double ar 3a's

[ironlake]

If I want to try double ar 3a's will it work to wire them in series so impedence will will 8 ohm to the amp?. I assume parallel with total impedence of 2.ohms would be too low for a solid state amp. Has anybody ever tried this?

[ra.ra]

This is a great question and very much in line with a set-up that I have been thinking about. Certainly ironlake deserves helpful response(s) to his (her?) specific situation with the two sets of 4-ohm 3a's, but I'd like to expand this question in a more general way regarding multiple sets of speakers and various wiring configurations.

The info and resources shared on this site are fantastic, but I'll admit to being in over my head very quickly when the talk turns overly techie. Electronics-wise, I cannot say I truly understand the complimentary workings of capacitors, resistors, magnets, potentiometers, or coils. I have similar experiences with much of the more advanced acoustics chatter, but I read and try to learn nonetheless.

Similar to the OP, the thought I've been pondering is whether four (4) pairs of 8-ohm AR classics could be wired in series-parallel in order to deliver a resulting 8 ohms to the amplifier? Assuming this wiring would do no harm and with these speakers being rather inefficient, what would this mean in terms of amplifier power required to (minimally but) adequately drive eight (8) cabinets with 16 or 20 drivers?

[soundminded]

This scheme is not advisable although there probably isn't any risk of damage to your amplifier. The voltages are not likely to divide uniformly at all frequencies if for no other reason than differences in the resistances of the potentiometers will make each speaker's complex impedance different from the others. It will also vary due to other seemingly minor differences such as different actual capacitance values. The effective amplifier damping factor will be below one for each of them. The best method is to use multiple amplfiers, one amplifier channel per speaker. If you have two pairs this means two stereo amplifiers or four monoblocks. Adjust the settings one pair at a time to try to get them to match each other as closely as possible.

[tysontom]

Stacking or "double" AR-3as makes for a great setup. Aside from the low-impedance issue, it is important to consider the way in which you mount the speakers. The best method (having two-AR-3as-per-channel) is to stack them vertically on top of one another, such that the woofer on one speaker is closest to the floor (but at least a foot off the floor), and the other AR-3a's woofer is closest to the ceiling. This therefore puts the tweeters and midrange drivers closer together, but in a "MTM" arrangement. This "MTM" arrangement is midrange-tweeter-midrange, such as espoused by speaker engineer Joseph D'Appolito several years ago with his Swan speaker, and it has had a wide following since that time. Having this arrangement creates a line source to some degree and suppresses the vertical dispersion somewhat while allowing horizontal dispersion to continue in the normal fashion. This is considered good practice, and reduces ceiling and floor reflection from the midrange/tweeter drivers, thus improving intelligiblilty and "focus" for those interested in imaging. Back in the far field, however, it really doesn't make that much difference, so it's just part of the plan.

Having stacked speakers also roughly doubles the power-handling capability, so the speakers can therefore be driven harder and louder without creating more stress and distortion, than with a single pair. Because of "mutual-radiation impedance," the output of one woofer will augment that of the other woofer to some degree (but not nearly as much as with mounting two AR-3as side-by-side), but this setup does *not* extend the actual deep-bass capability, yet it does increase by 2-3dB the output *below* the woofers' radiation impedance (for the AR-3a this would be from the 575 Hz crossover on down) all the way down to the 43 Hz resonance frequency, relative to midrange frequencies. What this means is that a stacked pair of AR-3as will sound "heavier" and "punchier" than a single pair, mainly because the bass is "turned up" relative to the midrange, due to this lovely phenomenon called mutual-radiation impedance. But since the resonance frequency remains unchanged, the bass below resonance is not changed. To be "blown" completely out of the room, try mounting "stacked" AR-3as in each corner! No, no, just kidding -- you *don't* want to do this, as the bass is overpowering and too heavy.

To handle the low-impedance issue -- which for the stacked AR-3a pairs-per-channel would be around 2 ohms on each channel -- you should use separate amplifiers for each pair, as Soundminded suggested, or you could employ just one of the Crown or QSC professional power amps. These beautiful brutes are extremely clean and effortless, and they are specifically designed to operate into loads down to 2 ohms -- all day long. The newest breed of amps are the class D or class H multi-tiered amps that are very small and lightweight, yet put out prodigious power without stress. Good examples of this high power are the new Crown XLS-1000 or XLS-1500 Class D amps (about 350+ watts/channel into 4 ohms or 500 watts into 2 ohms). These amps have fans, but the fan on the XLS amps seldom runs unless the amp is subjected to very high output levels and current demands. The XLS-1000 new is $299.00, so the price is very reasonable, and these amps are largely indesctructable.

I have attached some images of the Crown and QSC professional amps for examples. Also, I've attached an article from Harry Pierce of *The Absolute Sound* about stacking Advent speakers. The article is full of the typical nonsense one associates (coming from) this writer, but you can get a sense of some of the advantages of stacking speakers.

--Tom Tyson

Double_Advent_System.pdf

[ar_pro]

I actually owned and operated two pairs of final-production AR-3a systems in the late '70s.

