Jump to content

AR 78ls


Sonnar

Recommended Posts

post-116698-0-73864400-1406743374_thumb. Seen on ebay.it for 299,00 euros; and at the same time and the same price a pair of AR 58s ( the ones with same domes mid an tweeter of AR 9). Someone knows how sounds these speakers? Thank You, Adriano
Link to comment
Share on other sites

Hi Adriano,

The 78LS was part of the AR family that used the "dual dome" midrange-tweeter assembly in the early '80's. AR's research showed that the AR9 suffered from some vertical dispersion inconsistencies due to the spacing apart of the mid and tweeter domes. If the two radiators could be brought to within a very small distance apart at the crossover frequency (I believe it is a half-wavelength), then they will behave as if they are one radiator, like a "point source," resulting in essentially perfectly uniform, interference-free dispersion in both the horizontal and vertical planes.

The difficulty, especially in the early 1980's, was that the large ceramic magnet structures of the midrange and tweeter limited how physically close you could get those drivers. Maybe within three inches, that was it.

So AR decided to mount both the tweeter and midrange dome/voice coil assemblies very close to each other, within one common magnet structure.

Very clever.

They did this for the 9LS, the 98LS 4-way and the 78LS 3-way. The 78LS was the very last AR 12" 3-way speaker that used domes for the mid and tweeter. You can think of it as being very similar to the 58s, except that the 78LS used the dual dome mid-tweet, while the 58s used the separate midrange and tweeter.

From a tonal standpoint, I'd expect the two speakers to be virtually identical and quite excellent. I've heard the 58s (essentially an AR-91 in "bookshelf" format), but I've never heard the 78LS. I don't think there would be anything surprising or unexpected about the 78LS, however.

I've attached an April 1984 High Fidelity review of the 78LS.

Steve F.

post-100522-0-43829800-1406810962_thumb.

Link to comment
Share on other sites

Thank You Steve, very interesting test. But the room response shows a moderate extension in the low end, about 55 hz -3 db, while the 3 / 3a extends their low notes until 35 hz - 3 db . Also in the listening test they says needs a bass boost. Probably the tonal balance tends to a bright side. Surely brighter than classic models. I suspect that since late seventies AR indulge in market's trends , losing its identity. In Italy we says " neither meat nor fish ". In other words, an AR 3 is not a JBL 4311.

Link to comment
Share on other sites

I think AR missed an opportunity with the 78LS, afforded by that combination mid/tweeter.

Stand-mounted mini-monitors from Europe and the UK were gaining a foothold in the '80s, and most had a narrow-front to deep-sided cabinet format - a more modern-looking perspective that has pretty much been the norm for high-end speakers ever since.

The AR-9 and AR-90 were floor-standing examples of this "modern" look, but AR never really applied it to their bookshelf speakers.

The 78LS could have had a much more narrow front panel & deep sides, allowing a reduced overall height; and have had dedicated metal stands to achieve an optimum listening elevation.

Updating the appearance should have included the same attention to cabinet design & detail as they'd put into the classic AR-1 and AR-9 boxes, as well as a grown-up front panel (that vinyl face with the printed driver descriptions is très cheesey), and a premium grille cloth arrangement.

With their new mid/tweet, the tried & true 12" woofer, and a high-quality crossover, the performance of this speaker might have opened up a whole new upscale market to AR - one that was unnecessarily ceded to European models that could barely be described as "full-range".

Link to comment
Share on other sites

Thank You Steve, very interesting test. But the room response shows a moderate extension in the low end, about 55 hz -3 db, while the 3 / 3a extends their low notes until 35 hz - 3 db . Also in the listening test they says needs a bass boost. Probably the tonal balance tends to a bright side. Surely brighter than classic models. I suspect that since late seventies AR indulge in market's trends , losing its identity. In Italy we says " neither meat nor fish ". In other words, an AR 3 is not a JBL 4311.

