Jump to content
The Classic Speaker Pages Discussion Forums


  • Content count

  • Joined

  • Last visited

Community Reputation

0 Neutral


About tysontom

  • Rank
    Advanced Member

Contact Methods

  • ICQ

Recent Profile Visitors

10,142 profile views
  1. AR9 Teledyn Big LF Caps where?

    You might want to consider measuring the caps before you replace them.
  2. Carl, I tend to be a purists on these things, and it's not always the best approach I must admit! The 28w Scan-Speak woofer is expensive but very impressive and has most of the other parameters close to the AR-12W woofer, and it would be literally bullet-proof with any kind of input power. It does seem very over-damped to me (too much magnet), but it could certainly be made to work! If it could be made to operate flat down to resonance in the 3/3a cabinet, that would be a good thing along with a lower mid-crossover, but the 3a midrange can't go much lower. Some interesting challenges. --Tom
  3. Carl, even if it would work, it is too small for the AR-3/3a opening. It is 279.9 mm (11-inches) in diameter, and the AR woofer is 12-inches in diameter except along the flat sides, so it would never work without some sort of special flange. It's also a massive, subwoofer-type driver and as such is designed for a very low crossover into the midrange. It has output well into in the low midrange area, but it is very peaky and rough up there—as much as 10 dB rise at 1000 Hz—thus it would probably sound terrible above 200-300 Hz. In contrast, the standard AR 12-inch woofer is +/- 1.5dB from 38-1000 Hz. It is also highly overdamped (-3dB at 45 Hz in a sealed 1.5 cu. ft. enclosure vs. -3dB at 35 Hz for the AR woofer in the AR-3/3a). It would add little and take away a lot. I think this woofer would be far better suited for its own custom enclosure with a dedicated equalized amplifier and crossover network. This same question was raised about using the NHT-1259 woofer (NHT 3.3 woofer) in the AR-3/3a. It would work, but it wasn't perfectly suited. —Tom Tyson
  4. Masonite Ring

    I'm trying to remember what I've heard from AR over the years regarding the exact reasons for using the Masonite rings. Part of it was due to the process of alignment and final attachment. The cloth surround and spider material, a treated linen, apparently works better being glued to the Masonite ring than to raw metal, especially during that final alignment and gluing process. A lot of speaker manufacturers used a paper or cardboard spacer above the surround where it is glued to the frame in order to give extra strength to the attachment. This spacer also allows some speakers with half-round surrounds to be placed "face-down" on a table without flattening the surround, but AR did not want that spacer on the outside of the woofer. Once this process was started with the earlier woofers, it was easier to continue the practice rather than completely redesign the layout of the moving assembly and frame, etc. Part of the process of manufacturing the woofers was to attach the Masonite rings to the basket during the initial process and then align and glue the cone assembly later. --Tom Tyson
  5. Masonite Ring

    I failed to add that the inner suspension required the raised portion to prevent the spider from bottoming on the aluminum frame. The frame has offset, and the Masonite ring gives additional offset to allow the woofer mechanism to move >1-inch before the spider hits the aluminum casting. I was not able to upload another image for some reason, but I will try to do that in another message. --Tom Tyson
  6. Masonite Ring

    The reason(s) for the Masonite ring was to position the moving assembly in the correct orientation to center the voice coil in the gap. The gap height (top-plate thickness) was 0.5-inches, and the voice-coil length was 1.125 inches in the original-style (#3700) woofer. Without the ring, the cast-aluminum frame (and the stamped frame of later 200003-0 woofers) would need to be slightrly deeper (a taller aluminum casting) to accommodate the proper depth of the cone/voice-coil assembly. The cone dimension/voice-coil assembly was fairly critical. Also, gluing the spider assembly to the frame was made easier by having the spider already glued to an inner Masonite ring, and the process of gluing the fabric material directly to the aluminum material was made easier by the use of the Masonite ring on both the surround ("skiver") and the centering suspension, the "spider." In some earlier woofers, the Masonite ring does come unglued from the frame, probably because of hardening of the glue. It is not an uncommon problem. --Tom Tyson
  7. Masonite Ring

    I think all of them from the very first models had the Masonite ring. AR-1 SN 0006: --Tom
  8. Ar10-pi Woofer control

