AR-3a Possible Tweeter and Hello

[dynaco_dan]

Here's a modern version of your Microstatic Array that can be purchased today. The spec's freq. range is 1 kHz to 16 kHz. That range goes too low for an AR3a application. I'd install or, replace whatever caps are in there with a 1 uF cap like the Microstatics have. The upper freq. range is just as good as the Microstatics.

Oh, and the price is quite reasonable.

http://www.mcmelectronics.com/product/555-10345

Hi Carl

Thank you for that link.

My concern is they are advertised for PA and DJ work only, nothing even suggesting for hifi use.

The price is less than the value of the drivers or just the cabinet alone.

A steal, if they're hifi friendly.

[Carlspeak]

Hi Carl

Thank you for that link.

My concern is they are advertised for PA and DJ work only, nothing even suggesting for hifi use.

The price is less than the value of the drivers or just the cabinet alone.

A steal, if they're hifi friendly.

I'm not so sure the Microstatic Arrays were that 'hi-fi' friendly either with the cone tweeters.

[dynaco_dan]

I'm not so sure the Microstatic Arrays were that 'hi-fi' friendly either with the cone tweeters.

Hi Carl

I do not remember, but there may have been, very bad review's or very bad commentary's, of the Microstatic Tweeter Array speaker system, when used with AR or KLH speaker systems.

Yes, they are not perfect and the number of them sold is still up in the air.

A brochure they gave out had, on the cover, zillions, ok, not zillions, maybe many dozens of partially

finished enclosures stacked high, it appeared that they were selling them in quantity.

They did get a lot of good press, and of course they are not absolutely without shortcomings, but they were made to

suit a limited purpose.

I have yet to hear or read that someone threw there's in a dumpster because they didn't do what was claimed of them, increase wider dispersion.

We now know that there may have been at least, 5 versions or variations, including the original, Peerless, dual sized tweeters.

We will never know which versions were even being tested back then and whether they were equal in sound dispersion output, etc, as they evolved.

The same as Dynaco A-25's, AR-3's and AR-3A's, just to name three similar examples.

Just a thought for today.

[genek]

Mine are all the dual-size Peerless tweeters, and the sound output is quite similar to that of the original AR tweeter. I think the reason they don't make that much of a difference on my 3a's and 2ax's is because the room is highly reflective. But the 6's have tweeters with narrower dispersion to start with and are in a more average office, with carpeted floor, drapes, etc.

[dynaco_dan]

Mine are all the dual-size Peerless tweeters, and the sound output is quite similar to that of the original AR tweeter. I think the reason they don't make that much of a difference on my 3a's and 2ax's is because the room is highly reflective. But the 6's have tweeters with narrower dispersion to start with and are in a more average office, with carpeted floor, drapes, etc.

Hi Gene

Can I guess that your Peerless set has the screwed on back cover, rather than glued on.

As I mentioned earlier, when turning off the Microstatics, after using them for two relatively noisy days, I thought, "where's the AR-3A's tweeter gone".

A dramatic difference, black and white, in the loss of mid and tweeter output and/or dispersion.

I also had a pair of JansZen 1-30 tweeter array running in parallel.

There were stacked above the 3A's tweeter and I was just using my AR amplifier, tone controls set flat and also all of the speakers were set at maximum.

There was not enough time to evaluate whether or not listener fatigue would set in, with that added brightness.

I would have been divorced that weekend if I had bought them, but at least some friends and I had that positive, but brief sound experience.

It would have been nice to have had my loaner Crown equipment that weekend as well.

[genek]

Can I guess that your Peerless set has the screwed on back cover, rather than glued on.

Two pairs with screw-on backs, one that was originally glued on but which I converted to screws after the glue failed and the backs came off. Same HW in all of them.

[dynaco_dan]

Two pairs with screw-on backs, one that was originally glued on but which I converted to screws after the glue failed and the backs came off. Same HW in all of them.

Hi Gene

What I was suggesting was that perhaps the backs were screwed on

until they changed from Peerless to ? brand tweeters and also the same size for all four drivers.

These were un-announced changes to that same model speaker system.

I also realize that a dab of glue is cheaper and faster than screws and also the labour to install them.

[genek]

What I was suggesting was that perhaps the backs were screwed on until they changed from Peerless to ? brand tweeters and also the same size for all four drivers.

All three pairs use the different-size drivers scheme, but their cases were originally routed to accommodate same-size drivers, so I presume the company allowed for the possibility of driver changes from the beginning. But as far as I can tell the screw/glue backs are the only differences between the ones I have.

