RoyC
-
Posts
2,974 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by RoyC
-
-
Daniel,
There is no comprehensive document posted specific to restoring the AR-3.
The AR-3 has essentially the same cloth surround woofer as the early 3a, and the crossover is much simpler than the 3a. The level controls are the same, and will need the same attention as described in the 3a document. AR-3 tweeters and mids are different than the 3a and are often found to be lacking in output. The midrange drivers in particular are usually compromised and often need to be refurbished or replaced. Other than possible specialized repairs to these drivers, the crossover parts are easily acquired from sellers such as Parts Express, Madisound, and Ebay...assuming you are in the US.
The first thing to do is to listen to your 3's to see how they sound, and identify any possible deficiencies.
Roy
-
9 hours ago, Tom H. said:
Hi folks,
I'm restoring a very nice, one-owner, untouched pair of AR-3's They are 1965 vintage.
Looking at the history of the AR-3 schematics document, my speakers lack the factory .06mH coil on the mid-range driver. It looks like it was there to roll off the high range of the mid-range driver on the earlier versions. Any reason not to put one back in? One would think it'd be better to roll the mid-range highs off.
What caps do you guys like on these speakers?
TIA, Tom
The earlier 3's used a very powerful mid with a huge magnet. AR apparently tried the additional series inductor to calm it down at the crossover point, and then played around with tweeter polarity. Eventually the midrange was changed. Adding the inductor for your version is not necessary or recommended. Generally speaking the AR-3 needs all the help it can get at higher frequencies these days.
Roy
-
28 minutes ago, Brad1234 said:
Thanks everyone for the feedback. I dont think I have ever seen any model #'s stamped or written on any of the KLH tweeters, so if the 5, 12 and 23 have a different tweeter I dont know how KLH identified them?
Some tweeters used in the KLH 5 and 12 will have an "M12" very lightly stamped on them. Attached is a photo of a KLH 6 and a KLH 17. Can you tell which one is the KLH 12 tweeter from the KLH 17 tweeter? They are identical, and can easily be used interchangeably. It is really not something to be concerned with, especially considering the scarcity of fully functional used KLH tweeters.
Roy
-
On 7/29/2024 at 11:44 AM, Brad1234 said:
Sorry if this has been covered recently, but have a tweeter out (1.2M ohms thru it) on some Model 17's and trying to find a replacement. I believe all of these KLH 60's/70's vintage tweeters are the same for the models I put in subject line, is that correct? Maybe same for all vintage KLH?
There are no modern replacements for these tweeters.
Even though (slight) response differences have been alleged between the original tweeters, they are all close in response and impedance...and certainly much closer to each other than any other potential non-KLH replacement. The biggest difference is that some were painted black. Your best bet is to install any used original KLH tweeter of the type in your photo.
Roy
-
5 minutes ago, Lucky Pierre said:
Thanks Roy!
This is what I was referencing earlier about the correct amount of fiberglass. (there was approximately 1/3 more of this dense material in these cabinets compared to later cabinets with fiberglass.) This makes sense.
I have worked a lot with both rock wool and fiberglass. While I did take precautions (sleeves, gloves, mask), as always when working with this stuff, I did do a touch/feel test with bare hands. Unless early 60s fiberglass feels significantly different than modern fiberglass, this is certainly rock wool. It feels nothing like fiberglass.
I think I might give these a listen with one speaker overstuffed, and one with 20-24 oz, just to hear the difference.
Peter
...and the old stuff will also "poof" into the air with the slightest disturbance. It is very prudent to handle it outdoors.
Measurements actually showed rather small differences in Fc as material was varied, but subjectively it was noticed in upper bass frequencies. Too little material can become boomy, and too much material reduces bass response. In practice, it is also listening space and preference dependent. I don't think it was a coincidence that AR began using more series inductance in woofer circuits when the amount of stuffing was reduced, as both have an effect on this area of bass response. Fortunately the AR-3 (and early 3a) woofers have a relatively smooth natural roll off into upper frequencies allowing a more forgiving degree of adjustment.
-
3 hours ago, Lucky Pierre said:
Hi all,
I am getting back into my 1962 AR3s to replace the pots and the rock wool with fiberglass. I weighed the rock wool from each cabinet and it came out to 2 lbs., 7-7.5 oz per cabinet! All of the references I have read, here and elsewhere, states that there should be roughly 28 oz (1 lb, 7.5 oz) per cabinet. Any ideas on this? I am the first person that opened up the cabinets when I did the crossovers a few years ago, based on the condition of the screws and woofer sealer goo on the woofer hole.
I currently have 2 b, 7.5 oz fiberglass in cabinet one. I'll out 1 lb, 7.5 oz in cabinet two, just to hear the difference.
Peter
The stuffing in your 3's is the early type of AR damping material, which was used until 1970 coinciding with the introduction of foam surround woofers. There has been uncertainty as to whether it was a type of rockwool or just an early type of fiberglass (Tom Tyson maintained that the early material was simply early fiberglass.) Regardless of the definition, there was approximately 1/3 more of this dense material in these cabinets compared to later cabinets with fiberglass. When putting together the 3a restoration guide, and after many measurements, we concluded that 20 to 24oz of modern fiberglass was optimal for any model using any iteration of the AR 12 inch woofer in cabinets of this volume. Carl Richards independently later confirmed these conclusions. Additionally, Tom T. commented that Villchur thought that cabinets containing the early material were somewhat "overdamped".
One way or the other the early material is very nasty to handle, so take precautions.
Roy
-
On 6/24/2024 at 6:32 PM, Andre_Db66 said:
What is different about the improved drivers?
It seems they are the same only the xo is different.
Judging by above schematic, using a normal 3a xo scheme is not suited for a set of tonegem limited edition drivers, am i right? For this is what i have done.
The 3a Improved drivers are the same as the original 3a, but the crossover was changed primarily due to the elimination of the pots.
Tonegen tweeters were very similar to the AR-11 tweeter, but both were different from the original 3a and Improved 3a tweeter. The replacement you are planning to work with is different from all of them. Various iterations of the 3a can only serve as general templates or starting points.
-
On 6/24/2024 at 9:11 PM, ReliaBill Engineer said:
Hmmm…
Adding a resistor (series or parallel) only adjusts the impedance “seen” by the XO. It doesn’t change the actual impedance curve of the driver itself.
Parallel inductors, not resistors, were used with the new tweeters to change the nature of the crossover to match that of the original mechanical roll-off properties of the early tweeters by decreasing/shaping the new tweeter's stronger contribution to the mid frequencies. The dome mass and actual dcr of all AR's 4 ohm 3/4" dome tweeters were very close. The later tweeter's significantly stronger low frequency response was due to different mechanical properties (ie suspension), and was modified primarily through the addition of parallel inductance. The result was a much more uniform, predictable response. As we know, the original AR tweeters were difficult to consistently manufacture.
Regarding resistors, series resistors alone will attenuate overall output but will also affect the crossover point. Parallel resistors can be used to maintain or create impedance characteristics as you described. When parallel resistance is used in conjunction with series resistance it becomes an L-pad. Variable L-pads are used to better maintain the circuit's impedance at different attenuation levels.
-
21 hours ago, Andre_Db66 said:
Any thoughts would be appreciated
-It is impossible to provide specific crossover and attenuation suggestions to accommodate unknown response differences between the original and your rebuilt tweeters...much less your preferences. Once you get your tweeters functional, start by using everything "as is" to begin with.
-The best way to determine your preferred level setting for your rebuilt tweeter is to set up level controls outside the cabinet, and measure the series and parallel resistance values of your preferred setting. You can then replace those values with fixed resistors. You can also try out different capacitor and inductor values in this way.
-It should also be noted that the short-lived AR-3a Improved departed the most from the other iterations.
-
3 hours ago, ReliaBill Engineer said:
A stiff suspension and high damping of the early domes resulted in high mechanical impedance, restricting power response/output.
Yes, especially at lower frequencies. This was accomplished electrically (and more consistently/uniformly) with the later tweeter. The other drivers, cabinet volume, and crossover actually remained largely the same with later iterations of this design. The original tweeter's impedance (resistance/inductance vs frequency) is unique and cannot be adequately determined or compared by measuring dcr. The later tweeter was more uniform and reliable, and integration with a simple parallel inductor (not really a huge crossover change) was the beginning of the modern design. In the early days, AR's approach was to design and integrate proprietary drivers to physically work together with minimal electrical modification. The unusual suspension of the early dome was a product of this approach. It should also be noted that the AR 3/4" dome magnet and voice coil gap remained the same. I think we are in the same place with this, just using different terms.
Along with Ken Kantor, former forum member, Carl Richards, provided response measurements demonstrating the above relationships.
Andre...Ken Kantor's first recommendation was to use an AR-12 tweeter with a .07mh coil. The lower impedance Tonegen and AR-11 tweeters behaved similarly with slightly larger parallel inductors. Later work with the HiVi tweeter found a .05mh parallel coil to bring it relatively close to the original response, but dispersion was somewhat reduced. The 3a Limited of the early 90's used a .16mh parallel coil, and 4uf cap. Variable level controls were very helpful in all cases. (Bear in mind that this all had to do with replication of the original AR sound, which may not be relevant to this conversation.)
The replacement dome under discussion can be made to work fairly well. Its primary differences are a lighter cloth dome than any AR paper or cloth dome tweeter, and a much larger/softer suspension...resulting in a stronger and extended low frequency output. The replacement's (4 ohm) voice coil's dcr is a tiny bit higher than the original, and should not be a problem. In the absence of measurement tools, under the dome damping and treatment and suspension modification would have to be determined through trial and error. Crossover modifications may also be required for fine tuning. The good news is the goal is to simply find something subjectively acceptable. (Bill, Based on your work with other tweeters this could be a project right up your alley.)
Roy
-
Lance,
The replacement shown in Andre's link is the preferred replacement and is closest to the original. The original roll was about 5/8 inch. Frankly, it really doesn't matter what the original dimensions were, as there are only two available. It is not like you can shop for anything else anyway. Fortunately the most common one (shown in his link) is the best fit. It is the one on the left in the attached photo.
The foam shown in your photo appears to be the "other" replacement, which is primarily characterized by a larger outer mounting lip. I see no reason to replace it if it is still functional.
Roy
-
It is often overlooked that crossover/response characteristics are heavily influenced by the mechanical properties of the drivers. Low compliance and high damping of the early tweeters simply resulted in progressively restricted movement of the dome at lower frequencies, yet large magnets were utilized to maintain higher sensitivity at the highest frequencies. Interaction with the midrange is affected accordingly. It's why the unusually stiff and highly damped tweeter domes of the original AR tweeters were used with just a single capacitor and no parallel components to cut them off at lower frequencies. It is also the reason a "crossover" design cannot be based on dcr alone, and why response measurements are required along the way.
When AR adapted the later, more compliant dome tweeter as a service replacement for the 3a, they reduced the tweeter circuit's original capacitance to 2-ish uf from the original 6uf by the addition of a series 3uf cap glued to the back of the tweeter (placing it in series with the cabinet's existing 6uf cap). In the AR-11, which had the same configuration as the 3a, (same stated high crossover frequency), they used the more compliant cloth dome tweeter with with an added .1mh coil to cut it off electrically in a similar manner to the mechanically rolled off 3a era tweeters.
-
19 hours ago, Andre_Db66 said:
I thought the intended use of the vc was to facilitate a drop in tweeter for any of the older models. LST, 3A/IMPR, 11, 10Pi and ultimately 5 with 8R coil. I do know there are differences between each XO.
I fail to see what should be done in the case of LST implementation though. As far as i know only a resistor has been applied in the craiglist offering.
Overall output and sensitivity are not the only parameters to be concerned with. Increased electrical resistance decreases output, but also affects the crossover point. Suspension compliance significantly affects the crossover point as well. The replacement dome you are discussing in this thread has a higher resistance (dcr), lighter dome, and a much more compliant suspension than any of the original AR tweeters resulting in a different response curve. A parallel inductor coil (not "vc", btw) can bring it closer to original by somewhat compensating for AR's (deliberately) stiffer suspensions. For the LST and LST-2, all the tweeters in the cabinet must be the same before you can begin work on the appropriate modifications.
In my opinion, based on your previous posts, the best you can hope for through trial and error is something personally satisfying. Nobody can do that for you...and your geographic location is obviously a handicap in that regard. Your best bet would be to start a thread in the mods and tweaks section to discuss your project.
-
Just now, ReliaBill Engineer said:
Pretty much, this entire thread has more to do with tweaks and mods than anything original AR. Just my 2 cents.
Agreed.
-
1 hour ago, Andre_Db66 said:
My point and preference is to have a replacement vc in the original magnet which first of all doesnt alter physical appearance.
Secondly by tweaking the setup internal and external ie xo and wadding, to obtain a workable solution generally suited as acceptable replacement unit in old school AR. Much like the Tonegem version. User tweakable by adjusting the potmeter on the back so that it can have more output than original if preferred but basically the same intonation. I find the mentioning of old tired originals appropriate (they lack output) and would like the same sound but with more output available...
Whether this topic moves to mentioned group is not important to me. It is by all means a mod and tweak anyway.... by definition
Andre,
It should be noted that AR made crossover adjustments to modify the response of the later AR-11 style of 3/4" tweeter for use as a replacement for the early 3/4" tweeter. Specifically, a smaller cap was used for it in the original 3a and, later, a parallel coil was used in the 10pi and 11. Over a decade ago, Ken Kantor provided us with a number of response measurements for the NOS original 3a tweeter as well as later AR-11 and 12 tweeters. At the time he suggested a similar solution for restoration. Additionally, a perfect modern copy of the AR-11 tweeter already exists: https://www.midwestspeakerrepair.com/product/mw-audio-mt-4121-75-inch-dome-tweeter/
Look at the AR-11 and AR-3a Limited crossovers to see how it was done from scratch with the later tweeter.
If you just want to change things to your ear you should head to the mods and tweaks section.
Roy
-
2 hours ago, ReliaBill Engineer said:
Who mentioned that for a drop-in replacement for the 4X? It wasn’t me.
I was just responding to Andre, who apparently misinterpreted your initial statement ("I may try a Chinese dome tweeter on my AR-4X speaker project...") before you clarified your intent.
-
1 hour ago, Andre_Db66 said:
I meant that the chinese vc in the 3a magnet is an intended replacement for a supertweeter. Implementing it in a 4ax xo means it will crossover much lower. I dont know the data but it is indeed 2 way meaning the crossover point probably is around 3.5k or thereabouts. The tweeter might not perform that well taking on more midrange. I have not tried it except for in a 3 way xo.
Monday i will attempt the rebuild of the AR5 tweeters with resonance damping behind the dome. I will also prepare some coils with 0.01mH as proposed by Roy, gradually reducing the inductance.
Andre,
-You're correct regarding the small dome tweeter not being suitable as a drop-in replacement for the 4x.
-The coil value would be .1mh (same as the AR-11), not .01mh.
Roy
-
2 hours ago, powerglide said:
what is the roll on the more compliant surrounds? That info is not on the speaker works site and they are out of stock.
The more compliant surround is 5/8 inch. The other one in the photo is closer to 1/2 inch.
-
14 hours ago, powerglide said:
thanks Roy,
any links to a preferred supplier?
I've use this company many times with good results but the O/A dimension for their recommended AR3 doesn't add up to me 265MM
The surround in your link should work, but it is not the best choice. See attached photo...The surround on the left fits better and has a slightly larger, more compliant roll. The surround on the right appears to be the one in your link. I have no idea who will ship to Australia. If they do it, will likely be very expensive. This is a good source: https://www.speakerworks.com/11-Inch-Angled-Speaker-Repair-Kit-p/swk11a.htm
Roy
-
2 hours ago, powerglide said:
They are 280 diameter which makes the standard replacement surrounds too small on their overall dimension, only leaves 5mm of gluing area!
Should I be removing this timber ring or is there a surround that fits
appreciate any help, rings could possibly contain asbestos?
Powerglide
The ring is just a type of fiberboard similar to masonite (no asbestos!), and it is required. The proper surround will fit perfectly.
Roy
-
11 hours ago, Andre_Db66 said:
Roy, you are saying that in all speakers the foam could be replaced with fibre glass to what advantage may i ask when foam has not been tested as a replacement? The fibre glass is annoying to work with, so wouldnt it be preferable to place foam in the 3a and lst? Both options seem to work.
Density, type, and amount of damping material is important to the original design. Along with Fs and Fc, Q is another important parameter to consider, so just replacing damping material with something else is not as straightforward as it may seem. (I forgot to mention above that Q was something else we measured). Modern fiberglass is really not that difficult to use, and its properties are well known for the restoration of AR speakers. Advents were manufactured with more than one type of material, but the amount of each type is critical to its original design.
You simply may not find AR speakers to your liking. You can always head over to the Mods and Tweaks section of the forum to discuss what you are trying to change, and may end up with something you prefer.
-
9 hours ago, genek said:
Pretty sure that the more technically knowledgeable of our CSP members (Roy!) would tell you that an AR woofer retrofitted with a rubber surround is not performing comparably with its original surround.
Yes he would.
Modern rubber replacement surrounds tend to be rather heavy and are primarily used for subwoofer and automobile driver repair. Additionally, I'm not aware of a replacement that would even fit the AR 12 inch woofer. If there is one, there is a strong likelihood that it would add too much mass and/or significantly reduce compliance.
Regarding other aspects of the discussion above...This is a restoration forum, and there are still a number of us who are very familiar with the original sound of AR speakers. We owned them when new, and have been working (as enthusiasts and professionally) to maintain them for decades.
Andre...regarding your cabinet damping concerns. When we were putting together the AR-3a restoration guide, we conducted dozens of measurements of various types, amounts, and placement of fiberglass, polyester, and "Acousta-Stuf". Response measurements consisted of Fs of the individual woofers and Fc after installation. Measurements were verified between us, and valuable insights were offered by Ken Kantor along the way. The upper bass/lower midrange area you are dissatisfied with is not likely to be influenced by the damping material. Foam rubber was not tested, but it should be noted that Advent restorers often replace the foam rubber blocks with the earlier fiberglass, and that later Advents were stuffed with polyester. It could be that Advent tried the foam rubber for awhile simply because it was easier to handle in the factory, or less expensive.
Roy
-
On 5/19/2024 at 12:29 PM, Linedoggy said:
Hello, I just acquired a pair of the Smaller Advent loudspeakers. Both woofer surrounds have completely deteriorated on both. I played them at very low volume: one sounds good, no vibration, the other makes a vibration sound (but when I gently push the cone in a little, the vibration disappears. Is it possible that by reforming the surrounds, the one that has the vibration sound might be okay? Just wondering.... Thank you.
Yes..it will most likely be fine when re-foamed.
Roy
-
13 hours ago, Andre_Db66 said:
So a smaller value cap and start with say .05 mH inductor?
.1mh would be a good starting point.
AR LST strange things about filters
in Acoustic Research
Posted
You're correct, Giovanni...it's what AR had available to build the appropriate capacitance value. Not sure what the "American way" would be, as this kind of variation was common to many LST's (as well as some other models). Capacitance was determined for the circuit design, and achieved with available parts. "Remodeling" and "matching" was not part of the process.
Roy