Jump to content

Another AR-3a restoration


JKent

Recommended Posts

OK. My final comments on these. Harry may want to add his own after listening at home.

Roy suggested turning the tweeters all the way up and whadya know--cymbals had more sparkle and brushes had more sssshhhhhh. All the way may be the way to go with these tweeter rebuilds.

Bass is a little heavy but I have them placed on the floor. These are "bookshelf" speakers, not designed to be placed on the floor so stands would be advisable.

Kent

Link to comment
Share on other sites

Hi Kent, this is another beautiful restoration you've completed and a great discussion about the many factors and components that are involved in such a project. However, I remain a bit confused about certain parts of this tweeter discussion so I'm pretty sure I'm missing something here - - please clarify further if you can.

These speakers belong to original owner, never opened. I went back and re-read the brief statement where you said the mids were fine but tweeters were weak, hence requiring the re-build effort. Still having original caps and crunchy pots in the circuit, I'm curious about how this decision on the tweeters was determined. Basically, I'd like to learn more about proper diagnosis techniques. Was this a subjective evaluation while all drivers are in operation? If so, did tweeter pot offer a 'sweet spot' near max output?  Or, was the tweeter played by itself out of the box?... or, are there certain measurements that can be made without lab equipment to confirm weak output?     

I always appreciate your opinions and subjective evaluations and comparisons, but your most recent post surprised me a little bit because there has been no mention of room acoustics or interior environment. Having newly re-built tweeters, fresh film caps, and brand new L-pads, I would expect that the max setting on the tweeter L-pads would be a recommendation for only the plushest of listening spaces - - by this I mean "soft" - - carpet and rugs, upholstered furniture, drapery, and/or perhaps even absorptive ceiling treatment. Specific musical selection is somewhat of a factor, I suppose, but my inherent sense is that the "all the way" setting in a more austere or "live" room would result in an unpleasant or even overwhelming HF experience. 

My rooms tend to lean heavily toward a "soft" acoustic environment, so I typically have these rotary controls (or switches) on my speakers set to an increased HF output setting. However, based on many room interior pics I see displayed in this forum, this is definitely not the case for many vintage speaker lovers, and I suspect that there is no universal "one-size-fits-all" prescription for these controls that will satisfy every installation.  

Great project, great discussion, thx for sharing.   

 

 

Link to comment
Share on other sites

23 minutes ago, ra.ra said:

Specific musical selection is somewhat of a factor, I suppose, but my inherent sense is that the "all the way" setting in a more austere or "live" room would result in an unpleasant or even overwhelming HF experience. 

My rooms tend to lean heavily toward a "soft" acoustic environment, so I typically have these rotary controls (or switches) on my speakers set to an increased HF output setting. However, based on many room interior pics I see displayed in this forum, this is definitely not the case for many vintage speaker lovers, and I suspect that there is no universal "one-size-fits-all" prescription for these controls that will satisfy every installation.  

 

Ra.ra,...Do you own any of the AR models (3a/2ax/5) with the 3/4" black dome tweeter? These are unique animals, and the tweeter (crossed over at a very high 5000hz) is quite reticent by nature. AR's "normal" setting for these models was just off maximum, and I know of nobody who sets this tweeter below that point. The age degradation typical of this tweeter causes sensitivity and consistency issues, which has lead to confusion and some debate as to what the "proper" response should be. My repair is an attempt to consistently (that's the hard part) make it respond similarly to the original. The foam suspension material can be manipulated to result in very high to very low sensitivity, but its variable nature has led me to understand why the factory rejection rate was so high.

Recently I had a long discussion with Chris (chris1this1) regarding the construction of this tweeter and was planning to discuss some of the considerations and concerns we both share in the repair of these tweeters. I just didn't want to hijack Kent's thread.

Btw, very nice job, as always, Kent!

Roy

Link to comment
Share on other sites

Hey Robert

Thanks for the kind words. I'll try to address your questions and maybe Roy will comment as well.

These speakers belong to the original owner. Whether they have ever been opened is questionable. Harry said he never opened them but who knows if half a century ago they were serviced? Roy noted several bits of evidence: Too little stuffing and a different kind than was used in 1969 and use of soft duct seal for example.  Be that as it may, I tested the original tweeters out of circuit with a direct music source. One seemed OK or maybe a little weak but the other was nearly inaudible. Hence the rebuild.

You are certainly right about the effects of rooms and room placement. My audition of the completed speakers was under "less than ideal" circumstances: They are on a thick carpeted floor, far from the back wall and only about 3 feet apart. Listening was from a chair about 8 feet away and sometimes I just knealt in front of the speakers. And there's a lot of clutter because this is part of my work area. My intent was to be sure everything worked right and not to sit back and enjoy. Harry should be able to give better feedback once he gets these set up in his listening room.