[dynaco_dan]

All three pairs use the different-size drivers scheme, but their cases were originally routed to accommodate same-size drivers, so I presume the company allowed for the possibility of driver changes from the beginning. But as far as I can tell the screw/glue backs are the only differences between the ones I have.

Hi Gene

Is it possible to please post a detailed photo of the inside of each cabinet, Gene?

Also please post their serial numbers or in consideration of your privacy, digits please.

We can see what span of serial numbers we can come up with if all members post theirs as well.

I will look at mine to see if they have serial numbers, even the dead ones off ebuy.

Are you saying all of your's have dual Peerless brand tweeters, not any other brand?

Interesting about the routed out hole for same size tweeters.

The earliest of course was the dual size and somewhere down the road they switched to same size tweeters and several different brands.

I would have thought that the dual sized routed cabinet openings would been their first issue and then identical sized holes when they later changed sources.

My feeling used to be that they wanted to keep out owners when they changed tweeters, so they began glueing the backs on.

It did keep prying eyes out, at least for a while.

I've never read how many were produced or for how long.

Perhaps we should move this to the, "others," forum, Gene, what do you feel?

There is already at least 2 Microstatic topics quite recenetly started when I first posted my scanned review of them.

This is kinda focusing on the Microstatics now, which I do enjoy reading about, but this is the AR forum.

[genek]

At some point I will probably gather up all the Micro Acoustics posts and move them all at once. If there are enough of them, maybe we'll make a Micro-Acoustics forum. New York state is sort of close to New England.

[JohnHeubel]

Here's the latest iteration of my faceplate. The tab system, though workable, isn't very easily done to get all the holes aligned properly. So, I found and old license plate, I guess it's about 1/16" thick, and since it's metal, will stand up better to being screwed to the cabinet than the masonite. As you can see in the pics, it's virtually at the base of the dome, so dispersion should be basically as good as no faceplate, while looking much better. I think I may keep it brushed metal, but may still paint it black.

[JohnHeubel]

Here's the final tweeter iteration and the painted grill badges. I ended up going black with the faceplates since they're not perfectly round and will not be as noticable against a black baffle.

[JohnHeubel]

I decided to hook the "bad" tweeter directly up to a receiver today (thru the 6uF cap) and it worked. It always had good continuity once I repaired the tinsel wire on the front. But, I was getting a 1 ohm reading which seems way too low. I did a quick, rough comparison to the good (3 ohm Re) tweeter and the sound and output is about the same. Any ideas?

Here's the new crossover. I wired it for both front and rear hookups so I could try out the original pair vs the Peerless as a pair.

[DON]

Nice looking work John. I've enjoyed following your progress on these speakers. Keep it up please........I might just learn something.

[JohnHeubel]

Thanks Don. I'm listening to them right now, but with the Peerless tweeters in there. I still haven't re-stuffed all the fiberglass, but they sound mighty nice as is. I did hook everything original up (the last cap came today, more about that later) and there was virtually still no highs out of the one side. So, though it appears functional when directly hooked up and with no other musical info,

I just prefer hearing the highs that the Peerless has. Pics later of the finished product.

[JohnHeubel]

Well, here's the almost finished pics (I still plan to redo the grill cloth since there's a hole in one). Still, they're sounding very nice, and no inductor. I wonder if the higher impedence of the HiVi and other replacements (hence lower crossover point with the same 6uF cap) is what's causing the need for a steeper slope into the mids. Any thoughts on this? 1st 5 pics are self-explanatory. The last pic shows the new 14ga jumper vs the stock one. Do't know if it makes a difference (I doubt it sonically) but it sure is sturdier.

John

[JKent]

Beautiful work! Check the AR 3a Restoration guide for info on ordering linen from 1-2-3 Stitch. You "could" get away cheaper with linen from Michael's or a fabric store, but the 1-2-3 Stitch stuff is super-nice and I'd say your restoration deserves that treatment.

How about posting directions on fabricating the faceplate. You know; "first get one flat aluminum license plate" What state? Delaware?

Happy listening!

Kent

[RoyC]

I wonder if the higher impedence of the HiVi and other replacements (hence lower crossover point with the same 6uF cap) is what's causing the need for a steeper slope into the mids. Any thoughts on this?

