Jump to content

200028 vs 200032 1 1/2 inch MR


Carlspeak

Recommended Posts

The 200032 is the correct part for the AR-91, -915, -92 and -925. The 200028 is correct for the AR-9 and AR-90.

Your observation confirms Tom Tyson's comment that the 200032's lower resonance frequency is intentional, by design.

. . .


The AR-91 tweeter was identical to that used in the AR-9 and AR-90, but the midrange unit was modified. This unit has a different part number as well. The 1-1/2-inch midrange unit used in the AR-9 and 90 used Ferro Fluid on *both* sides of the voice coil, and sealed off the cavity under the dome, thus raising the resonance frequency of the dome; but in the AR-9 and 90, the crossover was set quite high and this was not an issue. In the AR-91 the crossover was 700Hz, so the voice coil in the midrange used Ferro Fluid on the *inside* of the voice coil only, and thus the cavity beneath the dome was larger, lowering the resonance frequency of the dome. It could then operate within the 700Hz crossover range more effectively. The small “semi-horn” appendage on the front did not affect the lower cutoff frequency of the dome, as it was designed to help maintain efficiency in the upper level of the operating range. It did nothing below 3kHz according to Tim Holl.

. . .

--Tom Tyson


Source: http://www.classicspeakerpages.net/IP.Board/index.php?showtopic=1331&p=56343

Personally, I like to use the correct part if possible, but if cutting corners, I'd be happier with a 200032 in an AR-90, than a 200028 in an AR-91.

Robert_S

Link to comment
Share on other sites

Having both 91's and 90's I find the 91 mid range to be thick and veiled compared to the 90 that's open and airy. I might prefer the 90 mids in the 91's.

What you hear may also be a function of the xover design and perhaps not the UMR per se.

Link to comment
Share on other sites

Carl,

I've compared AR 9, 90, 91, 58s, Cello Amati, later back-wired AR-3a/11/10pi mids, and 90's Tonegen replacements, and found some variations, but with no rhyme or reason. I have also dissected a number of them, and they all appear to be constructed exactly the same. One significant thing I found is the fiberglass under the dome will sometimes fall into the voice coil gap, causing response variations. I've seen a few that were almost completely bound with fiberglass! Perhaps the amount of fiberglass stuffed under the dome is the cause of response variations.

I suppose it is possible some had more or less ff...but the only differences I've observed between all of the back-wired versions of the AR-mid (after removing all faceplate adornments) are inverted role suspensions up until the late 70's, and thicker faceplates as time went on. From a practical point of view, the only significant differences between ALL of them seems to be the glued on plastic ring faceplates of the 9xx series and the glued on metal grille/fiberglass of the AR-3a/11/10pi era.

I recently used a number of AR-9xx series mids in AR-3a's after replacing the ring cover with the metal grille/fiberglass combo with excellent results.

The only really different version, imo, is the AR-5/LST2 higher impedance mid. All of the rest measure 3.5+/- ohms dcr, and are still the best possible replacements for each other as compared to any non-AR mid, past or present.

Roy

Link to comment
Share on other sites

I was hoping you where going to post on this Roy. I remembered you posted about this before. I just couldn't remember the details and if it included the 90 mids. The tonegens being similar is also interesting.

Harry.

Link to comment
Share on other sites

Thanks Roy for your input.

So, I guess the bottom line is the two UMR's in the topic line are indeed interchangeable.

From this info. I suspect the 200032 could be used in the AR90 but not the 200028 in the AR91 due to the differences in resonant frequency. Agree?

I recently made an acoustic measurement of an AR90 with the 200032 in place. I didn't check how the leads are attached, but saw a deep null around 1 kHz which suggests either the mid or tweeter is wired out of phase. I'll recheck this next time I set up to do similar measurements upon completion of the XO parts upgrade I'm about to undertake.

Link to comment
Share on other sites

From this info. I suspect the 200032 could be used in the AR90 but not the 200028 in the AR91 due to the differences in resonant frequency. Agree?

It is hard to know for sure. I've been having this conversation with Larry Lagace, who I've been helping out a bit, and who has provided me with many specimens (dead and alive) to inspect. We have mixed and matched the 4 ohm mids I mentioned above in a number of models, and came to the general conclusion they are all subjectively interchangeable. Perhaps the fiberglass-in-the-voice-coil-gap issue I mentioned above, or the differences in aging, are causing measurement differences at this point?

