to hear , long before reading the article, and I did not find pleasing to the ear the general character of this model, mainly the higher frecuencies.
        This was subjective and without the influence of technical design arguments that could have altered my hearing assessment.

        I'm not sure if this model was designed by E. Villchur , maybe you can confirm this.

        1. Out of the box AR9 was not a good sounding speaker to my ears but then none of them are. It took me several years to figure out why and how to improve it. They were purchased in 1984 and underwent their first improvement around 1989. They underwent a further improvement around 2006. The ability to adjust their FR with an equalizer gives them enormous flexibility but that alone was not enough. They do not meet my own design criteria but they are pretty good for a mostly direct firing speaker when properly adjusted.

          1. Your comment about the poor sound quality of the AR -9 ( stock unit) relieves me of any further comment.

            It is interesting to read again, your posts in this forum .

  3. Yesterday I commented on why one way systems can't cover the full audible spectrum which for most adults with unimpaired hearing is nearly ten octaves until they are old. I said that by their nature their requirements for reproducing the frequency extremes are mutually exclusive having opposite requirements. Two way systems fail in this respect for the same reason. It took extraordinary effort to the point of the absurd to successfully get the one way 901 system with 9 CTS acoustic suspension 4" drivers per channel to perform over the entire range as a two way system as an experimental device, not as a commercially viable product. Add a subwoofer to a two way system and you have a three way system. Ironically they also usually fail for the same reason. The useful range of each driver is restricted to the point where they cannot seamlessly cover the entire spectrum without substantial added efforts. Equalization whether passive such as in complex crossover networks or active through the use of graphic equalizers, parametric equalizers and even electronic crossovers and multiple amplifiers can help within the limits of power handling capacity on the low end and inertial mass and dispersion on the high end. Therefor the 4" CTS driver cannot produce the top octave no matter how much power you pump into it and can only produce adequate bass to very low frequencies through use of a large number of drivers and massive equalization of 30 db at 30 hz. 30 db means 1000 times as much power.

    Acoustic Research grappled with the problem of the 3 way speaker for decades failing every single time. For example, AR3 extended the woofer's range to 1 khz where it performed poorly. With an improved midrange dome the crossover was lowered to 525 hz where the midrange performed poorly. Eventually after nearly two decades AR threw in the towel and created a 4 way system lowering the LF crossover point to 200 hz and adding a lower midrange driver. Through equalization the speaker can have any desired FR across the frequency spectrum but the design of AR9 fails for other reasons.

    Infinity IRS overcomes the relatively limited output at the low end of the EMIM's range by providing a multiplicity of them to increase aggregate power handling. But IRS Beta with far less capability in the low end range of the EMIM requires a fourth lower midrange driver. Same solution as AR9 and for the same reason.

    Here's the Western Electric 1928 two way horn speaker.
    Western Electric was a division of AT&T. Somewhere in the recess of my memory Altec Lansing was originally associated with Western Electric. BTW, AR1 was a breakthrough design intuitively created by Edgar Villchur in 1954. Over 60 years later it still gives the best out there a run for their money and bests many if not most. Doesn't say much for the industry. AR1 woofer + KLH Model 9 electrostatic speakers = Infinity Servo Static? Well almost. The resonant frequency of AR1 and its successors was 42 hz in its enclosure and down 3 db at 35 hz. AR9's resonance is at 28 hz and can easily be equalized to well below the audible range, something that is not necessarily always desirable. The open loop response of IRS going back to Bascom King's explanation of servo feedback control was 60 hz. Sub optimal indicating the full potential of the acoustic suspension concept had not been realized and therefore required electronic correction to reach the bottom end of the lowest octave. You can think of negative feedback (servo control) in this sense as automatic equalization boosting power at the lowest frequencies to compensate for the FR falloff.

    1. A couple of comments/questions.

      First, was the original 901 acoustic suspension or merely closed box since there is a difference which seems to be forgotten in these days of ported boxes. Acoustic suspenstion is a proper subset of closed box.

      Secondly, I guess I was surprised that the IRS resonance was 60 Hz and then equalized to go lower. But there is a version of closed box in these days of very powerful amps where the bass driver is driven below resonance, in fact only used below resonance, because it gives very tight control of overhang. My 1st knowledge of this was the McIntosh speakers of the 70s. And both the Pipedreams and Scaeana speakers later. I know the original Scaeanas had a box resonance of 180 Hz and yet were crossed over with a steep slope well over an octave below that. And I once heard the most amazing bass response from them at Harry Pearson's where even a bass freak like Harry had bulging eyes for the performance.

      1. Original Bose 901and series II were true acoustic suspension speakers with long throw drivers. Among the design flaws, the resonance frequency of the system appears to be in the vicinity of 250 hz and is underdamped with a 7 db peak. The equalizer boost frequencies in the bass 6 db per octave but AS speakers fall off at12 db per octave. The equalization is therefore insufficient to produce the lowest bass audibly unless played at very loud volume. No highs, no lows, it's Bose. There is truth to that statement but they can be salvaged to make them into an excellent speaker. It was much cheaper and easier than starting from scratch.

      2. The IRS was not equalized in the conventional sense. The servo system compares the output of a transducer with the input voltage and makes a correction to compensate for the discrepancy. In this case the discrepancy is bass falloff below 60 hz. This is one reason why very high powered amplifiers are used. Additional equalization to alter the input signal can be added before the amplifier if desired.

        I have a pair of McIntosh ML-1Cs that need new surrounds for the woofer and lower midranges. The equalizer was sold separately. Without equalization its bass output falls off rapidly. The tweeter is not very good. The upper midrange is the same dome AR used for many of its TOTL speakers. Saul Marantz marketed AR LST under the Cello brand name.

  4. I was started on the Boston sound. My father had KLH 5's and my first speakers were KLH 32's. My friend's fathers all had various AR models. My favorite was the LST, an Allison design. When I was a few years older, I found a used pair of B&W DM5s that I still have today. Sealed two way with Qts=.65, the Euro equivalent of Villchur/Kloss/Allison.

    I converted to subs/sats after losing my DM6s, electronic crossover is really superior for the lower transition in a 3 way system and separate boxes allows for better acoustic integration with the room.

    1. How low do you set the transition between subwoofer and satellites? Are the subs designed for down firing or for producing an initial sound wave that emanates from the whole subwoofer baffle? Coax for the satellites?

      1. My subs are a variety of conventional designs, and one based on the Tymphany LAT-700. This is a novel driver with two vertical opposing motors each with multiple diaphragms. It achieves about 20dB of vibration cancelling, so is useful for combating resonances and spurious transmission/re-radiation in the structure on which it sits. This is the only vented box among seven.

        I have no coaxial drivers in my collection of 250.

      2. I use mostly electronic crossovers - I have a Velodyne sub with a built in plate amp, a Snell electronic crossover and a B&K Ref 30. For my custom stage speakers I use Behringer DCX 2496.

        Transition frequency varies form 80Hz to 400Hz depending on what I am doing. Some of my 15" "subs" are flat to 5KHz!

        My bedroom TV has a pair of EAW UB12's with a passive EAW SB48 designed for them. I bought them on eBay for a total of $400. Why have "Home Theater" speakers when you can have real theater speakers for less!

    2. I've always recalled the term that John Crabbe, the editor of Hi Fi News used to describe the imaging of the AR LST in his review. He called it a sonic wodge of sound, probably because of the way the multiple mids and tweeters combined their radiation patterns.

      I lived with a set of early DM6s for many years. I recall the 1st time I heard them. My friend and I were astounded by the clarity and focus which was spectacular for the day. Both of us were IMF owners at the time.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram