AR3a Midrange: Myth or Magic?

[Guest jdybnis]

Hi,

This is my first post on this form. I'm gearing up for a rebuild of a friend's AR3a's. I'm a bit incredulous of the 3a's 1.5" midrange. How can a 1.5" driver be crossed over as low as 525hz and still play at a reasonable volume? I'm an AR novice but I'll make an educated guess that the driver reaches it's mechanical limits at 1mm peak to peak excursion. A 1.5" driver has a surface area of about 11.5cm. The crossover looks reasonably close to a 4th order L-R (down 6db at the crossover point). To me it looks like the 3a can't play over 95db without destroying the midrange. Is this what the AR 3a owners have experienced?

-Josh

[Carlspeak]

>Hi,

>

>This is my first post on this form. I'm gearing up for a

>rebuild of a friend's AR3a's. I'm a bit incredulous of the

>3a's 1.5" midrange. How can a 1.5" driver be crossed

>over as low as 525hz and still play at a reasonable volume?

>I'm an AR novice but I'll make an educated guess that the

>driver reaches it's mechanical limits at 1mm peak to peak

>excursion. A 1.5" driver has a surface area of about

>11.5cm. The crossover looks reasonably close to a 4th order

>L-R (down 6db at the crossover point). To me it looks like the

>3a can't play over 95db without destroying the midrange. Is

>this what the AR 3a owners have experienced?

>

>-Josh

I presume you are aware that the midrange is a hemispherical dome. Are your max spl calculations based on conventional cone formulas?

In any case, these speakers were mfg'd in an era when high power amps >>100 WPC didn't exist and thus some caution should be exercised in powering these speakers.

What exactly are you planning to do?

Perhaps fusing should be considered in the rebuild if you think the speakers might see some fairly high power inputs well beyond 100 watts.

The pots alone are rated for only 25 watts.

Remember, it's all about the music

Carl

Carl's Custom Loudspeakers

[tysontom]

Hi,

This is my first post on this form. I'm gearing up for a rebuild of a friend's AR3a's. I'm a bit incredulous of the 3a's 1.5" midrange. How can a 1.5" driver be crossed over as low as 525hz and still play at a reasonable volume? I'm an AR novice but I'll make an educated guess that the driver reaches it's mechanical limits at 1mm peak to peak excursion. A 1.5" driver has a surface area of about 11.5cm. The crossover looks reasonably close to a 4th order L-R (down 6db at the crossover point). To me it looks like the 3a can't play over 95db without destroying the midrange. Is this what the AR 3a owners have experienced?

-Josh

This is a somewhat goofy assumption. The AR-3a midrange dome has sufficient excursion capability to easily exceed 95 dB, but playing loudly at a steady-state frequency, which is what you are inferring, is something no one would reasonably do. Having peaks occur that will exceed 100 dB is very common for this driver, and I have never heard of one “self-destructing” caused by mechanical limits, under normal circumstances. In any case, it is not physical self-destruction due to restricted excursion that occurs; it is almost always a thermal burn-out due to excessive power into a single frequency or set of frequencies, such as with electronic music and some types of rock music.

Look at it another way: the AR-3a requires -- and can absorb -- large amounts of amplifier power on a short-term basis, which relates to high output levels easily surpassing 95 dB SPL. This speaker system is capable of withstanding 400-500 watts, short-term peak power with music input, but not more than 5-10 watts long-term steady-state into a single frequency. Testing a nearly forty-year-old speaker this way is risky, of course, so the prudent thing to do is to back off the power and to fuse the speaker (Bussman FNM 1-1/4 amp dual-element). Also, insofar as the 1-1/2-inch dome has such wide dispersion, the acoustic power radiated into a room is actually much greater than a similar midrange cone that is capable of generating higher SPL output on axis.

In any event, you should look for quality, not quantity, when it comes to the sound output from that speaker. It will play plenty loudly, but it was never intended to play like a Klipschorn (thank goodness).

--Tom Tyson

[charger3834]

>The pots alone are rated for only 25 watts.

Just curious, how did you arrive at this conclusion ?