Larry and I have concluded that AR part number differences were sometimes due primarily to cosmetic differences (ie woofer dust caps, trim rings, color of flanges), which may or may not include slight performance alterations. The fact that 200028 and 200032 have different colored stick-on plastic rings may actually be the only intended difference. We would have to compare more measurements to know for sure, but in practice, any variations, intended or otherwise, appear to be insignificant for those of us trying to repair these old models.

Roy

Link to comment
Share on other sites

"... but in practice, any variations appear to be insignificant for those of us trying to repair these old models but in practice, any variations appear to be insignificant for those of us trying to repair these old models."

This is a major statement, one that runs counter to what we'd previously accepted as true and accurate. It had always been thought that the UMR in the 9/90 did indeed have a higher resonance than the mid in the 58/91/92. Tom's previous explanation about ff on "one side vs. two" may or may not be exactly right, but in any event there was always thought to be a difference in the two domes' resonant frequencies, and thus their usable passband.

If indeed all AR 1 1/2" dome mids are essentially interchangable (assuming the same z), it's a revelation akin to the discovery that the 2a/old 2ax and the 3 all used the same 4-ohm 1 3/8" tweeter--whereas previously, it was thought that the 10" models used an 8-ohm version of the 3's 4-ohm tweeter. Not so, after all.

If the 91 and 9 use the exact same 1 1/2" dome, then that's big news.

BTW, Classic AR-ADD-Vertical series speakers pushed the heck out of their drivers, didn't they? Using a 500 Hz resonant driver and crossing it over at 700 Hz (or using a 900 Hz resonant driver and crossing it over at 1200 Hz) was typical of AR in those days.

These days, conventional design practice is to cross a driver over at least an octave above resonance, for lower distortion and better power-handling in the application.

For the 3a or 11 to use a driver spec'd at 485 Hz FAR and cross it over at 525 Hz was nuts. Nuts.

Steve F.

Link to comment
Share on other sites

"... but in practice, any variations appear to be insignificant for those of us trying to repair these old models but in practice, any variations appear to be insignificant for those of us trying to repair these old models."

This is a major statement, one that runs counter to what we'd previously accepted as true and accurate. It had always been thought that the UMR in the 9/90 did indeed have a higher resonance than the mid in the 58/91/92. Tom's previous explanation about ff on "one side vs. two" may or may not be exactly right, but in any event there was always thought to be a difference in the two domes' resonant frequencies, and thus their usable passband.

Steve F.

Hi Steve,

I have had the opportunity to work with dozens of these mids, and stand by my statement. One thing is for sure...ALL of them have the same dome size, construction, and voice coil. What AR was doing with damping fiberglass under the dome, and ferro-fluid is anybody's guess, but in practice the sonic differences are not apparent. Once the metal grilles and plastic rings are removed they are identical, all looking and sounding like the naked Cello Amati AR mid, of which I have four. When looked at from a restoration perspective this provides more possibilities.

I completely agree regarding the crazy implementation of these mids. :rolleyes: It contributes to the unique nature of the various AR-3a iterations, but power handling is always a concern.

I've attached a photo of a project 3a cabinet I'm currently using to test modifications and crossovers for the AB Tech tweeter. Can anyone spot the AR-92 midrange in the photo? It sounds great!

Roy

post-101150-0-48381700-1362591975_thumb.

Link to comment
Share on other sites

I have no reason to doubt that all these dome mids are essentially the same.

Here's an interesting question (two actually):

1. How closely-related are the original AR 1 1/2" dome mids to the one Ken used in his 302/303? They certainly look similar from the outside. I'm sure that was intentional.

2. Ken's 338 from the 302/303 series was an 8" 3-way with a 1 1/2" dome mid. However, the 338's mid did not have the 3a/5-styled metal mesh/fiberglass cover like the 302/303 did. The dome mid in the 338 was cleverly called "exposed dome technology" as if that was something special and beneficial.