John

It appears you are making an assumption that the HiVi (or other modern tweeters) require the crossover mod/inductor, and your Peerless tweeter does not...simply based on your satisfaction with the sound. You don't know, however, what any of the other modern tweeters sound like, with or without the inductor. If the goal is to approach the response of the original tweeter, a crossover mod is needed. The inductor brings modern tweeters, including yours, closer to the original tweeter's response. Measurements would confirm this. If they sound good to you, don't worry about it. If you want to know why, much more work needs to be done. Impedance is a contributing factor to a driver's response, but it is only one of many.

Your project looks great!

Roy

[DON]

Beautiful job John. That grille cloth is somewhat rare.......the same as on my European AR2ax.

[JohnHeubel]

Thanks for the comments. Don, perhaps I'll fashion completely new masonite frames and cover those with the linen (though I'll most likely use the burlap I have from another project) if the grill cloth is somewhat rare. Perhaps stiching the hole and doing a light cleaning would help.

Roy, I'm probably oversimplifying things a bit, and no, I haven't heard the modern replacements with/without inductors added, but that was one of the reasons for my question. In your testing of various replacements, was the determination about the inductor made by measuring the original tweeter's response curve "out-of-system" and then seeing what needed to be done to the replacement "out-of-system" to get its response curve to match (i.e. a very analytical approach) or was it determined after wiring it into the crossover to see how it reacted in the entire system?

I'm just trying to increase my understanding of some of the nuances here, because in my simplistic way of looking at the situation, the crossover limits what frequencies the driver "sees" and has to reproduce. So, assuming the substitute driver is at least as capable as the original at reproducing the band of frequencies allowed by the crossover, the worst thing that happens is you'll be "wasting" some of the capabilities of the new driver. Different materials, etc will effect the tonality so again, I'm simplifying the discussion and "throwing out" some variables. Obviously, if you try to sub-in a driver that can't cover the band needed, it's not going to work.

Now to the specific driver and determination about the interaction of the tweeter with the mid. You stated the reason that the inductor was needed was to "roll off" the low end to simulate the mechanical damping (?) of the original driver. I'm guessing that it was because with more bleedover into the mids, they became too forward and out of balance? But if the crossover is limiting the frequency bands of the drivers, it would seem to me that the problem area would have to be with harmonics of the driver. I don't know too much about 2nd, 3rd order harmonics so it's just a guess. BUT, if we simplify the crossover to just look at where the frequency shifts, in a simple 1st order crossover a 6uF cap feeding a 4 ohm driver gives a crossver frequency of 6625Hz. That same 6uF cap feeding a 6ohm driver shifts the crossover frequency lower to about 4416Hz. That's obviously going to affect the prominence of the midrange, bringing it much more forward. In your testing, did you happen to try a 4uF cap for the tweeter to keep the crossover frequency about the same as the original?

As I stated in an earlier post, I did try out the good original tweeter and found the highs rolled off more than I like. So, yes, my goal is not necessarily to mimic the original tweeter's response curve. What I observe with the Peerless tweeters wired into the system is that, on the white dot setting, they seem hardly there, fairly laid back, and I was finding that I preferred the sound with the Hi pot "increased" a bit to between "9 and 11 o'clock" if the knob were the center of a clock face and the white dot setting was 6 o'clock. I did end up adding a little extra fiberglass because, when I restuffed the cabinets, the woofer section seemed a little too empty. I didn't even take the insulation out of the tweeter and mid area, so I don't know why it looked too empty, it just did. Could the extra stuffing have had an effect on the presence of the highs? Maybe that's why I don't see the need for the inductors, though if I had them on hand, I'd probably wire one in and see if I could tell a difference. Anyway, it's been a fun process this restoration.

John

[JohnHeubel]

It's official...these speaks will get to stay! Did some more listening tonight with material that's not only very familiar, but also very revealing and spatial. The speakers just disappeared. I did set a recliner in the sweet spot, so to a degree I should have expected some of that effect, but there was a really nice depth and height to the soundstage. Brought a smile more than once as I listened and they're not even in the main system with the more powerful amp yet. It's actually kind of funny how little the system cost which I use in the basement where I do most of my work. Pioneer SX-780 (free...bulk trash day, but now completely re-capped, probably about $20-$30 in caps since I replaced the main cans also http://www.audiokarma.org/forums/showthread.php?t=295300) and a Sony CD player (also free on a bulk trash day).

[JKent]

5 bux for AR3a's and piles of cool free stuff curbside. I want to move to YOUR neighborhood!

