Ar3a Midrange issues...

[Mexicomike]

Hi all,

I recently picked up a set of 3a's and am in the process of getting them into shape but I have some issues with the Mids. ( I sold my original pair of AR3a's back in the 70's when I was young and stupid!)

The speakers have alnico magnet, cloth surround woofers which appear original and are in excellent shape. The tweeters are also original. The mids are different. One appears original and seems to have an alnico magnet (very deep unit) while the other has a ring-type magnet. The ring magnet unit appears to be an AR - it has the "cage" and some foam under it but it looks different than the other one - the cage material on the alnico one goes much further out toward the screw holes than does the ring magnet one. Both are "outside" (front) wired. It appears to me that one was replaced, perhaps by AR at some point.

All the crossover components appear original. The only mod that I can find is that the mid/tweeter wires were removed from terminal 3 of the pots and soldered to the "full" position. The pots, of course, were corroded and didn't work at all when I replaced the wires to the original locations.

I cleaned the pots and they are working fine BUT the two midranges sound completely different. The older one seems harsh and missing the lower end of the midrange compared to the newer one. They sound different with the pots in or out of the circuit.

So...I'm not sure of a couple of things...How would I test the crossover systems to ensure that they are both operating in spec? I suppose the difference could simply be the x-over, right? Also, If it is an actual speaker problem, I see that many companies supply replacements though none of them look at all like the originals. Does anyone have a good view of what a current midrange sounds like as opposed to the original?

I realize I can "upgrade" these speakers with new mids/tweets and Xovers, etc. But I don't really want to do that.

So I'd like some input on the best way to approach this. I want them to be AR3a's, not "formerly AR3A's" if at all possible but I realize this may NOT be possible. If I could find an original AR mid of the newer type (with cage) that would be fine I guess, assuming the crossover isn't causing the problem in the first place.

Thought are most welcome!

[johnieo]

Hi Mike:

From the sound of it, no pun intended, one of your midranges is an AR-3 driver (your description of physically deep driver fits that, but a photo would confirm). Perhaps a prior owner blew one 3a driver and installed the incorrect replacement? The lower crossover frequency for the AR-3 is 1 kHz; the 3a is 575 Hz. Operating the AR-3 midrange that low is one explanation for its harsh sound.

One economically viable option would be to consider swapping your 3 for a 3a with one of the CSP folk who restore AR speakers. Alternatively buy one on e-Bay and then sell the AR-3 driver afterward.

Cheers,

[Mexicomike]

Here's a pic of the two AR3a speakers. The one on the right is the one with the alnico tweeter. The one on the left sounds a lot better. THe speakers serial numbers are within 5 of each other; both woofers are stamped Sep 23, 1968 on the rear of the magnets.

oops - pic didn't work...

[Mexicomike]

Speaker pic is now attached (I hope).

The dome diameter of the two mids appears the same to me but the cage material extends further out on the Alnico version. Also the foam between the dome and cage is soft on the ring-magnet mid and very hard on the alnico mid. I talked to ABstech and asked them if their mid is sonically and mechanically interchangeable with the original and they told me that they absolutely were and that the cage on the original ARs were for "cosmetic purposes" only. Since the original ARs were never designed to be seen (and the cage is ugly anyway, this seems to be an ABstech marketing statement, not an accurate one. From what I have read, the screen/foam was placed there precisely for sonic reasons. Am I wrong in my understanding?

oops - didn't work either.

How about this:

[Guest matty g]

Hi Mike -

I'm pretty sure that what you have there is an AR3 midrange on the right and an AR3A on the left. They do fit but they don't the sound the same (as you know). John's suggestion is an excellent one, and I'll bet that one of the good folks here would be willing to make just such a swap.

