Jump to content

Tweeter for AR-3a Improved


Guest mrhagerty

Recommended Posts

Guest mrhagerty

I'm sure I'm asking a question previously answered elsewhere in the vault, but I can't seem to get Search to work properly.

Is the tweeter for an AR-3a a good replacement for the 3a Improved system? Finding an exact replacement online is not easy, while 3a tweeters seem to abound.

Informed advice or references would be much appreciated.

Mike

Link to comment
Share on other sites

Is the tweeter for an AR-3a a good replacement for the 3a Improved system? Finding an exact replacement online is not easy, while 3a tweeters seem to abound.

Don't use an original 3a tweeter, it won't have the right response. A lot has been written recently about finding new tweeters for older 3a's, and one solution provided by Ken Kantor is the 3a replacement sold by AB Techservices with a mod to the crossover (an inductor across the tweeter terminals). As it turns out, that mod is already in your 3a Improved crossover, so the AB replacement should be a drop-in fix for you. You'll probably be better off replacing tweeters in both your speakers to ensure a match.

Link to comment
Share on other sites

Don't use an original 3a tweeter, it won't have the right response. A lot has been written recently about finding new tweeters for older 3a's, and one solution provided by Ken Kantor is the 3a replacement sold by AB Techservices with a mod to the crossover (an inductor across the tweeter terminals). As it turns out, that mod is already in your 3a Improved crossover, so the AB replacement should be a drop-in fix for you. You'll probably be better off replacing tweeters in both your speakers to ensure a match.

Hello, about 15 years ago i blew one of the tweeters in my improveds.At that time you could buy replacement tweeters via ar in Holland, so i installed 2 new ones.I was told that they were of the ferrofluid type and they performed excellently uptill now!!

Fed with white noise gives an balanced, non agressive sound. With music they sound warm, with an open mid and non-irritating highs.

Link to comment
Share on other sites

Is the tweeter for an AR-3a a good replacement for the 3a Improved system? Informed advice or references would be much appreciated.

Mike

Mike:

No it is not. The tweeter used in the AR-3a is not the same tweeter that was used in the AR-3a Improved. The AR-3a used tweeters numbered A.18, A.19, A.20 or A.23. (Driver numbers are from the photos on pp. 31-32 in the Appendix of the AR-3a restore document.) You might wish to read Tom Tyson's text of Section A.3 accompanying the driver photos.

The AR-3a Improved used ferrofluid cooled tweeters numbered A.22 or A.24 in the same document. These two models appear to be electrically the same; A.24 has a foam covering.

First, and most importantly to avoid confusion, ensure that you do have the "Improved" model. Many of the AR-3a sold in Europe before 1976 were identical to the US AR-3a with the exception of the cabinet frame and gille cloth; their badge read "AR-3a." The badge on those made after 1976 read "AR-3a Improved." They used level switches not potentiometers and the sticker on the rear side read "AR-3a Improved."

The "improved" does not have the same crossover as the AR-3a. It uses a second-order crossover (C-L) with the same 6-uF capacitor as in the AR-3a plus a 0.1 mH inductor across the tweeter. (Same style crossover as the Mark II AR-11 and other speakers using ferrofluid tweeters.)

You could probably use Ken Kantor's crossover modification for the AR-3a shown in the restore document with the 2006 version of the AB-Tech ferrofluid tweeter. That modification uses a 0.07 mH coil across the tweeter.

My choice would be to contact a reliable seller like Vintage AR and try to purchase a pair of either A-22 or A-24. In this way you know for sure what will arrive in the mail. You might try to buy a pair of AR-11 or AR-10pi tweeters on ebay -- they do not appear very often -- but then you might get A.21, the non-ferrofluid tweeter used in their respective Mark I versions. Best to deal with a supplier who understands what they sell.

Before doing that, is there a particular reason why you wish to change the drivers? Is because of all the posts discussing tweeter degradation? Please bear in mind that those discussions concern AR-3a tweeters (A.18, .19, .20, .23) and not the ferrofluid, higher sensitivity tweeters used in the "limited."

Cheers and hope this helps,

Link to comment
Share on other sites

AB-Tech ferrofluid tweeter. That modification uses a 0.07 mH coil across the tweeter.

and not the ferrofluid, higher sensitivity tweeters used in the "limited."