[Guest jdybnis]

Don't worry I'm not going to destroy a speaker to figure out how loud it can play a sine wave. I am just curious if there is anything special about the construction of these drivers. I've heard that, like you say the 3a's can soak up tons of power. It sounds like the only times they fail is when someone drives them with a clipped signal coming from an amp that can't deliver enough power.

I was looking on the web for replacement drivers in case the ones I'm working with are not functional. I found a few options: Simply Speakers sells the genuine AR part for $150, AB Tech sells a substitute part for $100, and ebay (none currently available). The only modern 1.5" dome I know of is made by ScanSpeak. I wouldn't normally cross the ScanSpeak driver at 525Hz. I think it would exceed it's mechanical limits at that level too and break when playing loudly.

Thanks for the replies!

[tysontom]

Don't worry I'm not going to destroy a speaker to figure out how loud it can play a sine wave. I am just curious if there is anything special about the construction of these drivers. I've heard that, like you say the 3a's can soak up tons of power. It sounds like the only times they fail is when someone drives them with a clipped signal coming from an amp that can't deliver enough power.

I was looking on the web for replacement drivers in case the ones I'm working with are not functional. I found a few options: Simply Speakers sells the genuine AR part for $150, AB Tech sells a substitute part for $100, and ebay (none currently available). The only modern 1.5" dome I know of is made by ScanSpeak. I wouldn't normally cross the ScanSpeak driver at 525Hz. I think it would exceed it's mechanical limits at that level too and break when playing loudly.

Thanks for the replies!

Josh,

There isn’t anything magical about the AR-3a dome construction, but it is a well-made treated-cloth dome that has pretty decent excursion. The dome has a compliant cloth “half-round,” inverted suspension that does allow it to move quite a bit. What can happen, however, is some non-linearity when the dome is driven abnormally hard, but this is normal for any driver. A clipped signal usually affects primarily the tweeter, but might take out a midrange as well. Clipping usually indicates an underpowered amplifier, a common occurrence driving into a power-hungry, low-impedance AR-3a. I once heard that the #1 cause of burned-out drivers in older AR speakers was “under-powered amplifiers.”

AB Tech has a 1-1/2-inch replacement service-part dome that you could use if the existing AR-3a midrange is defective or “suspect.” This midrange is the later AR-9-style driver with basically similar characteristics to the original AR-3a-type driver, but it is a compromise replacement driver, and there will be some slight differences. It is probably the next-best thing to the original driver, and I believe that it has ferro-fluid in the voice coil to protect it against higher input power.

http://www.classicspeakerpages.net/dc/user_files/1734.jpg

AB Tech Replacement 1.5-inch Midrange Dome (Image from AB Tech Site)

If you replace one midrange, however, you should replace the midrange in the other AR-3a as well to make both speakers sound alike. You should also inspect the crossover capacitors to be sure that they are within 10% of their rated values, as many on this forum have reported problems in this area. There are several bona fide crossover experts on this forum, and they could help you with this if you need it.

--Tom Tyson

[Guest jdybnis]

Thanks Tom, that's the info I was looking for. Is the replacement sold by Simply Speakers the same? http://www.simplyspeakers.com/12diaphragms.htm

http://www.classicspeakerpages.net/dc/user_files/1735.jpg

[RoyC]

Josh,

The midrange in your photo is the replacement driver for the AR-303 and 302 systems produced in the 90's. The dcr is almost twice that of the original midrange (3+ ohms vs 6+ ohms), and the frequency response curve is probably somewhat different as well. Also, it is likely the cabinet hole will need modification if for no other reason than to accomodate the terminals, which can be seen in the photo as different from all versions of the original midrange..

Ken Kantor is the designer of those systems, and is a member of this forum. He may be able to provide insight into the suitability of using this midrange as a replacement for the original.

Roy

[Carlspeak]

>>The pots alone are rated for only 25 watts.

>

>Just curious, how did you arrive at this conclusion ?

>

It's not a conclusion but, instead, a fact. There are some real AR experts who frequent this site and they confirmed it.