Concerning the comments made by ABtech - I'm not the expert here, but I do know that AR never did anything with their drivers in those days for "cosmetic reasons". They had a very good reason for EVERYTHING that was done. Cosmetics would never be a factor, as they didn't even want the owner/user removing the grills (as is painfully evident whenever we need to service them). To the best of my knowledge, the screen is there to hold the foam over the driver which is purely for sonic reasons. These guys were highly respected engineers, and they knew what they were doing. That's why these systems are still so highly regarded after all this time. The ABtech replacement tweeters don't sound the same, from what I've read here, so I would question the midrange replacements. I believe a crossover tweak may correct that (I think that is still in the works) but then they aren't the same speaker. If you can find a 3a tweeter, that would be the best solution.

Hope this helps.

Matt

[johnieo]

Mike:

Thank you for posting the photos. It looks to this observer that both midranges are indeed AR-3a midranges; the one on the left - later, the one on the right - earlier. The AR-3 mid looked different.

Attached below are two photos. One is of an early AR-3 -- photo source forgotten; the second is a composite that Roy C gave me some time ago showing four versions of the AR-3a-style midrange from early to that used in the AR-11. It would appear that your two midranges are the first and second itterations of the AR-3a midrange driver and not the AR-3 driver. Second opinions, anyone??

Two other possiblities for the harsh sound:

- the older midrange is damaged.

- the midange capacitor has increased in value so much that it has reduced the lower mid crossover frequency and affected the output of the driver.

Your cabinets show the old five-pound paper box capacitor containing the woofer and midrange cap (three wires: woofer - common - midrange) It is my experience that these units no longer meet specifications after 38-39 years. I would replace the caps in both cabinets. 150 uF and 50 uF for the woofer and lower-mid crossover, and 6 uF for the tweeter. Most folks use non-polarized electrolytics (npe) for the woofer and mid. A mylar or polypropylene capacitor is often used for the tweeter, although that choice is subjective; many restorers use npe for all three capacitors.

Lastly, could you provide the serial numbers on your cabinets? Are the woofer coils (largest of the three coils) in the crossover labeled in felt pen with the number "7" or number "9" ? (The midrange coils should bear the numbers "4" and "1.") Becareful of the midrange 0.51-Ohm resistor. In early cabinets it was a single piece of Nichrome wire that was sometimes misidentified; later cabinets a fixed resistor. When tracing the circuit please follow the AR factory crossover diagram, not the Layne Audio circuit. Thanks.

Now, let's see if the images load!

Cheers,

2593.txt

[Mexicomike]

The serial numbers on the cabinet papers are: "AR3a 15170" and "AR3a 15175"

The number 7 is on both woofer coils; the nichrome wire resistor is present on both speakers; the medium sized coil has a 4 on both speakers, the smallest coil has no number on either speaker

Both crossover's are now wired in accordance with the AR factory diagram. As I mentioned in my first post, the mid and tweeter wires (green/black) were wired to bypass the pots. That was the only non-factory change but they are now back to factory configuration.

Thanks for the excellent info and the pics of the various mids!!!! I'm really looking forward to getting these speakers "humming" again!

[johnieo]

Hi Mike:

Tried to post this two times, but did not work. Hope this is not a third copy!!

-------

Please consider replacing all the capacitors, then re-stuff and listen again. Personally, I would not mess with the midrange drivers until you replace the aging caps. From the date/serial# it would appear that the later midrange was a replacement unit.

The Alnico-magnet woofer was used in the AR-3a up until 38,2xx to 38,5xx or so (Jan/Feb 1970), then AR switched to the newer ceramic-magnet woofer. your woofer dates and serial numbers fall right on the line of the data I have collected.

AR changed two other component values as well. The woofer inductor value was increased from #7 (1.88 mH, 0.58 Ohms) to #9 (2.85 mH, 0.80 Ohms). We do not know the date of this change; Roy C and I have changed the coil used with our early woofers to the new value and prefer that value. AR stated that the purpose of the inductor increase was do remove a +2 dB "bump" in the woofer's upper range response - again we do not know if that statement pertained to the Alnico or ceramic woofer.

AR also reduced the stuffing weight of yellow fiberglass from 28-30 oz. to 18 oz. This may have been done when the ceramic woofer was introduced. Did you save the fiberglass removed from each cabinet? If so please weigh it for your records.

Let us know how they sound!