Cheers and hope this helps,

Mike:

No matter how clear I try to be, fog or errors sneak in. The first sentence above should say that if you were to install the AB-Tech tweeter you would need to remove the existing 0.1 mH coil and replace it with a 0.07 mH coil. (or unwind the 0.1 to 0.07 mH)

The last sentence should end with the word "improved" not "limited." Sorry!

Cheers,

John

Link to comment
Share on other sites

Guest mrhagerty
Mike:

No matter how clear I try to be, fog or errors sneak in. The first sentence above should say that if you were to install the AB-Tech tweeter you would need to remove the existing 0.1 mH coil and replace it with a 0.07 mH coil. (or unwind the 0.1 to 0.07 mH)

The last sentence should end with the word "improved" not "limited." Sorry!

Cheers,

John

Actually, John, now I'm a little confused on the crossover. According to genek above, the Improveds already incorporate a crossover mod that allow the AB Tech replacement to "drop in" as-is. But your comment says I need to do a mod.

Which is correct?

Mike

Link to comment
Share on other sites

Actually, John, now I'm a little confused on the crossover. According to genek above, the Improveds already incorporate a crossover mod that allow the AB Tech replacement to "drop in" as-is. But your comment says I need to do a mod.

John is *more* correct. The Improved has a .1 mH coil across the tweeter and the Kantor 3a mod calls for a .07 mH coil at that same place (the 3a has no coil at all there). I don't know how much difference that .03 mH delta between the two values will make on your speakers, but if they were mine, I'd connect the tweeter to one speaker with the existing coil first and see if I could hear any difference between it and the original tweeter in the other before making any changes in the crossover.

Link to comment
Share on other sites

Guest mrhagerty
John is *more* correct. The Improved has a .1 mH coil across the tweeter and the Kantor mod calls for a .07 mH coil at that same place (the 3a has no coil at all there). I don't know how much difference that .03 mH delta between the two valus will make on your speakers, but if they were mine, I'd connect the tweeter to one speaker with the existing coil first and see if I could hear any difference between it and the original tweeter in the other before making any changes in the crossover.

Thanks for the clarification.

JOHN - I do in fact have the Improveds because the badging was correct and there is a single toggle switch adjacent to the terminals, as opposed to pots.

Mike

Link to comment
Share on other sites

Actually, John, now I'm a little confused on the crossover. According to genek above, the Improveds already incorporate a crossover mod that allow the AB Tech replacement to "drop in" as-is. But your comment says I need to do a mod.

Which is correct?

Mike

Mike:

The crossover for the improved tweeter has a .1 mH coil wired in parallel. Ken Kantor's measurements showed that the correct value of parallel coil for the AB-Tech tweeter (2006 model) is 0.07 mH. He obtained the best (flattest) response with this value and the existing 6 uF cap. Ken's data show that the 0.1 mH coil does not sufficiently attenuate the AB-Tech in its low range. You could either replace the coil or add a 0.2 mH in parallel, keeping it, say, an inch away from any other coil. That combination would yield 0.067 mH, which is as close as one can get using commercially available coils.

But back to the beginning. For what reason are you considering replacing of this tweeter? This tweeter is not the same model that was used in the AR-3a. It has ferrofluid cooling and high sensitivity. Do you know for sure that it is defective? The same model was used in the AR-11 and AR-10pi and those are not going south.

My concern is that one does not allow the leaves obscure the view of the tree. Yes, the 0.07 mH coil is "the correct" value of parallel inductor, but it would not be prudent to replace the existing the hi-range drivers without solid evidence that they are defective! ;)

Cheers,

Link to comment
Share on other sites

Don't use an original 3a tweeter, it won't have the right response.
This tweeter is not the same model that was used in the AR-3a. It has ferrofluid cooling and high sensitivity.

Hold on a moment—lest we consider the "Tweeter for AR-3a Improved" to be objectified—it remains a subject open for discussion. It would appear that more that one flavour of driver was used, during the production run.

I have a pair of AR-3a Improved which date from late '74, which therefore pre-date the introduction of the ferrofluid-cooled hi-range, #A-22 (reported as 1976, in the 'Restoring the AR-3a' document). When re-foaming the woofers, I noted they were date-coded year 74, week 24, and the mid-range that I examined was 561 7436 with a barely decipherable six-character stamp reading 4503?? (the last two characters were smudged beyond recognition). The woofers look like type A.3; the mid-ranges, like type A.13 (back-wired). I did not remove the hi-range drivers, but can identify them as looking, in every way, like the A.19 version. (They are front-wired to the terminal tie-point on the baffle board front.)