I wonder if the 338 and 302/303 dome mid were essentially identical except for the cosmetic metal mesh covering? I'd like Ken to chime in and tell us if the 3a/5-look that was used in the 302/303 was just for the sake of nostolgia (nothing wrong with that, BTW--it's called Good Marketing and knowing your customer), but the actual acoustic performance of the 303-302-338 mids were all about the same?

Does anyone have experience either comparing 3a-91-9 mids to 303's or comparing 303 mids to 338's?

Just curious.

Steve F.

Link to comment
Share on other sites

  • 4 weeks later...

These midrange drivers were definitely not the same. Yes, the magnet and voice coil/dome were virtually identical in most versions, but one quick observation should be obvious: the half-roll skiver on the earlier midrange driver was inverted whereas the later ones used in the AR9, etc., without the fiberglass and screen over the top, were in the more conventional shape. This likely affects system performance in a subtle way, but the only way to definitively know that there were differences is to actually measure the drivers in an anechoic chamber and to measure the driver's resonance and so forth. True, these drivers were fundamentally the same up through the series, but there were subtle differences in performance. I have anechoic measurements on many of these versions, and at some point I will post the fr curves for these drivers.

--Tom Tyson

Link to comment
Share on other sites

but one quick observation should be obvious: the half-roll skiver on the earlier midrange driver was inverted whereas the later ones used in the AR9, etc., without the fiberglass and screen over the top, were in the more conventional shape.

--Tom Tyson

Hi Tom,

Based on many observations and dissections, the inverted roll appears to have been discontinued with the front-wired version of the midrange. All back-wired versions I have seen, including the version used in the later AR-3a, the AR-11 and 10pi, have the more conventional roll suspension. All voice coil measurements are the same as well. That pretty much leaves ferro fluid and the amount of damping material under the dome as the most likely variables.

I am looking forward to any AR data you may have.

Roy

Link to comment
Share on other sites

Hi Tom,

Based on many observations and dissections, the inverted roll appears to have been discontinued with the front-wired version of the midrange. All back-wired versions I have seen, including the version used in the later AR-3a, the AR-11 and 10pi, have the more conventional roll suspension. All voice coil measurements are the same as well. That pretty much leaves ferro fluid and the amount of damping material under the dome as the most likely variables.

I am looking forward to any AR data you may have.

Roy

Roy,

Well, that's not quite right either. I don't know exactly when the inverted-roll suspension was initiated, but it was not with the beginning of the back-wired, recessed-screen version of the midrange. A dimension drawing of this back-wired version of the midrange clearly shows the inverted skiver.

By the way, the response curves of these drivers are invariably shown with crossover in place; therefore, some response-shaping is occurring as a result of the crossover. AR rarely did raw-driver curves except for the initial development. Nevertheless, there are lots of curves (literally thousands) on the midrange driver in various configurations and development, along with the updated iterations with the semi-horn, etc. The fact that the semi-horn drivers had various and different part numbers -- even though the drivers were identical in appearance -- indicates that there were differences when used with different model speakers. For example, the UMR used in the AR9 was not completely interchangeable with the midrange used in the AR-91, and so forth. But what you say about the construction of the voice coils, magnet structures and so forth, is true in that there were very little differences in these drivers over the years except for the fiberglass or lack thereof, semi-horns, crossover differences and the use of Ferrofluid cooling in the coils.

--Tom Tyson

post-100160-0-48511500-1364612150_thumb.

Link to comment
Share on other sites

Tom...Thanks for posting the drawing.

If the only data available was collected by measuring the mids in their various applications, differences are bound to be the result.

Imo, for today's restoration and repair purposes, the only truly useful way to compare them is out-of-cabinet and on the bench, with the horns and grilles removed. That is what I have done, and I am not seeing the alleged technical differences in practice or application...whether the suspension rolls are inverted or otherwise. The bottom line is, in the absence of the original mid, any of these mids may be used in the last version of the AR-3 right through the AR-9 series by simply swapping faceplate adornments. It is fortunate that this is the case, as we often have no other choice these days.

As for part number differences, I believe the only differences between many AR drivers had to do with dust cap variations, trim rings, color, etc...and had little to do with technical differences. I'm seeing many AR drivers on a regular basis working with Larry, and I'm measuring and comparing everything I can get my hands on. As an interesting aside, I am seeing differences between the various iterations of the AR-4x and AR-2ax/5 woofers, with no variation in part numbers. The differences (some significant) are primarily related to Q measurements.