    Thanks for your message and the description of your background. I hope you don't feel offended if I critiqued the changes that have occurred to your speakers, such as bypassing the woofer crossover circuit, changing the amount of cabinet Polyfill or fiberglass, and that sort of thing. Acoustic Research arrived at those parameters through testing and development, and none of that was done by "voicing" the speakers or seat-of-the-pants design. Good loudspeaker design is (and should be) a science more than an art—aside from the boutique designers over the years who sell exotic, cost-no-object speakers selling for outrageous prices. AR did not alter their basic engineering approach to speaker design because of a change of "taste" in speakers or even a change in engineering management over time, even though the company constantly investigated new ideas such as the tower speakers, vertical alignment of midrange-tweeter (Lambda), acoustic blankets, boundary effect and that sort of thing. "Taste" should never be a factor in determining a speaker's sonic accuracy. However, the basic engineering goals of uniform acoustic output, low distortion and wide dispersion did not change over time except to become more rigorous and detailed. It became computerized, and more and better methods of evaluation were developed to improve quality, durability and performance. The speaker-design process at AR was done more in the lab and test chambers than by simple trial and error. Obviously, there was a lot of tweaking and listening to the final product in the end; nevertheless, "taste" is a personal decision: something an individual likes or dislikes, but it doesn't (or shouldn't) have anything to do with accuracy or low distortion in the initial design process. For an optimized speaker design such as the AR9Ls or the AR9, changing one parameter by the end user will usually adversely affect another parameter—experimenting with the amount of fiberglass or Polyfill with change the Q, causing a peak or dip in output around resonance, which is undesirable. Changing the woofer crossover in the AR9 will have a hugely negative effect on both the flatness of response as well as the impedance, which we have discussed previously. In short, experimenting with parameters on these speakers—changing things around—is very different from changing the compound of the tires you are using, re-calibrating the fuel-management or spark-management system on your race car for best results. It's just not the same thing. Also, it's really not possible for an AR-3a to have better bass output than an AR9 or AR9LS, unless there are problems with the latter two speakers, and there is essentially no difference in the ultimate bass output of the AR9 and the AR9Ls. I think you are having difficulties with your AR9 and AR9LS systems because: 1. the proper amount of damping material in the enclosures has been changed and is incorrect; 2. the "Q" and impedance setting for the AR9s is completely invalid due to the removal of the resonant circuit in the woofer crossover; 3. Speaker phase (woofer) should be reexamined to be sure that a driver is not out of phase somewhere. or even the system. In the end no one is criticizing your approach or your decision to modify these speakers to try to make them better. They're obviously yours to do what you wish, but the performance will be affected by those changes. I hope you get them sorted out! —Tom Tyson
  9. Ar10-pi Woofer control

    I realize now that I misunderstood you regarding the open area in your basement. That open area to the side will also cause some problems in bass reproduction, but not as severe as having the rear wall missing behind your couch. If you reinstall the AR9's woofer crossover, you will restore a lot of the bass as long as you have not upset the proper amount of filling inside the enclosure. Without that Q circuit, bass response will be severely rolled above the resonance frequency. In other words, I wouldn't experiment too much with the crossover or the amount of cabinet fiberfill material. All of these settings were painstakingly optimized by Acoustic Research using their large anechoic and reverberant chambers to get flat response down to around 30 Hz, with strong output down to below 20 Hz in both the AR9 and the AR9Ls/AR9Lsi. By the way, I recently rebuilt a pair of AR9Lsi speakers for a good friend, and he has them in a large living room with plaster walls (25' x 18' x 10' ceiling) and he uses a large McIntosh 2275 amp. The amp goes into "Power Guard" protection frequently, meaning that it is surpassing 500-600 watt peaks. You can feel the bass even at fairly low output levels, and the bass output is so strong at times it rattles the windows and pictures on the wall. You can feel the bass anywhere in the room and right up through the floor on organ or electronic music. I guess what I am saying is that the speaker never has a sense of "lack of bass" that I could detect, but then again, his room is ideal for this type of sound reproduction. I own a Velodyne HSG-18 subwoofer, and it doesn't subjectively outperform my friend's 9Lsi setup in his living room, about the same size as our living room (8 foot ceilings). The Velodyne can ultimately produce more bass output down below 20 Hz, but there isn't a sense that it is much stronger than the 9Ls, from my subjective experience. --Tom Tyson
  10. Ar10-pi Woofer control