AR 78LS has an effective Q of 0.55 somewhat lower than 0.7 or grater quoted for AR3/3a. So the bass of AR78 would be better damped(controlled) vs more relaxed for the AR3/3a.

Link to comment
Share on other sites

ar_pro's comments about the "missed opportunity" with the 78LS are very interesting, indeed. AR was such a conservative and unimaginative marketing company that doing something way out of the ordinary likely never even occurred to them. The 78 was the natural, "automatic" follow-on from the 1, 3, 3a, 11, 58s to the 78LS. A bookshelf box, about 25-27" x 14-ish x 11-ish. Automatic--never even another consideration. Too bad. A "modern"-looking 78LS like pro describes would have probably been hugely successful and ushered in a new era for AR.

As to the measured results in the High Fidelity test, I don't think you can really compare the DSL (Diversified Science Labs) low-end results of the 78 with the curves that AR ran on the 12" woofer in the cabinet for the 1/3/3a. Low-end measurements from different organizations made under widely different conditions always yield different results. AR spec'd the 78 as being down -3dB @ 39 Hz, very close to what the 3a did. I have no reason to believe that the low-end of the 58 or 78 was much different from the 3/3a.

As to it needing a "bass boost," that was with the speaker placed well out into the room, a placement that AR stated in their owner's manual that they did not recommend.

Brighter than the 3a? Sure, every 12" 3-way AR speaker after the 3a was brighter--the 11, 10 Pi, 58s, 91, and 78LS. I don't think there is any real difference in the tonal balance of any of the post-3a speakers, nor do I think there is any meaningful difference in their low-end responses. All AR single-12" woofer speakers had excellent bass (tweaked a little here or there, Q's of .7 or .55, -3dB @ 35 or 39, etc.), and I wouldn't think that either the 58s or the 78LS was "deficient" in bass compared to the 3 or 3a.

I'd have a lot more confidence in the accuracy of DSL's mid and treble FR curve, and here you can see two things: the almost unbelievably phenomenal dispersion of the 78LS, as all the curves sit right on top of each other, and the gently downward sloping response in the treble, indicating that the speaker is not harsh and shrill.

This is obviously a very, very good speaker.

Steve F.

Link to comment
Share on other sites

AR 78LS has an effective Q of 0.55 somewhat lower than 0.7 or grater quoted for AR3/3a. So the bass of AR78 would be better damped(controlled) vs more relaxed for the AR3/3a.

No, this is not correct.

The AR78 was designed for 1π floor placement; i.e., on the floor back toward the wall but away from the corners, so the bass output at resonance is doubled (3dB) by virtue of this placement. The low Q of the AR78 was designed to compensate for this placement, as it was anticipated that the woofer-floor intersection would otherwise cause "heavy" or "boomy" bass. Witness to this would be to place a 2π speaker—such as the AR-3/AR-3a—directly on the floor close to the wall. If you have ever done this, you will realize why is it usually necessary to get the AR-3 or AR-3a up off the floor at least a foot or so. Therefore, the AR78 is more aptly a 1π speaker, whereas the AR-3/3a are 2π speakers, designed to be placed up off the floor (at least one foot or more) or on a shelf, but back against the wall. By the way, the lowered damping came about because of the slightly larger enclosure.

The AR78 has neither better nor worse damping because of its low Q; it does, however, have an over-damped bass response in the region of resonance when measured with the usual 2π protocol, and it would necessitate tone controls to bring up the bass output of this speaker if it is mounted within a bookcase or up on stands, etc. Proper damping—or the prevention of overshoot ("ringing") in output at resonance—is no better with a Q of 0.5 than a Q of 1.0 or 0.7 with regard to AR woofers, because of the presence of acoustic and mechanical inherent damping in addition to electro-magnetic damping. In other words, the AR-3's Q of 1.0 exhibits no ringing in the output, and the response at resonance is up only about 1.5 dB at resonance, whereas a Q of 0.5 will be attenuated by about 4 dB at resonance—when measured at the usual 180˚ solid angle.