Attached below are several reference images (all are of the AR-3a Improved, from various sources). The keen-eyed viewer will notice (at least three?) variations:

post-101656-1207355733.jpg post-101656-1207355761.jpg post-101656-1207355648.jpg post-101656-1207355791.jpg

As someone once wrote "ARs often have the appearance, once you remove the grills, of having been designed and built by drunken hobos. That's part of their charm." ;) Hand assembled beauties, even from this late in the classic-AR day, can startle contemporary sensibilities—especially those informed by and accustomed to machined surface uniformity, which often masquerades as a signifier for good design—I digress!

Robert_S

Link to comment
Share on other sites

Robert:

Thank you for your photos. Looking at them from left to right, I would say there are two models: No's 1, 2, and 4 are pre-1976 early models; #3 is the post-1976 model.

- #1, 2, 4 have front-mounted terminals for mid and tweet

- #1, 2, 4 have one of the several AR-3a tweeters

- #1 clearly shows the two potentiometers on the backside (early crossover).

- backsides of 2, 4 are not shown, so potentiometers are not visible; #1,2,4 would have predated the later version.

- #3 has back-wired mid/tweet, with the A.22 ferrofluid-cooled tweeter.

The remaining issue is the early model logo. For some time I had believed that at least some of the early models also used an "AR-3a limited" logo, but was "convinced" by others that I was daft. Your photo shows that this is not universally true. Perhaps it depended on where in Europe they were assembled? Holland? England? ... ? That I don't know, but I do know that pre-1976 units with the standard "AR-3a" logo do exist.

So I will modify my definition and say that the later (post 1976) version can be identified by its level switch, back-wired mids/tweeters, and A.22 tweeters also used in the AR-11 and AR-10pi. The "limited" badge appears not to be unique to the 1976 model.

Thanks for your information, and cheers,

Link to comment
Share on other sites

Hold on a moment! Lest we consider the "Tweeter for AR-3a Improved" to be objectified, it remains a subject still—for it seems that more that one flavour of driver was used, during the production-run life-span.

Hi John and Robert

All the 3a Improved I have seen, pictured as well as the set I own myself, has the AR-3a type driver, front or backwired with the three orange/yellow butyl foam dots for suspension, the ones type numbered A.19 or A.20 in the restauration document (The A.20 was also used in the late versions of the AR-3a). On all the AR-3a speakers you have pictured in your post, I see this type of tweeters, also in #3, not the type A.22 John notes, since it has the three yellow/orange dots.

And as for the #1, John, I do not see any potmeters, I see one switch with the A/B settings as on all other AR-3a Improved and I see the frontwired AR-3a tweeter the type A.19. They are S/N 760, pretty early for AR-3a Improved. Here in Denmark, the 3a Improved was marketed from 1974.

Is it my eyes failing?

I have enclosed a picture of my late (S/N 5022) AR-3a Improved. They have the backwired tweeter, note the connecting block, used for frontwired drivers, are completely gone and the tweeters are the A.20 type; they are dated 75 week 41 on the back.

BRgds Klaus

post-101646-1207394685.jpg

Link to comment
Share on other sites

A few too many thoughts (improved, or otherwise).

Mike, if as-close-as-you-can-get to truth-to-the-original (in restoration terms) is your concern, then I believe that the AR-3a tweeter is correct for your AR-3a Improved, unless you are certain that your speakers had something different, factory-installed, to begin with.

Klaus, I agree with your observation, that is what I was hoping to show; that the AR-3a Improved can be seen with several variations of the AR-3a hi-range (Fig. A.19, A.20 and A.23, in the restoration document), consistent perhaps with the date of production and/or availability.

John, I think you may be arriving at an incorrect conclusion, regarding the Improved, in that you appear to hold the belief that the 'Improved' crossover was specifically designed with the ferrofluid-type driver in mind, and therefore that it dates from around the time said hi-range was introduced. This perhaps leads you to believe that there are two distinct crossover designs, one pre-'76 (without a switch?), one later (specifically for A.22); therein lies the source of (what I believe to be) your confusion! Please correct me if I am wrong. You get closer to how I perceive events—although just a year or two out of sync—in the post linked here: topic 2214, post #25, than in the information that you are conveying above, in post #5.