Steve...All I know about the AR-303 style midrange is that the same one was used in all the models using that type of mid. The magnet and structure look much like the last AR Tonegen replacement for the mids discussed above. The only differences I am sure of would be the 6 ohm impedance (vs 4 ohms) and the position of the connecting terminals in the back.

Roy

Link to comment
Share on other sites

Roy and Steve,

The point I was making is that there were subtle differences, and that all of the 1-1/2-inch drivers were not exactly the same. Overall, I agree with you (Roy) that the voice-coil measurements (except for impedance differences) and magnetic flux and so forth were nearly identical throughout the series. Once stripped down, they all look pretty much the same except for the inverted-roll skiver in the early versions. But you must remember that the versions with the screen on top would definitely have a different response characteristics from those that were open in the front, such as the AR9 versions and the AR-3 special replacement version. Those with Ferrofluid damping would be different, as well, from those without it. Besides, there were many, many iterations of crossover differences among the many AR speakers using that driver, so one could not just replace a bad driver with any version of the 1-1/2-inch MR as a "drop-in" replacement. I am not sure that the amount of fiberglass damping material under the dome was the same across the entire series; it may be that the original version had a bit more fiberglass to offset the amount on the outside of the unit. But the treated-cloth dome and the aluminum-former, copper-voice coil midrange and the ceramic magnet/Armco Iron circuit were pretty much unchanged throughout the history of that driver. I think that is the point you are making. It's probably true that you could drop in any version of the midrange driver into any of the AR speakers without too much problem, but the only way to know for sure would be to run some frequency-response curves on baffled drivers to compare their performance. I'm guessing that the Ferrofluid models, such as used on the AR9, might not do so well in the original AR-3a speaker without some crossover modifications to compensate.

Steve, the AR-303 1-1/2-inch midrange is nearly identical to the earlier AR-3a version, but the magnet was somewhat smaller, probably indicating that KK and Bill Bush decided that the original design was -- if anything -- a bit over-damped with regard to magnetic flux and so forth. They were probably able to get the same basic performance with the newer midrange without the massive magnetic structure used in the earlier AR mids. The attached measurement was made out at AR in California prior to the Julian Hirsch AR-303 review. The AR-3a pair belongs to me, and they wanted to measure them before sending them on to Julian Hirsch for the review. Both speakers have a very similar response profile, especially throughout the midrange, even though there were sensitivity differences. The 303 curve, on the bottom, shows an actual rising characteristic from the mid-bass down to resonance, but it is actually smoother overall than the AR-3a. The AR-3a's tweeter roll-off is evident in the comparison, but the integrated power response would be closer due to the 3a's superior dispersion.

The second attachment is a fax I received from Julian Hirsch shortly after he did the AR-303 review in Stereo Review magazine. From his comments it can be seen that he felt the 303 was very similar to the 3a, but superior in response smoothness, etc. He was extremely impressed with the AR-303's low-frequency response, and I think it is evident that the excellent design work KK did on this speaker is very evident in this respect. I also noticed that his reply was hurried and had some errors, probably indicative of his progressing, slow deterioration through his final years. To this day I consider Julian Hirsch -- through his insight and technical review in print -- to be the most important influence to the field of high-fidelity sound during the formative period of the early 1950s through the mid-to-late 1990s. He will always be missed; I used to wait with eager anticipation to his comments and reviews in each issue of Stereo Review! And Steve, I think you visited with him on one occasion, so I think you also felt the same way.

--Tom Tyson

post-100160-0-50523000-1364662309_thumb.

post-100160-0-15072400-1364662858_thumb.

Link to comment
Share on other sites

I'm guessing that the Ferrofluid models, such as used on the AR9, might not do so well in the original AR-3a speaker without some crossover modifications to compensate.