—Tom Tyson

post-100160-0-70367000-1406907481_thumb.

Link to comment
Share on other sites

In typical AR poor-marketing fashion, they design a speaker that is barely 27" tall to sit on the floor, looking ridiculous, and firing its mid-treble energy at the seated listener's waist! Idiots.

The 11 and 10 Pi were spec'd with a Q of .7, I believe. I'm not aware of 3a lit that published this spec (1.0), although I've heard Roy say on many occasions that he always used to shoot for a Q of 1.0. I think in the 3a's timeframe, the Q spec was not widely published.

Attached is another "Q" diagram, with a few more options and a little more clearly labeled.

Steve F.

post-100522-0-51342400-1406908828_thumb.

Link to comment
Share on other sites

In typical AR poor-marketing fashion, they design a speaker that is barely 27" tall to sit on the floor, looking ridiculous, and firing its mid-treble energy at the seated listener's waist! Idiots.

The 11 and 10 Pi were spec'd with a Q of .7, I believe. I'm not aware of 3a lit that published this spec (1.0), although I've heard Roy say on many occasions that he always used to shoot for a Q of 1.0. I think in the 3a's timeframe, the Q spec was not widely published.

Attached is another "Q" diagram, with a few more options and a little more clearly labeled.

Steve F.

The graph shows at resonant frequency there is about 3 db difference between Q of 0.5 and 0.7 as well as between 0.7 and 1.0. Do you think 3 db difference is perceptible?

Link to comment
Share on other sites

In typical AR poor-marketing fashion, they design a speaker that is barely 27" tall to sit on the floor, looking ridiculous, and firing its mid-treble energy at the seated listener's waist! Idiots.

The 11 and 10 Pi were spec'd with a Q of .7, I believe. I'm not aware of 3a lit that published this spec (1.0), although I've heard Roy say on many occasions that he always used to shoot for a Q of 1.0. I think in the 3a's timeframe, the Q spec was not widely published.

Attached is another "Q" diagram, with a few more options and a little more clearly labeled.

Steve F.

The AR-11 Q was advertised as 0.75; the AR-1, AR-3 and AR-3a had a Q of 1.0 as described in Ed Villchur's disclosure articles, but there was no mention of the Q in any of the literature to my knowledge. There was also some doubt that the Q was exactly as measured: Villchur felt the original AR-1 woofer was somewhat over-damped, and for this reason, he used a damping factor of 1 on the Fairchild 75 watt amplifier used to test the original AR-1 buried in a hole outside. This was to be certain that the response was not attenuated at resonance, and the response measurement was 38-1000 Hz, +/- 1.5 dB.

When the AR-3a was developed in 1967, it used the identical AR-3 Alnico-5 woofer (#3700), and the damping was same as the earlier AR-3 and AR-1, all set at a Q of 1.0. Roy Allison and Chuck McShane used the same guidelines for the 1969 AR-3 with the new ferrite-magnet (200003-0) 12-inch woofer, and the Q was set at 1.0 for this speaker. The differences between 0.7 and 1.0 are slight, of course.

Attached image shows an AR-1 12-inch woofer with impulse testing in the region of resonance, with a Q of 0.5, 0.7 and 1.0, showing no perceptible ringing at any of the damping figures.

--Tom

post-100160-0-10400100-1406950104_thumb.

Link to comment
Share on other sites

The 10 Pi and 11 had a Q spec of .7, as I said, not .75. The ADD lit that I gave to the Library is there, if you'd like to see it. In any event, not a huge deal between .75 and .7 (BTW, probably ".707," right?).

More interesting is that the 10" ADDs--the 12 and 14-- have a Q of 1.15, which is definitely a little "fat." One wonders if AR was trying to give the 10-inchers a bit more "apparent" bass, so they'd seem a little closer to the 12" models in that regard. It's also likely that AR knew that the vast majority of readers would have absolutely no idea what "Q" meant, and would therefore not raise their eyebrows at the very "un-audiophilish" 1.15 number. You know, like spec-obsessed crazies like me did.