Incidentally, can you be sure that your 3a Improved have (or had, when they were new-to-you), the original factory-installed drivers?

Referring to the images I previously posted: 1, 2, 3 and 4; I've pictures that show the back of all of them (bar image 3): they all have the single A/B switch, no pots, thereby establishing that they are not incorrectly-badged AR-3a sporting European-style cabinets. The example show in image 3 is somewhat curious though, in that the paper-label refers to the adjustment controls found on an AR-3a, and the serial number would seem to run quite high (16242), though it is obviously a late-production model (confirmed by the absence of a terminal block), and possibly just mislabelled ("Hey, we've run out of 3a-imp labels." "Here, just use this, it won't affect the sound!" :)).

My understanding of the AR-3a Improved, is as follows:

The defining characteristic of the 3a Improved, is the presence of the crossover-type, as shown below. This crossover has an A/B switch, by design (no potentiometers), and was, I believe, introduced around 1974:

Image 1. Detail of two crossovers, both from early units (no serials), that have all the original drivers (front-wired hi-range, as in figure A.19), dating from late 1974:

post-101656-1207600445.jpg

Image 2. Another example, date unknown, but the serial number is shown (note fiberglass stuffing):

post-101656-1207600474.jpg

Image 3. The schematic deduced and complied by John and Minh; also showing, for reference, what I presume to be the working source-photograph:

post-101656-1207600515.jpg

Drivers:

Woofer: AR-3a type (Fig. A.3 in the restore-doc, or late production models may have A.4).

Mid-range: AR-3a type, back-wired 4-Ohm.

Hi-range: AR-3a type, with the attendant front/back-wired design variations, A.19, A.20 or A.23 (though perhaps the ferrofluid tweeter was specified, post 1976?).

Stuffing: Typically, multi-coloured polyester, though one example above is shown with fibreglass.

Minor front baffle board variations can be seen: Terminal block and strip (See first image below); Terminal block, no strip (see second image below); Terminal block absent (see image posted by Klaus, in post #13):

post-101656-1207605532.jpg post-101656-1207600137.jpg

The Cabinet: I've only seen the 'European' style narrow edge:

post-101656-1207599921.jpg

Steve F makes an interesting comment, in the following topic, post #6, as does Minh Luong, in post #5.

http://www.classicspeakerpages.net/IP.Boar...;showtopic=1287

If there was a post 1976 variation, produced with the 'A.22' tweeters, factory-installed, it could perhaps be considered—informally—as the AR-3a Improved/x. If the crossover design was altered, we have yet to see and document those changes.

Hopefully this discussion will continue and bring more clarity and cohesion to our collective understanding, concerning the history of the AR-3a Improved. I presently believe there was only one model, with minor variations in cosmetics and stuffing material.

Robert_S

Link to comment
Share on other sites

Robert:

All of the speakers you are describing seem to be pre-1976 if they have AR-3a drivers and fiberglass stuffing. There is no question that their drivers are the same as those used the AR-3a.

The post 1976 Minh and I described has ferrofluid cooled tweeter, a 0.1 mH coil across the tweeter, and red/blue polyester stuffing. The coil across the tweeter is the unique unambiguous identifier of the ferrofluid cooled tweeter from a crossover perspective, but that tweeter can also be identified from its front view. Sometime around late 1976 or early 1977 AR stopped using fiberglass. One can see this change in the AR-11 Mark I to Mark II, where 20 oz of FG was replaced by 11 oz of polyester fiber.

The crossover schematic Minh and I drew is from a late model, perhaps even 1977 ? Martin from Germany posted the schematic of his two years ago; it is the same as the pair Minh and I described. Coil across the tweeter = ferrofluid cooled tweeter.

Cheers,

post-100900-1207634919.jpg

Link to comment
Share on other sites

Hello John,

All the speakers I am describing are AR-3a Improved, regardless of age!

