johnfalc

Thanks for doing the legwork on this. The only thing I have to offer from my own experience is that the 200027 and 200045, while they seem almost identical in many regards, use different cone materials and, at least to me (and others who commented here in earlier posts) sound different enough that it makes sense (if you have AR-9s or AR-90s - I have both) to use only the originally-intended 200027. My "discovery" of this occurred because a pair of AR-9s I purchased last year had the 200045 LMRs and didn't sound "right" until I replaced those with the 200027. Another column showing cone material (and perhaps even details specific to surround attachment) would be a nice addition to your work.

Glad to see you're back!

I did a very uncontrolled test by swapping between the newer (stamped frame) and older (ALNICO with cast aluminum frame) woofers ... I have two pair of AR-3a and these woofers are their originals in each case. I didn't hear any significant difference and didn't attempt any quantified measurement. I left the cast (older) woofers in my main pair of AR-3a (the one's in our living room). If a pair of the newer woofers has had their surrounds replaced with overly stiff surrounds then I'd expect perturbation of the LF response, something one dodges with the older, cloth, surrounds.

I run my AR-9s, AR-90s, and AR-3as all on Carver M-1.5t amps which are very similar (check the schematics and you'll see what I mean) to the PM-1200. I've been doing so for a number of years ... I generally buy M-1.5t amps that need repair and do the repairs myself. The PM-1200 and the M-1.5t share a three-tiered power supply to maximize linearity of output while keeping the output devices in their safe operating areas. The most common points of NATURAL failure in these amps at this point are the filter capacitors for the 36 and 76 volt rails and both the capacitors and circuit boards designed to ease the repair are available on eBay. The high voltage rail (+/- 126 volts) capacitors seldom need replacement. The most common points of UNNATURAL failure have to do with unqualified people working on and "buggering up" the amps, either in the output sections or in the circuitry that controls the conduction angles for the line-control triac. Once checked out and repaired on an as-needed basis, these are, to me, a great value in sound and reliability and the source clean high output nicely in to AR's 4 ohm (nominal) load. One final thought ... these amps do place significant demands on your home AC wiring when driven hard ... that can be an issue with old, low amperage, circuits (e.g., you may see lights dimming as the amps suck current from the line).

From a site with a review of the AU-D11 ... plenty of power for 4 ohms: http://www.hifi-classic.net/review/sansui-au-d11-332.html "The ultra-low distortion of the Sansui AU-D11 was demonstrated by its performance into 8-ohm loads at 1,000 Hz, where the distortion was between 0.0005 and 0.001 per cent for power outputs up to 140 watts. When the AU-D11 was driving 4-ohm loads, the distortion did not change very much, reading less than 0.0014 per cent up to 190 watts. Many amplifiers, over the limited power range through which they can drive 2-ohm loads, have substantially higher distortion driving such impedances than they have with higher ones. Not so with the AU-D11, whose distortion was under 0.0032 per cent into 2 ohms from 1 to 200 watts." I see similar performance when checking the AU-D11 with my HP 339A ... some clever circuitry (feed-forward in the current gain stage) and a solid integrated amp with a good MC/MM phono section.

I ran a Sansui AU-D11 with my 3a's and 90's to high levels with no problems at all for a year or two. This was, BTW, the original D11, not the D11 II. You'll have no trouble at all, I expect, with the 11's or 3a's.

Both of my sets of the 200027 (i.e., all 4) have eight mounting holes and seamless cones.

I really haven't taken the time to give the AR-90s a critical listen since I installed the correct drivers. I definitely heard (and measured) a difference when I removed the 210045 poly cones from the AR-9s and replaced them with the 200027 drivers. That difference had to do with balance/coloration in the lower mid-range ... most noticeable on female voices (I'm a big Kate Wolf fan). I never played around with the level switches when comparing, and the 210045 were "hotter" in their band-pass as noted in my original post on this thread.

Here's the eBay item number for those who might benefit from these 9LS / 9LSi 8" drivers: Item ID: 283735409408

FYI - I managed to find (on eBay) a good pair of 200027 LMR drivers which are correct for my AR-9s, so I've listed the 210045 8" drivers I took out of my AR-9s on eBay. These (the 210045 drivers) are correct for the AR-9LS and AR-9LSi and were replacement parts (maybe originals also, IDK) for some other AR speakers with 8" drivers. Anyway ... making this post to alert anyone who's looking for these - they're perfect except for a small tear in one of the surrounds. DCR on the voice coils both measure exatly 5.28 ohms. I refrain from posting the listing URL since I don't know that to be allowed on this board (which I value highly). John

I have a pair of these AR metal stands that I used for many years with one of my my AR-3a pairs. They fit perfectly and I liked the result.

Here's a LINK to Audio Magazine, July, 1955 which on page 18 has an easier-to-read version of the first article Pete B posted above. BTW ... the RADIO HISTORY site is a joy to me, audio nostalgia extraordinaire! The site itself looks like something I might have created (style-wise) 20+ years ago ... probably usable with Mosaic! Here's a link to its home page: https://americanradiohistory.com/ Oh, and just a warning, if you haven't been to that site before you might want to grab a glass of wine and plan on hours of fun time reading.

Sorry to hear of your illness and, BTW ... lovely examples of one of my favorite speakers. John

Very nice ... I always liked those best of the bookshelf models. I still have the metal stands for them.

Giorgio - Thanks but those have a DCR too high for application in the 90s or 9s. John