I actually don't think the 78LS was really intended for direct floor placement--not even AR would be that stupid--but probably meant for placement on a low (6" or so) stand or bench. So placed, their bass is undoubtedly just fine, right in line with any good AR 12-incher. Adriano--sitting on top of those big Altec boxes that you use for speaker stands in that room is probably not the best for the 78LS.

If I remember, the AR9 had a Q of .5, which is a good choice for that model, given how much LF energy it developed and the way it integrated into the room.

Steve F.

Link to comment
Share on other sites

  • 3 years later...

Thanks!

 

Odd crossover, and 400hz cut off on woofer!  If that is accurate.....

I'll throw caps in it, and reform it....then scrutinize it.

 

I will say, once sorted out, there is no problem with that mid,tweeter assembly.  The tweeter in that series is really nice

Link to comment
Share on other sites

Slapped on a set of 12''s while I refoam.

 

I don't know what Tom Tysons is talking about with pi's, but these need stands.

 

Needs caps and a light resistor on miss for my taste, but these are great speakers.  GREAT tweeter.

 

A little overstuffed too......it's muting some bass output.   I won't even touch the "they had it perfect in the chamber" crowd, but I'll do my magic.  These are pleasing speakers and simple. 

15310021136791858123599.jpg

1531002161204687525313.jpg

Link to comment
Share on other sites

All done.

 

Put poly caps on 5mf.tweeter, the rest Dayton npe

Added 1.0 ohm to midrange circuit

Removed 1/2 of polyfil.

Results?  Smoother, more balanced, and overall a great small speaker.

Ar3a bass?  No, or not yet.  I suspect changing the polyfil to fiberglass will help it some.   I'm going to allow the surrounds to break in.  BTW Springfield surrounds SEAL.  I'm focused on that, as MIDWEST speakers surrounds all leaked for me.   Complete garbage and a ton of wasted time and effort.   They are not all the same.

Crisp highs, detailed mids, crystal clear.

They won't replace a ar9, but they are going into office for a longer term listen.  So far, I Like Them alot

20180708_073146.jpg

20180708_073140.jpg

20180708_080317.jpg

20180708_080323.jpg

20180708_092227.jpg

20180708_101427.jpg

20180708_101435.jpg

20180708_101447.jpg

Link to comment
Share on other sites

On 8/1/2014 at 3:54 PM, ligs said:

The graph shows at resonant frequency there is about 3 db difference between Q of 0.5 and 0.7 as well as between 0.7 and 1.0. Do you think 3 db difference is perceptible?

Ligs,

That was a loaded question.  

Yes it's perceptible.   

 

Leads me to next question on how stuffing effects Q?  The amount and the material chosen.

 

 

I can say the dispersion and warmth of these are excellent, and they are extremely clear, all they need is a little more bass output.  

 

My measuremedy of about 4000ci vs 5000ci of cabinet on ar78 vs ar3a.....shows more than 20%more cabinet size on ar3a.

Link to comment
Share on other sites

  • 2 weeks later...

The lower Q with significantly less output at Fc and smaller woofer inductor for 

less baffle step will make these have much less bass than the 3a/11 type speaker.

I like the mid/tweeter combo, nice!

Link to comment
Share on other sites

Hi Pete

 

Im listening to them right now.  Bass?   Has it.   Yes it does.

Placement is the magic.   Im comparing to the 58s right now.    58s is surprising me.  A touch deeper,  more bass output from 58s, but its not crazy different.

I need to learn more about "Q", but these 3 ways are real good.  

Will they replace my ar9?    No.

3a has magic in the midrange.....unsure if its the fiberglass ring on top of the dome, as my brother thinks, or the circuit....but the 3a has a great midrange.    The 58 doesnt have that.   But the 58 has a totally superior tweeter.   Best tweeter ive heard.    

 

Im lucky to have experienced all these models.   I did work hard fixing them all.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...