--Tom Tyson

Tom,

The later mids are more sensitive, but whatever slight overall response differences exist compared to the front-wired version are not great enough to require a crossover modification (unless the AR-3 is not of the last version manufactured with this type of mid). For the most part, the level controls work well to compensate for the sensitivity issue. The later back-wired 3a to AR-9 series type of mid with the appropriate, transplanted, faceplate adornment sounds just fine in the AR-3 as long as the last version of the AR-3 crossover is used. Granted, in the case of the AR-3 and front-wired 3a, replacing the mids in pairs is the safest way to go. We should bear in mind that AR was replacing the front-wired midrange with the back-wired version from the mid-70's on when repairs were required. I should add that the differences between the various iterations of the AR 12 inch woofer have proven more discernible than the mids.

One of the reasons I began to explore this issue was the revelation that around 95% of AR-3 mids I have seen in the recent past have serious response issues. The small differences between the later mids we have been discussing above are absolutely trivial in comparison. The AR-3 mids are showing the same type of suspension problems the early dome tweeters have... mainly stiffening and reduced/compromised output.

Roy

Link to comment
Share on other sites

I realize I'm drifting a bit off topic now, but I'd just like to add that I agree completely with Tom about Julian Hirsch.

I knew Julian very well. I was at the "Hirsch-Houck" labs several times for the different companies I was with over my decades-long career when he reviewed one of our products. I remember him taking me upstairs to their living room and him showing me the speaker system he'd selected as his personal favorite to use in his own system.

He swore me to secrecy forever, and I've never broken that promise. And I won't, so don't ask! But, no, it's not the AR-3 or 3a!

When he passed away in late 2003, I wrote an e-mail letter to the then-publisher of Stereo Review, Bob Ankosko. My letter was published virtually verbatim in the magazine very shortly thereafter.

I have attached a copy here.

Steve F.

Julian Hirsch.pdf

Link to comment
Share on other sites

  • 2 weeks later...

Just a quick note of appreciation for this thread. I just bought a pair of back-up 200028 UMRs for my recently acquired AR-90s. It is interesting to see the different prices people ask for these on eBay; it is as though the don't realize they're all participants in what is, in essence, a single market.

Link to comment
Share on other sites

  • 11 months later...

Thank you, Robert_S, for bringing back this relevant thread that my foggy memory was attempting to recollect. Your posts are always among my very favorites on this site, but it just now dawns on me why you have such an interest in these particular speaker models - - - looks like you've got an AR-91 or 92 in your avatar along with a healthy jade plant and the corner of a Mark Rothko painting (we can wish, yes?).

Of equal interest in this document by Tim Holl is the confirmation that the AR-92 utilized an entirely new woofer, different in fact from the one used in the AR-90, which I had suspected after looking at the various assembly drawings in the library. The magnet structure is explicitly mentioned, but no mention is made of cone, spider, voice coil or surround components.

Link to comment
Share on other sites

  • 4 weeks later...

I just want to add that given that the resonant frequencies of the two midrange

drivers in question are so different they are not interchangeable. It was made

very clear, many, many years ago that the electroacoustical response of a

driver such as these midranges has to be accounted for in the crossover, in

fact the 3a employs something very close to what we now call a LR 4th order

crossover where the 2nd order highpass respose of the mid is one section

of the design. The resonant frequency and Q are both important to provide

the proper response.

The mids in question probably have different suspension compliance and

perhaps moving mass.

Link to comment
Share on other sites

Hi Pete,

Since this thread was started I have had a chance to examine more of these mids. Frankly, after the introduction of the back-wired AR mid around 1975 there are no easily observable physical differences between AR dome mids with the exception of the faceplate adornments AND the amount of fiberglass under the dome (which is where the resonant frequency issue you descibed above comes into play). After lots of cutting and dissecting I am fairly certain there are no differences in dome mass, suspensions, size etc. Once the metal grills and the plastic donuts are removed from the faceplates they are all the same, with the exception of the amount of fiberglass under the dome...and the reported variations in the use of ferro fluid, which I have not yet been able to verify.

I have used these mids interchangeably a number of times now and, after installing the appropriate faceplate treatment, I would defy anyone to hear the difference under typical use. I don't disagree that there are technical differences, but given the scarcity of good used parts for restoration there are often no alternatives. Whatever differences may exist between these siblings, they are very minor compared to any potential modern day replacement driver.

It should also be noted that subjective differences between later AR service replacement woofers and tweeters and their original counterparts are much more noticeable than anything I have been able to discern between the mids under discussion.

Roy

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...