    One additional thought about the AR9 without the special circuit. If you bypass the woofer circuit, the "Q" of the speaker will drop to something like 0.3 or 0.4 or so, greatly attenuating the bass response in the low end of the speaker due to extreme over-damping. So, the special woofer circuit does more than just even out the impedance of the paralleled 4-ohm woofers to keep them from dropping below 4 ohms; it is also about getting the "Q" of the system into the normal range in order to have flat output down in the region of resonance. Some of that Q can be restored by reducing the amount of polyfill in the cabinet, but this can cause other issues throughout the bass range, too. There is an optimum amount of material to be in there for proper performance. --Tom Tyson
  11. Ar10-pi Woofer control

    Hi Carver, Thanks for your detailed description and pictures of your beautiful collection of AR speakers! That is an impressive collection of some of AR's best and most-beautiful speakers. I simply can't understand exactly what is going on with the bass response in your systems. I suspect that the AR9Ls might be accidentally wired (internally) out of phase somehow, because there should never be the impression of lack of bass from that speaker -- or from the AR9s. Note, however, that a hallmark of low harmonic distortion in a bass system is the "lack" of bass when none is there in the program material, but if there is known deep-bass content, you should be feeling it as well as hearing it. Something is wrong somewhere, and now I suspect that the "Q" of the 9Ls speakers might be out of spec if you have changed the amount of polyfill. But you say that the bass in the AR9s is not as good as in the AR9ls, and I don't understand that either. With the AR9's woofer resonant circuit bypassed, bass response would be somewhat out of whack, and I'm surprised that it doesn't cause your amp to get quite hot. Without the special crossover circuit, the impedance would be very low, on the order of 1.5-2.0 ohms in the 80-100 Hz and below 35 Hz range, and this would relate to a lot of current passing through the amplifier output. Some amplifiers could handle this, of course, but it could cause serious over-heating of some designs. All of the speakers in your collection have excellent bass response by any measure, but the AR9, AR9Ls, AR90s should provide the deepest and strongest bass. I don't understand how the AR-3a in your collection could have better or more powerful perceived bass output than the 9LS. It doesn't make sense to me. As for your 3-way AR speakers with 12-inch woofers, all of them have nearly identical bass output, and the only perceived differences would be due to the spectral balance differences of the midrange and high frequencies, and there shouldn't be big differences between the newer 3-ways (the AR-3a is a bit more laid back if it has the original tweeter). One big issue is that you have have described a large open space back into the basement area behind your listening couch, and you are not getting proper reinforcement of deep bass in your sitting position as you would if you were in a normal room. You could also be right in the middle of a standing wave, cancelling much of the low frequencies at your listening position. If you were to go back into the area well behind your listening area, you would sense a very different balance of low frequencies. But this does not explain why the AR-3a in your setup has stronger perceived bass output than the AR9, AR9Ls or perhaps even the AR90. The position of the woofer facing you or facing the side will have no effect on the amount of radiated energy, as those bass frequencies are omnidirectional. Well, it's a big mystery. Concerns are: (1) change of crossover components, (2) changes in the amount and arrangement of the polyfill material in the vertical cabinets, (3) location of equipment facing into a large area behind your listening couch and (4) possible phase issue with woofers. I also assume you are running your system "flat" without any filters or equalization through the amplifier-input system? Anyway, I hope we can help you figure out what's up with your systems. Describe, again, what changes you have made to the AR9 crossovers for the woofer and treble sections. --Tom Tyson
  12. Thanks. Both AR-3s are in good, but not excellent, condition. They work, but the level controls are badly oxidized. I have not opened them up yet to check everything out, but I will do that probably in the fall. At least they are original. One was missing a "3" badge, but I had an extra, so I fixed that issue. I have apparently run out of file-size memory on this site, so I had to compress this picture of AR-3 C-0005. Not only that, but the picture seemed to involuntarily rotate to the left, so I had to manually rotate the picture to the right! Who know how it will turn out. --Tom Tyson
  13. Ar10-pi Woofer control