As you know, Roy is constantly tweaking and fine-tuning his work. Based on the feedback he receives he may tweak future rebuilds. You're right--one size never fits all. 

Hey! While I was typing Roy posted his reply.

 

Link to comment
Share on other sites

Hey, thanks, you guys are great - - this is all helpful and interesting.

Kent, this is yet another terrific re-build and I enjoy the way you take us through the various steps and decisions. I've seen pics of your work space and sort of imagined that your audition occurred there, which would certainly confirm the increased settings. Nonetheless, sometimes I almost cringe when I see unadorned rooms with smooth, hard walls and ceilings, tile floors, and large glass patio doors adjacent to speaker locations with virtually no absorptive materials to be seen - - - so I am reminded to bring up the issue of room acoustics when the topic of these variable controls comes up. Sorry if I diverged too far off course here. But Roy raises some other interesting points.......

Yes, Roy, I do currently own one pair of speakers (2ax) with this 3/4" tweeter and this is largely why I was asking specifically about diagnostic techniques. Mine are old yet functional and probably much like Kent's recent pair. They're capable of making a little noise but nothing very impressive is coming out of them - even with fresh caps and pots at max setting. Like many members here, I've been very pleased to see at last two more good options (in addition to the Hi-Vi solution) for these tweeters crop up in recent years: this re-build of the originals (by yourself and Chris), and the new replacement driver from Midwest Speakers. With several suitable paths available, I'm beginning to feel guilty for not having remedied my own speakers yet - - I convince myself I'm still in the data-gathering stage, otherwise known as procrastination :P .

Roy, your remarks about the questionable "proper" response, difficulty with performance consistency, and variable sensitivity of the suspension material are all very revealing and make it easier to understand the factory rejection rate - - also, they suggest the uncertainty in defining the target objectives during a re-build. If you have more to share on this, eager ears await. The comments regarding near universal agreement on the max setting for these controls makes me think my statements about "hard" rooms do not apply to this situation, but it also makes me wonder why a different fixed circuitry - - with no L-pad, since no adjustment is ever required - - has not become a more favorable solution for contemporary re-build projects with this tweeter. Perhaps the words 'restoration' or 'collectible' or 'vintage' get in the way, I dunno.

In any case, this thread has been another successful example of giving two old walnut boxes a new lease on life, and this latest discussion has given me greater admiration for the skills and knowledge that are required for these delicate re-builds of these old drivers. Thanks to all who contribute and share here.       

Link to comment
Share on other sites

On 6/4/2019 at 3:13 PM, ra.ra said:

Roy, your remarks about the questionable "proper" response, difficulty with performance consistency, and variable sensitivity of the suspension material are all very revealing and make it easier to understand the factory rejection rate - - also, they suggest the uncertainty in defining the target objectives during a re-build. If you have more to share on this, eager ears await. The comments regarding near universal agreement on the max setting for these controls makes me think my statements about "hard" rooms do not apply to this situation, but it also makes me wonder why a different fixed circuitry - - with no L-pad, since no adjustment is ever required - - has not become a more favorable solution for contemporary re-build projects with this tweeter. Perhaps the words 'restoration' or 'collectible' or 'vintage' get in the way, I dunno.

Robert,

I'm convinced this tweeter was a nightmare for AR to manufacture consistently. The suspension material's characteristics are affected by mixing ratio, amount applied, mixing temperature, and degree of migration into the voice coil gap as it expands. Other variables were the clear rubber coating applied over the voice coil gap, and the nature of the now-dried-out foam under the dome (my dome has a cloth roll over the gap in lieu of the coating).  I have listened to and tested many hundreds of these tweeters over the years, and though they were all reticent in nature, the output differences have been all over the place. Ken Kantor also provided some data awhile back confirming the nature of these tweeters...so at this point I don't think its response is all age related. It also explains why some people prefer a smaller capacitor with the HiVi, and some don't, based on what they are used to or expect to hear. For this reason I'm back to recommending the use of the original capacitor value as a starting point with the HiVi tweeter.

I believe the key to the successful use of the original tweeter was, in fact, the availability of level controls, along with a very high crossover frequency. (Listen to a 5000hz test tone sometime...not much to hear up there by itself.) I believe these were "variably reticent" tweeters to begin with, and there was a range of acceptability. In other words, the tolerances were not as tight as we might like to believe. Replacements such as the HiVi (with the requisite parallel coil) and Chris's rebuilds are actually on the "higher output" side of acceptability, imo, and the level controls are useful for these. My effort is simply to provide a less labor intensive repair of the original tweeter within the acceptable response range using the original cap value, without the need for a parallel coil. It is also a way to more easily re-purpose a tweeter body with a missing or badly damaged dome.