Cheers,

[mluong303]

Hi John

The midrange on the right is AR-3 2" hard dome midrange which is orange brown color if one look closely through the cloth donut ring. Also pay attention to the much wider diameter of the metal mesh compare to the 1.5" black soft dome AR-3a midrange even with the earliest front wired version of Roy's picture. The earliest AR-3 midrange didn't have the round fiber glass in front of it as your picture shown.

These speakers look like the earliest AR-3a with big block of 50uF + 150UF combo wax paper covered capacitor. Somewhere along the line an AR-3 midrange was installed which will be wrong substitute for this speaker to sound balance from one to the other!

"....Your cabinets show the old five-pound paper box capacitor containing the woofer and midrange cap (three wires: woofer - common - midrange) It is my experience that these units no longer meet specifications after 38-39 years. I would replace the caps in both cabinets. 150 uF and 50 uF for the woofer and lower-mid crossover, and 6 uF for the tweeter...."

Sir, your statement will hold true to me and to the people who is planning to restore the AR-3a speakers for themselves. But for the sellers who are planning to re sell their AR-3a speakers to get maximum Money from over sea buyers. They better of not touching those wax paper type big block capacitors at all. The 6uF also use the same type in the earlier production. The AR collectors swear this type of capacitors sound the best with tube amp?!

I am just telling what I see and what I know. Whatever one decided to do for personal satisfaction or enjoyment is up to his own choice.

Minh Luong

http://www.classicspeakerpages.net/dc/user_files/2599.jpg

[johnieo]

Minh:

Thank you for picking up my mistake! I was correct the first time (Mike described the mid as having a deep magnet, which the 3a does not). However, I did not look carefully enough at the metal mesh -- it certainly does extend close to the screw holes and is an AR-3 midrange! I was fooled by its light colored center. Like the 3a, the 3 must have had more than one version.

So back to square one for Mike - replace the midrange and leave the capacitors alone for now. Minh is correct Mike, collectors want original capacitors even if they have drifted 50% in value!

My apology and thanks again!

Cheers,

[Carlspeak]

Thanks John for summing up all this valuable information (which I've seen in numerous separate messages) into one post.

It's all about the music

Carl

Carl's Custom Loudspeakers

[Mexicomike]

So...does anyone have an original (working) AR3a Mid that they would sell me or partial trade for the AR3 Mid? Or can anyone point me to someone/somewhere I could find them?

I will, of course, keep checking ebay though there doesn't appear to be any there at the moment. Ebay is where I got these particular units - which were "Guaranteed original." I looked at the pics and asked questions but never noticed the different mids. Oh well, the cabinets are in excellent shape!

[johnieo]

Mike:

For one matching original early AR-3a, try Larry LaGace "vintage-ar"

ljlagace@aol.com

I checked with one member, but he does not have an individual driver that is identical to your unit. Anyone else?

Cheers,

[RoyC]

Mike, I have some extra front-wired ceramic magnet 3a mids (circa 1970 vintage). Any one of them should be a good match. I've attached a photo of one of them (they are all the same). Send me an email through the forum, and we should be able to work something out.

John, Larry was actually looking for a couple of mids a few weeks ago as well:-). They appear to be getting more difficult to find.

Roy

[johnieo]

>AR also reduced the stuffing weight of yellow fiberglass from

>28-30 oz. to 18 oz.

I'm not doing too well today! Let's change that 18 oz. to 20 oz.

[johnieo]

Mike:

Here is the AR-3 midrange photo I was first seeking; found it in an old post from Tom Tyson. This is the easy way to distinguish between the AR-3 and -3a midrange. The -3a mid is 1-3/8" deep from front of faceplate to back of magnet. The -3 driver is at least twice as deep. Once the profile views are compared there is no mistaking one for the other!

Cheers,

[Mexicomike]

Roy,

I sent you an email...Thanks!

[Mexicomike]

Thanks John, your pic is exactly the same mid that is in my questionable 3a.

[Mexicomike]

I reinstalled all the stuffing and the woofers in my 3a's and fired 'em up! Despite having the mismatched mids, I managed to get them to sound fairly balanced by adjusting the now-perfectly-working pots appropriately. My wife even commented on how good they sounded!

