200028 vs 200032 1 1/2 inch MR

[Pete B]

Hi Roy,

It may be true that it is the best alternative at this time but I'm still not convinced

that they are really the same.

The suspensions may look the same but it is very possible that they were

treated differently for different compliances and that is the main source of the

difference. We should measure some with the front fiberglass removed, but

the two in question do not even have front fiberglass and Carl has measured

a difference in the resonant frequencies. It would seem that one is for a

three way and the other for a 4-way but both the 9 and 90 are 4-ways so it

doesn't really make sense. It is possible that early on they used the same or

similar driver to those used in the 3-ways and refined it for specific use in the

4-ways but there _should_ be crossover changes to go along with the driver

changes.

[RoyC]

The suspensions may look the same but it is very possible that they were

treated differently for different compliances and that is the main source of the

difference.

It is possible, but I doubt it, Pete. In Tim Holl's document above, he only mentions variations in the use of ferro fluid with the AR 1 1/2" dome midrange driver.

I was referring to the fiberglass under the dome in my post above. It seems the otherwise identical version of the mid used in the AR-58 had no fiberglass under the dome. ALL others, including the earliest mids and the AR-9xx era mids have fiberglass under the dome.

Roy

[ra.ra]

My interest in this thread began when Carl posed the original question, having myself become curious about any differences in these two mid-range drivers. This discussion is already over a year old and still gathering excellent thoughts and observations, and basically, I think everyone is raising valid points.

Not that I have ever performed any measuring, testing or dissection on these particular drivers, but there is much illuminating detail found in the AR production drawings, and this is what I want to share here - - - this information comes from the drawings in the library for the 200028 and 200032 mid-range drivers. Interesting to note that the original date on the 028 dwg. is April 78; while the 032 dwg. is dated one year later in April 79; disappointed to see no date on the Holl technical paper. In the several side-by-side comparisons attached here, 028 is on the left; 032 is on the right.

Just to summarize briefly: the 200028 mid was used in the AR-9 and 90 (four-ways), has a black plastic semi-horn with silver paint, crossover at 1200 Hz, and was tested to AR spec 206019. The 200032 mid was used in the AR-91 and 92 (three-ways), has a brown semi-horn with silver paint, crossover at 700 Hz, and was tested to AR spec 206023.

The section views appear fully identical, while the materials lists show three very minor itemized differences: the label (duh!), the semi-horn (see note above), and the ferro-fluid. Further, the semi-horns have exactly the same dimensional profiles. As best as I can determine, it appears that the 028 driver originally had 0.7cc of fluid, whereas the 032 driver had 0.4cc of fluid - - - this would seem to support the one side vs. both sides of voice coil theory. Reading the revisions column on the 028 drawing, it appears that the fluid for this driver was increased to 1.0cc in March 82. (Note on attachments: I have purposely included reasonably sized files here, so much of the fine print might get a bit blurry - - - for better views, use the Library.) Lastly, it seems that an original foam button was deleted from both drivers in June 80.

What to make out of all of this? With the fluid quantity being the only meaningful physical difference, the original designs of these two mids were indeed intended to provide different measured performance for their particular applications, and Carl's original tests seem to bear this out, at least for the drivers he was comparing. While I would be curious to know if similar findings held true for other measured comparisons between the 028 and 032 (i.e., are all of these old drivers still performing to original specs?), I am strongly inclined to keep in mind Roy's observations that he was unable to detect any appreciable sonic differences, and, from a practical standpoint, that either of these drivers probably still serves as the best alternative in restoration efforts.

[ra.ra]

Ane here are the full drawings.

[RoyC]

Thanks for the info, Ra Ra...

I do find it very interesting that ferro fluid was used in this way.

Roy

[ra.ra]

One further thought - - it occurred to me that there is yet a third version of what is basically the very same driver, and I became curious whether it, too, had any original performance characteristics that were also intentionally different from the 028 and 032. As has been noted in this forum, the somewhat rare AR-58s (three-way 'bookshelf', final descendant of the AR-3 series) employed a mid with part number 200044 (drawing and pic attached). In another line-by-line comparison, with the exception of the label and the semi-horn (all black), this driver is fully identical to the 032 driver - - - 0.4cc of fluid, tested to AR spec 206023. Summary: the 044 driver is exactly the same as the 032 driver, but with the 028 black semi-horn plastic faceplate, sans silver paint. The pic suggests that at this point, designers were a bit more conscious of "grille-off" appearance of the entire baffle board layout, which would be the only reason to create a new part number for this 044 driver.

More informed discussion of these drivers took place in this thread:

http://www.classicspeakerpages.net/IP.Board/index.php?showtopic=6137&page=2

My own experience is insufficient to determine exactly how, or maybe more importantly, just how well the construction of these drivers was able to contain and keep-in-place the original quantity of fluid. My blind guess is that, after 35 years or so, many of these mids have not been able to sustain 100% containment of the fluid, and therefore may now deliver performance characteristics that are not fully original. No doubt that Carl's pair (see post #1) were possibly testing near to original specs, but they might be the aberration rather than the norm. Just to add to what Roy has already stated regarding best-practical drivers for restoration efforts (balancing performance and availability), the 200044 driver should also be considered to be a suitable replacement for the various speaker models noted in this thread.

[Pete B]

Hi Roy,

It may be true that it is the best alternative at this time but I'm still not convinced

that they are really the same.

The suspensions may look the same but it is very possible that they were

treated differently for different compliances and that is the main source of the

difference. We should measure some with the front fiberglass removed, but

the two in question do not even have front fiberglass and Carl has measured

a difference in the resonant frequencies. It would seem that one is for a

three way and the other for a 4-way but both the 9 and 90 are 4-ways so it

doesn't really make sense. It is possible that early on they used the same or

similar driver to those used in the 3-ways and refined it for specific use in the

4-ways but there _should_ be crossover changes to go along with the driver

changes.

I must have been getting tired in the above post because I was originally going to

write that one was used in 3-way systems and the other in 4-way but then

changed it thinking that I read that they were both for 4-ways. ra.ra statements

reminded me and based on my above post it really does make a lot of sense

from a crossover design point of view to have different resonant frequencies.

I recall from my work on the AR-11 that even that mid had a resonant frequency

very close to the crossover point which makes perfect sense when designing for

and electro-acoustic 4th order crossover. One second order section is the

electrical network and the other second order section is the sealed box rolloff

of the midrange driver.

From memory, I believe the AR-11 and 3a had crossover frequency around 500 Hz

whereas the later 91 and 58s had it around 700 Hz.