Chris and I largely agree on the issues mentioned above, and will continue to compare notes. I'm certain his excellent rebuild, however, will remain the most authentic in terms of construction.

Roy

Link to comment
Share on other sites

Thanks, Roy, for the excellent response. That is a very educational brief summary regarding the varying outputs of these tweeters - - both original and re-built units - - as well the rationale and usefulness for the pot control in this circuit. Thinking about this over the past few days, I started asking myself: Is it possible that I have not listened to these tweeters in their "fresh" state in 45 years or so, ever since a couple well-heeled friends of mine purchased new AR-3a's back in early-mid 1970's? I know I've heard plenty of "tired" 3/4" tweeters, and I still have no personal experience with any of the recent re-built variety or the Midwest replacement driver. 

Please correct any misinterpretations here - - but if I understand your statements, it sounds like you and Chris employ somewhat different steps and/or techniques in your reconstruction efforts and, in fact, do not have fully identical benchmarks for output of the rebuilt tweeter. Your process results in a reconfigured tweeter which works best with the pot set to max, as discussed in Kent's project; and Chris's tweeter possibly delivers a bit more HF energy and might sound best with the pot dialed back a bit. Am I beginning to get this right? Either way, it is good to know that we have several options to save or replace these aging drivers, and we doff our hats to you both for carving this new path.

The remark about the number of variables involved with the new foam suspension blobs clearly suggests the challenge of the re-build process and explains the difficulty in achieving consistent results. (The statement about the AR manufacturing "nightmare" immediately brought to mind that 60's AR advertisement which showed barrels full of rejected drivers in the old Cambridge facility.) What is most surprising, however, is that despite the challenge of the initial manufacturing process to produce tweeters with consistent response performance, this driver was assigned high-profile duty in two of AR's biggest Classic sellers - - the 2ax and 3a - - as well as several other top line models (AR-5, LST/2, and LST). The reviews were great, if I remember correctly, but one wonders what/where was the level of corporate confidence regarding longevity? My suspicion is that - - much like the new-fangled urethane polymer (foam) surrounds - - no one really knew how to predict a meaningful lifespan.   

This discussion is tempting me to tear into my 2ax speakers later this summer at the same time when I assist my friend with his 3a's, and I assume I will follow Kent's procedures for diagnosing these tweeters after removal from their cabinets. And this brings up another question: exactly what are the physical differences between the 4-ohm and 8-ohm versions of this tweeter, and what should one look for if examining a pair of each simultaneously?

Kent - - once again, your restoration project here is terrific - - but like many threads, spurs are created which must be explored. :rolleyes:      

Link to comment
Share on other sites

On 6/8/2019 at 9:55 PM, ra.ra said:

Thanks, Roy, for the excellent response. That is a very educational brief summary regarding the varying outputs of these tweeters - - both original and re-built units - - as well the rationale and usefulness for the pot control in this circuit. Thinking about this over the past few days, I started asking myself: Is it possible that I have not listened to these tweeters in their "fresh" state in 45 years or so, ever since a couple well-heeled friends of mine purchased new AR-3a's back in early-mid 1970's? I know I've heard plenty of "tired" 3/4" tweeters, and I still have no personal experience with any of the recent re-built variety or the Midwest replacement driver. 

Please correct any misinterpretations here - - but if I understand your statements, it sounds like you and Chris employ somewhat different steps and/or techniques in your reconstruction efforts and, in fact, do not have fully identical benchmarks for output of the rebuilt tweeter. Your process results in a reconfigured tweeter which works best with the pot set to max, as discussed in Kent's project; and Chris's tweeter possibly delivers a bit more HF energy and might sound best with the pot dialed back a bit. Am I beginning to get this right? Either way, it is good to know that we have several options to save or replace these aging drivers, and we doff our hats to you both for carving this new path.

The remark about the number of variables involved with the new foam suspension blobs clearly suggests the challenge of the re-build process and explains the difficulty in achieving consistent results. (The statement about the AR manufacturing "nightmare" immediately brought to mind that 60's AR advertisement which showed barrels full of rejected drivers in the old Cambridge facility.) What is most surprising, however, is that despite the challenge of the initial manufacturing process to produce tweeters with consistent response performance, this driver was assigned high-profile duty in two of AR's biggest Classic sellers - - the 2ax and 3a - - as well as several other top line models (AR-5, LST/2, and LST). The reviews were great, if I remember correctly, but one wonders what/where was the level of corporate confidence regarding longevity? My suspicion is that - - much like the new-fangled urethane polymer (foam) surrounds - - no one really knew how to predict a meaningful lifespan.   