Remember, it's all about the music

Carl

Carl's Custom Loudspeakers

[soundminded]

While I'm confident AR3a can play considerably louder than 95db, the purpose of this speaker should be kept in mind. It was intended to reproduce classical music, jazz, and other "serious" music as accurately as possible in a home. It was never intended for very loud rock at deafening levels nor as a sound reinforcement speaker for a theater. Tom is right, the goal was quality, not quantity of sound. This is comparable to the goal of a full range electrostatic speaker like KLH Model 9 although AR3a is more robust than that speaker. At the first sign of distortion from being overdriven with any speaker, it is always a good idea to back off the volume control immediately. If higher SPLs are required than can be produced by a single pair, multiple pairs are a good solution and multiple amplifiers if necessary as well. Consider fusing your speakers if you intend to push them very hard. This will save you a lot of money in repair costs.

AR grappled with the problem of designing a 3 way loudspeaker for decades. Eventually they caved in to the very problem you infer, bridging the gap between the woofer and midrange by going to a four way design. This restricted the upper midrange dome to producing all of its sound in a range best suited for it while doing the same for the woofer. Therefore, AR9 can play even louder than AR3a without distortion.

[dynaco_dan]

Hi Soundminded;

Thank you Tom, Carl, Brad and Roy also.

You wrote a reality check of what the purpose of the AR-3A and all other conventional hifi speakers are made for.

Check out the Crownaudio website regarding their view of speaker powerhandling and expectations.

For really loud sessions, perhaps really efficient disco speakers should be used.

Fusing, in my opinion, should be manditory, as pre-designed by the manufacturers for each level of drivers.

For example, slow blow for woofers and separate fast blow for mids and highs.

Sadly, even though AR, Advent and Dynaco, did their best at the time.

They recommended and supported a single fuse size per speaker system, they could not foresee a need nor justify the increased cost of this deluxe safety feature.

An appropriate comparison in specific fusing is, in old housing electrical panels that use the standard screw-in base glass fuses.

Original fuses could be bought from at least 15 amp up to 30 plus amps.

Inside the fuse was colour coded but the threaded base could fit any socket opening, allowing a 30 amp fuse to protect a 15 amp circuit, NOT.

For a number of years special screw-in fuses, coloured for amperage, and a suitable sized base adapter were assembled together as a unit.

They were then screwed into the panel and when a fuse blows, you unscrew the fuse and the adapter remains in the socket.

Only that size fuse will screw into that same base, as the fuses are threaded different sizes.

This is not practical for our purposes but the size of an auto fuse would be more appropriate if they were physically not interchangeable and engineered by the manufacturers.

Regarding the midrange driver, I'm really scratching my head here but I seem to remember an article on the AR-3A, a long time ago though, some where out there, that over 1,000 watts was briefly used with a pair.

The comment I remember was, how robust the drivers were.

Just my 2 1/2 cents worth.

[dynaco_dan]

Hi again;

I remember an old AR ad and it mentioned that some organization bought 16 mids for sound or hearing tests, I believe.

It shows the accuracy of them at least.

Also Cello used them and there was a lot of choices globally to pick from even then.

[charger3834]

I was just wondering if that was the result of some research on your part. Sounds like it is just common knowledge.

[dynaco_dan]

Hi Brad;

Whom is that question directed to?

[dynaco_dan]

Fusing, in my opinion, should be manditory, as pre-designed by

the manufacturers for each level of drivers.

For example, slow blow for woofers and separate fast blow for

mids and highs.

This is not practical for our purposes but the size of an auto

fuse would be more appropriate if they were physically not

interchangeable and engineered by the manufacturers.

Just my 2 1/2 cents worth.

Hi again;

I was unable to return to add to my previous writing and so, here goes an additional thought to that writing.

I was not suggesting that speaker manfacturers create a dedicated fuse, but use an imaginary fuse similar in size and compactness to the auto fuses of today.

It would not be safe or practical to have such a monopoly.

In practice, you could insert maybe 3 fuses in about 1 square inch of terminal space.

These fuses would need to be used in other fields, not just for speakers or they would be doomed for supply and availablity worldwide.