    It's getting late here now, so I will try to answer in more detail at a later time. However, I don't understand what is going on with your description of the differences in bass output in the AR9 and the AR9Ls. Describe those differences again, and also describe how you have the speakers mounted (my assumption is that you have both in the same room for comparison) and what you have done to each speaker. First of all, the AR9Ls does not use the same impedance-matching network configuration as the AR9; the 10-inch woofer is a higher impedance than the 12-inch woofer, so the "automatic transmission," as they called it, isn't used in the AR9Ls to my knowledge. There is some crossover "tailoring," but not the same as in the AR9. Technically, the AR9 has a more potent setup with the two 12-inch woofers -- both with long-throw 2-inch voice coils -- but AR cleverly used a tuned chamber in the 9Ls for the 10-inch woofer (1.5-inch voice coil with less excursion), and it compensates for the difference with the tuned chamber output. As for the cabinet filling, I didn't have the impression that the 9Ls had filling all the way to the bottom woofer, but I can't recall the exact configuration. I think it has poly-fill material is used only in the top part of the cabinet, as with the AR9. If it was all the way to the bottom, you could have overdamping and the result of thinner bass sound. It is advisable not to change the weight and amount of filling in the cabinet one iota from the original configuration, as it will definitely adversely affect the smoothness and damping at resonance. Overall, the bass output should be pretty close between the AR9 and the AR9Ls, as both have the same -3dB point of 28 Hz., and I don't know which has lower distortion, but both are very powerful low-frequency reproducers. If anything, both speakers have been criticized for having too much of a good thing, but I've never heard of either of those speakers being considered weak in low-frequency output. In addition, both the AR9 and the AR9Ls also need to be back against the front wall, but the 9Ls can be pulled out from the wall without affecting the "boundary" dip issue. The 9 really needs to be back against the wall at all times with no object too close to either woofer. Anyway, send some pictures of your speakers and more description of what is going on. --Tom
  14. I still have AR-3s with serial numbers C0005 and C0020. I haven't seen any earlier than this, but I'm sure they are out there somewhere. I suspect that the very first ones were used for testing, advertising pictures and for audio shows. I also have AR-1 No. 0006, and it was one of the first hand-built version and a prototype of sorts, and I haven't seen any earlier than that one, either. It strangely ended up in a thrift shop in Palm Springs, California! --Tom Tyson
  15. Ar10-pi Woofer control

    No, there is no functional similarity in those particular circuits. The AR9's bass section crossover is a full-section bass-extension and Q-optimizing circuit because the speaker uses two 4-ohm, 12-inch woofers in parallel. Ordinarily, this would result in less than 2 ohms impedance at frequencies below and above resonance, which would destroy all but one or two power amplifiers trying to power it. The internal, sealed volume of the AR9 is 4.24 cu. Ft (120 liters), or roughly 2½-times larger than a single AR-3a or AR-11. The larger enclosure would also result in over-damping, thus making the speaker sound as though it had even less bass. The two woofers in the AR9 are the same as used in the AR-3a, AR-11, etc. Thus, the AR-9 uses a cleverly designed type of resonant circuit corrects the "Q" and which adds resistance in series with the woofers above and below resonance (the point in which impedance is highest anyway), preventing the very low impedance that would occur with the woofers wired in parallel. The AR9's system resonance is 31.8 Hz., with the -3dB point of 28 Hz., and this speaker is a powerful low-frequency system with few peers. In terms of low-frequency extension, low distortion and frequency-response smoothness, it's still one of the best speakers ever designed. The other advantage of the AR9 circuit is the nearly seamless integration of the two 12-inch woofers with the lower-midrange 8-inch driver. Few listeners realize that when listening to an AR9, you cannot ever detect the 200 Hz crossover frequency or the transition from the woofers to the lower midrange. There is nothing to identify the crossover from the woofers through the lower midrange driver, as it is seems like the output is only coming from the 8-inch driver itself, itself quite an accomplishment by the AR9's engineers. This is very difficult to accomplish, and this is the big issue with subwoofers: integrating the midrange and lower bass with deep-bass reproduction from an outboard subwoofer. --Tom Tyson