This discussion is tempting me to tear into my 2ax speakers later this summer at the same time when I assist my friend with his 3a's, and I assume I will follow Kent's procedures for diagnosing these tweeters after removal from their cabinets. And this brings up another question: exactly what are the physical differences between the 4-ohm and 8-ohm versions of this tweeter, and what should one look for if examining a pair of each simultaneously?

Kent - - once again, your restoration project here is terrific - - but like many threads, spurs are created which must be explored. :rolleyes:      

I'll go with bullets so you and the rest of the gang don't have to wade through a long narrative.

-The only difference between the 4 and 8 ohm versions of the tweeter is the voice coil resistance. You can't tell by looking at them. The 8 ohm version is, of course, less sensitive.

-The Midwest tweeter, like the HiVi tweeter, requires a parallel coil. The goal of these rebuilds is to eliminate the coil and to use the original crossover cap, though a modern tweeter/coil combo is likely to be the most consistent and long lived solution.

-It is very likely you have not heard any of these tweeters in a "fresh state" in lots of years....don't know how many. As with the AR-3 drivers, the coating over the voice coil gap changed over time as well. It should be noted, however, that AR had the tweeter controls maxed for their listening vs live demonstrations. (I believe Tom T and/or Steve F mentioned there was some treble boost as well for those occasions.)

-Output and overall response can be manipulated. The tweeters I sent to Kent were deliberately on the more reticent side of the replacement options. It is not just the sensitivity, however, that can be varied with the suspension foam. AR manipulated the physical and mechanical properties of drivers to dial in response characteristics...so crossover frequency can also be affected during the manufacturing and rebuild process. The cloth dome I'm using is lighter than the original paper dome, and its voice coil has a slightly higher dcr, so Chris and I are dealing with somewhat different issues to achieve similar results. Since we are not working with large volumes, we are both making an effort to match the response of tweeter pairs. (It is recommended to have tweeters rebuilt in pairs for obvious reasons.) Ideally these rebuilds will result in matched tweeters with characteristics as close as possible to AR's intentions. I have yet to hear a rebuild that was not a significant improvement over a typical original tweeter in its original state.

-AR speakers of the era had inherently wide dispersion. The completely exposed tweeter dome and the tweeter's extremely high crossover frequency were its primary contribution to this attribute. For this reason AR "got away" with this tweeter's probable response variations. With two speakers playing under typical conditions, the vast majority of users were never going to hear any variations between a pair of these tweeters. Once again, I encourage the more technically inclined members to get an idea of what tones at 5000hz and above sound like...and to compare tweeters during restoration work. Frankly, as time went on, some of the back-wired versions of this tweeter I have seen should have been in that reject barrel. (Tom T has confirmed that quality control was not very good near the end of this tweeter's production.)

Roy

Link to comment
Share on other sites

7 hours ago, RoyC said:

I encourage the more technically inclined members to get an idea of what tones at 5000hz and above sound like.

https://www.google.com/search?client=safari&rls=en&q=5000+Hz+tone&ie=UTF-8&oe=UTF-8

Also google other test tones: 14000 Hz, 12000 Hz. How high can you really hear? And how critical can we be of various speakers and headphones that are rated 20 - 20KHz? When I crank my tweeters up all the way it may sound fine to me but drive the neighbor's dog crazy ? Just sayin'

Link to comment
Share on other sites

Thanks again, Roy, for another terrific write-up. Your re-build objectives are clearly stated; possible differences with Chris's re-builds are clarified; and the options to use new replacement tweeters are well summarized. Also, I appreciate the remarks about two speakers in use possibly masking the response variations within a tweeter pair; and again, your advice to repair (or replace) tweeters in pairs rather than singly is always a helpful reminder. The reject barrel bin comment gave me a chuckle, and I enjoy the personalized anecdotes.

I agree with you both about those high frequencies - - the oxygen gets pretty thin at that altitude. On top of that, many of us suffer from the typical aging male syndrome of gradually diminished hearing loss, particularly in the HF spectrum. With respect to the high 5000 Hz crossover frequency for this tweeter, it is worthy to note that in the subsequent top AR models, this x-o point increased to 7000 Hz in the four-ways (AR-9, 90) and 7500 Hz in the three-ways (AR-91, 92, 58s) with the newer ferrofluid tweeters. While the newer midrange (or upper mid) driver was able to extend up to this range, it is interesting to note this division of frequencies among drivers resulting in an even smaller work load for the uppermost driver.      

 

 

 

Link to comment
Share on other sites

  • 3 weeks later...

I'm the happy owner of the speakers that are the subject of this thread. Kent and Roy did an absolutely great job on the restoration. I couldn't be happier! The only problem is I have to wait till my wife goes shopping before I can really crank them up. 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...