Only one example is shown as having fiberglass stuffing, and I note it only because, in my experience, it is unusual to see fiberglass in a 3a Improved. I am aware that when it comes to stuffing, you certainly know your stuff. What you may not know, is that in Europe, the polyester stuffing was introduced at an earlier point in time, than the period you mention. Therefore it is perhaps unwise to consider the stuffing-fiber-type to be a global-constant, with reference to dating an AR speaker. See these posts, for examples of local variation:

http://www.classicspeakerpages.net/IP.Boar...ost&p=63200

http://www.classicspeakerpages.net/IP.Boar...ost&p=73205

When stating "My understanding of the AR-3a Improved", I describe the stuffing as "Typically, multi-coloured polyester . . .", this description is empirical, I am just literally describing what I have observed. Here, for example, is a sample of the stuffing from my AR-3a Improved (photographed in sunlight):

post-101656-1207681753.jpg post-101656-1207681777.jpg

It's perhaps best not to let this discussion veer too far off-topic (as you've gone deep on trilobal polyester, elsewhere), so to get back on track, you say "Coil across the tweeter = ferrofluid cooled tweeter."

Pictured below, from one of my AR-3a Improved, the (all original) hi-range and crossover (showing, I believe, the #16 0.1 mH coil, and wired, as I understand it, as per your schematic):

post-101656-1207681819.jpg post-101656-1207681836.jpg

What am I to conclude? Am I missing something really obvious?

The open question seems to relate more to the crossover in the 3a-Improved (we've established that the 3a type hi-range driver was used). Was there any design variation throughout the production period? If so, this has a bearing on the topic subject.

Robert_S

Link to comment
Share on other sites

Robert:

I stand corrected.

Minh Luong says the ferrofluid tweeters were added to the pair by an owner after they were manufactured. Learn something new every day! One definitely would not need an inductor across the the classic AR-3a tweeter!

Cheers,

Link to comment
Share on other sites

Only one example is shown as having fiberglass stuffing, and I note it only because, in my experience, it is unusual to see fiberglass in a 3a Improved.

Hi Robert and John

Just for info, I have fiberglas stuffing in my 3a Improved

BRgds Klaus

Link to comment
Share on other sites

Ah, sorted!—the 3a tweeter it is then. Cheers John.

Robert_S

Hi Robert_S,

Just to be complete here, have you checked the polarity of all the 3a

improved drivers? Wondering if it is the same as the 3a, since the

tweeter inductor difference might have required a change.

Link to comment
Share on other sites

Guest mrhagerty

All -

Well here is the sordid and ugly tale (which will no doubt subject me to questions of sanity).

My left tweeter failed some time after buying the set. I could tell simply by the noticable muddiness on the left and did a sort of cupped hearing test to isolate the tweeter and it was indeed dead.

I purchased the AR-3a tweeter finding no improveds on eBay and when it came and was installed, it was definitely not balanced with respect to the right. Just not as crisp. I used certain test recordings I've come to know for standard response and sure enough, the L was weak.

To test, I substituted a tweeter from my old pair of Dynaco A-25's and, bingo, there was a fine restoration on the left.

I've since let this arrangement stand until I locate the correct tweeter. No doubt someone here will consider me extremely fortunate that I did not damage the crossover or the Dynaco tweeter, but the planets must have been aligned since there have been absolutely no deliterious effects.

The only issue is that the Dynaco of course does not fit the cut out and the cabinet is not air tight. Since I am supplementing with the AR 12 subwoofer, I currently have the EQ toned down on the 3a's as not to overdrive the woofer with air leakage.

Finding the right tweeter and re-tightening the cabinet is my quest at the moment, and I am very sensitive to hearing any overdrive conditions, which are non-existent in my listening, and I believe I have an excellent ear.

So there it is, the ugly truth. Let the punishing blows commence.

Mike

Link to comment
Share on other sites

The only issue is that the Dynaco of course does not fit the cut out and the cabinet is not air tight. Since I am supplementing with the AR 12 subwoofer, I currently have the EQ toned down on the 3a's as not to overdrive the woofer with air leakage.

Finding the right tweeter and re-tightening the cabinet is my quest at the moment, and I am very sensitive to hearing any overdrive conditions, which are non-existent in my listening, and I believe I have an excellent ear.

You could probably reseal the cabinet temporarily by cramming caulking putty into the gaps...?

Link to comment
Share on other sites

Just to be complete here, have you checked the polarity of all the 3a

improved drivers? Wondering if it is the same as the 3a, since the

tweeter inductor difference might have required a change.

I checked my 3a Improved, and it has reversed polarity as pictured on the schematic in this post ie opposite the 3a, where all three drivers have the same polarity. The addition of the inductor across the tweeter makes it a second order filter for the tweeter, and thus the polarity is reversed like on the 10Pi and 11.

BRgds Klaus

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...