I don't hear any obvious problems with the crossover components. Clearly there is a difference between the two speakers so I will be looking forward to replacing the AR3 mid with the proper one for a 3a. (Roy did you get my email?) At the moment they are set up in a small room of the house we are renting in Mexico City. We are still moving in and there are boxes everywhere. I thought I'd provide a pic of this exceptional stereo system!

Yes, sadly the Ipod on the left speaker is providing the signal at the moment! Despite their best efforts, even a tube preamp (Conrad Johnson) and a Moscode 600 power amp can't change the Ipod into an Oracle!

Next thing is putting the new grill cloth onto the frames. Once that's done I think I'll try adhering the grills to the speakers with velcro so they are easily removable. But I may not like it if it makes the grills stick out too far due to the thickness of the velcro. We'll see...

One quick addition - I ran some test tones and measured with a radio shack db meter at about a foot from each speaker. There is a large difference in the 1000hz output at the same Pot position - around 10 db. I figure that's due to the improper mid. In the higher ranges, there was NO db difference between the speakers at 10,000hz. At the low end there was only a 2db difference between the speakers at 35 HZ. That's pretty impressive for 41 year old speakers!

[RoyC]

Nice speakers, Mike...well worth the restoration effort.

I replied to your email yesterday morning, and just sent another short email to see if we're connecting. Hmmm, maybe that beefed up Mexican/US border is the culprit :-)!

Roy

[Mexicomike]

Roy,

I just checked my old email address - I got one of the emails and have sent you one from a better email address for me. I don't know why I used that one for my profile here - wasn't thinking I guess and now can't figure out how to change my email address in my profile!

[Guest matty g]

Mike -

Very nice setup. If those are tape boxes in the foreground, then the next box I'd unpack is the open reel tape deck (then put the i-pod in the empty box). When I moved in here, I think I had the stereo set up before the bed or the kitchen!

Matt

[Mexicomike]

Yep, they are open reel tape boxes. Some of them are 10" tapes which I no longer have a machine to play. However, I also have a fair number of 7" tapes and an Ampex F44 which I bought in 1965. It still looks almost new and works perfectly. I had new old stock Ampex heads installed and aligned about 5 years ago. It's an amazing old machine and parts are still available - or at least they were a few years ago when I replaced all the rubber parts with NOS parts in 2003.

For years when I bought records, I took them home, played them the first time and recorded them on that Ampex. From that point on I played the tape instead of the record except on rare occasions.

[Guest matty g]

Those Ampex machines are darn near bullet-proof. I had an AG-350 that took a tremendous amount of abuse and never gave up. Finaly sold/traded for a newer Teac 6300. They are now my favorite tape machines. With a great playback system (built around AR3a's) there's just nothing that sounds as good as a reel to reel machine (provided it's used at 7.5 IPS or higher). Just my opinion, but I'd rather listen to open reel or properly mastered vinyl than most digital sources - especially with accoustic suspension loudspeakers of the 3a's caliber. I still record vinyl on to reels so as not to wear the records out....sometimes burn 'em onto a cd, for the car.

Hope your midrange issues work out to your liking - 3a's are worth the trouble!

Matt

[Guest matty g]

Those Ampex machines are darn near bullet-proof. I had an AG-350 that took a tremendous amount of abuse and never gave up. Finaly sold/traded for a newer Teac 6300. They are now my favorite tape machines. With a great playback system (built around AR3a's) there's just nothing that sounds as good as a reel to reel machine (provided it's used at 7.5 IPS or higher). Just my opinion, but I'd rather listen to open reel or properly mastered vinyl than most digital sources - especially with accoustic suspension loudspeakers of the 3a's caliber. I still record vinyl on to reels so as not to wear the records out....sometimes burn 'em onto a cd, for the car.

Hope your midrange issues work out to your liking - 3a's are worth the trouble!

Matt

PS - please disregard the multiple posting of this last message - apparently the computer doesn't agree with my comment about digital audio sources!