Having been aware of the "Double Advent" system, I was interested in achieving the same sort of "punchier" sound that Tom has mentioned.

The initial postioning was a vertical stack, with woofers at the top & bottom.

Based upon the Advent paradigm, I knew this to be the "correct" setup, but in fact, it was surprisingly underwhelming.

It was only when the systems were stacked horizontally, with the woofers aligned, and at the outside left & right positions that the speakers sounded significantly better than a single pair of 3a's.

I'd purchased two pairs of the then-new steel support stands that AR was offering with the AR-10pi and AR-11 speakers, and positioned them side-by-side, so that the stacked 3a's could lie horizontally across one pair. The dimensions worked out perfectly - the height was excellent, and the stacked systems almost had a custom look to them. I wish that I'd taken photos.

My set-up used a pair of stereo McIntosh MC2205 amplifiers, which did the job nicely - the two ohm autoformer taps on a single amplifier could run all 4 speakers, but the performance was much better with a pair of amplifiers.

I did experiment with a series connection to a single stereo amplifier, but the result was laughably awful.

Tom's observation regarding bass ouput is correct - there was greater perceived bass, but not an increased extension.

The AR-9 was released within the next year or so, and the improvement in extension and output was profound enough to warrant selling the 3a's and buying my first pair of 9's.

That said, the stacked, twin-amped, 3a system was the best I'd ever heard from AR up to that point.

[tysontom]

ar_pro's AR-3a setup with the speakers mounted horizontally one on top of the other would *definitely* maximize the "mutual radiation impedance" mentioned in my previous post by about 3 dB in the region below the frequency of ultimate radiation resistance, which in the 3a would be below about 575 Hz. Therefore all bass frequencies would be elevated slightly going down in frequency to the point of system resonance, at which point the enhancement ends. This makes for a punchier sound, but still a clean sound. Few speakers could approach this level of bass power before the AR-9!

AR did a similar thing for Harvard Medical School to get the ascultation of the human heart and reproduce it to a large group of medical students. Four AR-1x systems were wired together, powered by an AR Amplifier and placed in the corner of a large auditorium/classroom area. By mounting the four woofers close together -- and in a corner -- the bass response was greatly amplified and worked perfectly for this application. AR took advantage of this opportunity to make an excellent, if typically understated, AR advertisement (see attached).

--Tom Tyson

AR_AR-3a_Ad_Heart-Sounds-Ascultation_Harvard_c1970.pdf

[ironlake]

I have come up with in my opinion a super sounding system. I have my 4 ohm ar3a,s wired in series with my 8 ohm advent heritages which are superior sounding speakers on their own. due to height of the advents stacking them is out of the question but my big jcv reference amp just loafs along at rather loud settings with the clearest sound I have ever heard. The advents make up for the very high frrequencies not strong on the ar;s. Being a band director I am used to listening to real music at full volumes every day so now with this system I can have that level with no strain on the speakers.

The review by julilan h on the advents is right up there with the 3a,s. So putting these two speakers together on each channel for me is a real sound experience. Total impedaince measures 12 ohms dc resistance on my fluke and would guess similar to the amp.

105 db sound levels and the amp does not even get hot. I have measured a full 120 db in the band room with the full 80 piece band playing pop music, so for me to like it I need the level. Most of you would not like this level of sound but at 46 years of teaching it is what I am used to.

On the other side when I am just laying back in a mellow mood and have some brubeck etc on soft Desmonds alto sax is so super smooth with this speaker system.

[soundminded]

Having two pairs of AR3as allows for some interesting experimentation with placement. Were I in such a fortunate circumstance the first thing I'd try is placing each pair side by side towing the outer speakers outward and the inner speakers inward in an arrangement similar to LST. This would further increase AR3a's already excellent lateral dispersion. Then I'd try them with the inner pair toed out and the outer pair toed in which should have a similar effect except you'd never be far off axis of at least one midrange and tweeter for each channel. Then I'd pull them away from the walls and try them as a direct/reflecting pair. If you have four separate channels of amplification and at least one equalizer you can experiment with the front facing and backward facing pair at different loudnesses and equalizations to copensate for the absorption/reflection properties of the reflecting wall. You could angle the backwards facing pair toed in or out to determine how different dispersions affect the overall sound field.

Two pairs of AR3as will not achieve the same low frequency reach as a single pair of AR9s, they are tuned differently but they will double the power handling capacity to the point where any perceived loss of bass by pulling the speakers away from the walls and rasing them on stands could be compensated for with equalization without overdriving them (not easy to do in any event.)