Nice work on the Pioneer btw

Kent

[RoyC]

Roy, I'm probably oversimplifying things a bit

Yes, you are.

The integration of a potential replacement driver's inherent mechanical and electrical characteristics with an existing crossover is not at all simple. As I said before, the replacement candidate's rated impedance alone will not tell you a thing. Different tweeters with exactly the same rated impedance will ALL have their own sonic character with the very same 6uf AR-3a capacitor. The crossover frequency and response curve is not determined by the driver's rated impedance and crossover capacitor alone.

The intent was not to re-engineer the 3a with the HiVi and ABT tweeters, or try to make the 3a "better". Their response into the midrange needed to be reduced to be similar to the original tweeter's characteristics (measurements were taken in and out of the system). Using the parallel inductor for this purpose is the easiest approach, and can be accomplished at the tweeter without disturbing the original crossover. Higher order filters and different capacitor values could be used as well, but would entail more work. Fortunately, you were not trying to retain the original tonal balance of the 3a, and subjectively like your results. All is good.

This is a good book to own if you are really interested in the details:

"Loudspeaker Design Cookbook" by Vance Dickason.

http://www.amazon.com/Loudspeaker-Design-Cookbook-Vance-Dickason/dp/1882580478/ref=sr_1_28?s=books&ie=UTF8&qid=1308113367&sr=1-28

Roy

[JohnHeubel]

Yes, you are.

The integration of a potential replacement driver's inherent mechanical and electrical characteristics with an existing crossover is not at all simple. As I said before, the replacement candidate's rated impedance alone will not tell you a thing. Different tweeters with exactly the same rated impedance will ALL have their own sonic character with the very same 6uf AR-3a capacitor. The crossover frequency and response curve is not determined by the driver's rated impedance and crossover capacitor alone.

Roy

Roy, thanks for the further comments as to the testing of the other drivers. And yes, since it is all about the sound afterall, I'm happy with my replacements. It seems though, that while I may be perceived as placing too much importance on the speaker impedence, the way I'm reading your post seems, to me, as placing too little importance on it. I know you've done far more experimenting and research into these things than I have, and if I can get my hands on the referenced book it will certainly shed more light on the topic, so I know I have a lot to learn.

But, take for example the issue of recapping a speaker. Common sense dictates one to go with the original value because the xover was designed with specific driver requirements/capabilities in mind, to include a specific driver impedence. If you change the cap value, without changing anything else, you'll be changing the crossover frequency plain and simple. Whether that's detrimental or not would have to be determined by the degree of change in value, and subsequent tests/listening.

It's only logical then (to me at least), that if you change the speaker impedence, and leave all else unchanged, you've still changed the crossover frequency. Again, the degree of change in impedence may or may not have a detrimental effect, but will certainly have an effect.

So, when I looked at the recommended HiVi's specs, and what I knew of the originals (which wasn't much), and knowing that it's been several years that the HiVi's been used, I looked to see if perhaps there was a newer driver similar to the HiVi, but ALSO in a 4ohm version. That's what led me to the Peerless, as I was trying to remove one other variable, that of the difference in impedence between the HiVi and the original. My thoughts being that perhaps, when the HiVi was selected and tested, it was the best driver at the time as a replacement, but that there could possibly be something else now. Is it exactly the same as the original? No more so than the HiVi. But, in my listening, as compared to the original tweeters (which I reinstalled last night to give them a second chance and see if I could get used to the AR sound), I ultimately went back to the Peerless tweeters, as I prefer the "air" and presence they provide to cymbal brushes, etc due to the extended frequency response.

As an old commercial used to say "try it, you'll like it."

John

p.s. Here they are with new grills:

[genek]

It's only logical then (to me at least), that if you change the speaker impedence, and leave all else unchanged, you've still changed the crossover frequency. Again, the degree of change in impedence may or may not have a detrimental effect, but will certainly have an effect.

Keeping original values and not changing the crossover frequency or slope is a good thing if you're installing a replacement driver with performance identical to the original. But since there are no replacements with performance identical to the original, some crossover mod, either frequency, slope or a combination of both, is inevitable if the goal is to keep the performance of the speaker as a whole as close to original as possible. You have to compensate for differences in driver output.

It is conceivable that someday someone will produce a replacement driver whose performance exactly reproduces that of the originals, and then you'll want as original a crossover as possible. But since there doesn't seem to be anyone even trying to make one, the odds are pretty slim.