[tysontom]

Soundminded, you are so right that two AR-3as/ch cannot equal the bass output of a single AR9/ch; however, two AR-3as/ch with an equalizer such as the Allison Acoustics ESW (I have one in my office system) provide absolutely flat output down to 20 Hz by providing up to 12 dB/octave compensation below resonance, matching exactly the standard 12 dB/octave rolloff in frequency below resonance. The ESW does require a powerful amplifier (much like the needs for the Bose 901) to apply the necessary energy to move the woofers all the harder below 40 Hz. Taking two AR-3as/ch -- and using said ESW (or similar) equalizer -- will give greater deep-bass output when compared with a standard unequalized AR9s per channel, simply because the AR9 is not flat to 20 but rather 28 Hz. The AR-3as would definitely have to be close to the floor or floor-wall boundary -- or even close to a corner -- to best the powerful AR9, and even then there would be the dreaded boundary-interference effect, also known as the Allison Effect, whereby there is a notch in the low-frequency output caused by a frequency of cancellation. Notice that the AR9 and the AR9LS and other tower speakers do have the woofers close to the floor-wall boundary to eliminate this so-called "Allison-Effect" cancellation notch (or "suck out"), and overall the AR9 would be flatter in the bass down at least to the 3dB point of 28 Hz.

By the way, for those interested in getting even more deep-bass output from their AR-3as, AR-10s, AR-11s, AR-LSTs, etc., and also the 10-inch and 8-inch AR speakers, the Allison ESW has approximate contours for those speakers to extend the low frequencies. The ESW was originally designed specifically for the Allison: One, Two and Four systems, which correspond closely with the AR-3a, AR-5 and AR-4x in bass resonance. There is one caveat: using an accurate and flat equalizer will not give a sense of more bass output except for the very low percentage of music that resides in this nether region -- organ and synthesizer music -- so about 99% of the time one doesn't even notice the effect of the thing. But when there is good organ music, etc., the ESW definitely makes a subtle difference. The ESW was not made in large numbers, and the instrument has been discontinued for over twenty years, so used ones are difficult to locate.

--Tom Tyson

[ar_pro]

I've heard of the Allison ESW but never experienced one, first-hand.

I believe that there was some criticism of the ESW regarding its alleged ability to destroy woofers when used improperly, but it came from one of the audiophile magazines, so I took it with a grain of salt.

At one point, I was vertically bi-amping the AR-9, and using the Audio Control Richter Scale equalizer on the LF portion of the speakers.

At the time, my system had a passive preamp, feeding a pair of McIntosh power amplifiers, and it was very easy to insert the Audio Control EQ into that set-up.

The Richter Scale provided five bands of control from 22.5 HZ up to 125 Hz, and its effect was subtle & natural, giving the 9 just a tad bit more authority on the lowest notes.

[tysontom]

ar_pro,

The ESW should be used with extreme caution when driving some types of speakers such as ported or transmission-line designs that have no acoustic-resistive element to attenuate the movement of the cone at subsonic frequencies. As you know, the entrapped air in the cabinet of a sealed cabinet restricts the movement of the cone to prevent over-excursion at subsonic frequencies. The ESW does have a <20 Hz subsonic filter that was designed originally to prevent high power into very low frequencies, such as when a tone arm was accidentally dropped on a record, etc.

With a vented (bass-reflex) speaker, there is nothing -- other than the mechanical suspension of the speaker itself -- to prevent excursions from exceeding limits, and with an equalizer, this danger is greatly magnified. Considering that 12 dB is more than a fifteen-times power gain, an input of just ten watts at 40 Hz will result in a power of about 158 watts at 20 Hz, enough to rough-up some speakers. This may be the origin of this "woofer-destroyer" rumor, but Allison specifically warns against using the ESW with bass-reflex designs for this reason. The AR 12-inch woofer, however, can easily handle the ESW when driven reasonably.

--Tom Tyson

[dynaco_dan]

If I want to try double ar 3a's will it work to wire them in series so impedence will will 8 ohm to the amp?. I assume parallel with total impedence of 2.ohms would be too low for a solid state amp. Has anybody ever tried this?

Hi there

Here is a Dynaco schematics for multiple speaker systems.

It is also posted recently in the Dynaco section under, "Dynaco Dynamax" and was originally posted by Dynacophil, thank you again, Phil.

Be sure to see the total load before turning on the amplifier.

The drawings are primarily for Dynaco classic speakers rated at 8 ohms but should apply to most all equivalent brands.

Dynamax_small.pdf

[volvo245ls]

Really appreciate the pdf of the TAS Double Advent Report.

Thanks for providing a professional dialogue on one of the most famous "pairings" in the history of Audio High Fidelity!

[samberger0357]

What about stacking a pair of 3's with a pair of 3a's? Should one expect a similar positive outcome as 2 pairs of 3a's, or will the differences between the two potentially negate that?

[samberger0357]

So even though there's another thread going about stacking, I would welcome thoughts on my previous post re: stacking a pair of 3's with a pair of 3a's. Obviously these are not the same speakers, and so I would welcome those that are more technically minded then me(that would be almost everyone here) to comment on what they think the potential positives/negatives of such an arrangement might be. Yes, stacking them does sound good to me, but while that's the most important thing, I'm also here to learn. And while I haven't tried it yet, what do folks think about stacking a 3/3a, vs stacking a pair of 3's, and a pair of 3a's?

 

Somethings to ponder over Saturday morning coffee.