Jump to content

Yet another AR-3 project


mvboland

Recommended Posts

I recently came to possess a pair of AR-3s that my father bought new way back when (serial numbers C-0977 and C-0978) and I would appreciate the group's wisdom on how best to approach them.  They are both functional with sound coming from all 3 drivers when connected to the Dynakit Mark III amps that they have been paired with for as long as I remember (amps were the last update project!).

The potentiometers are functional in that there is sound from the mid and hi drivers with them turned all the way up and no sound from them when turned down but there is some "scratchiness" in the middle.  The audio coming from them sounds muted in the mid-hi frequencies but I am not old enough to remember what they are "supposed" to sound like.

I was able to successfully remove the front grille from one of them (have not touched the other one yet) and the drivers are not obviously damaged, unlike like some I have seen online!  The cloth surround on the woofer is also intact (see attached).  The woofer cone is easily displaced but returns faster than I expected based on online comments.

I measured the frequency response of both speakers with all of the potentiometers turned up using a calibrated microphone (see attached). 

Based on all of the above, I am hoping those of you with deep knowledge of these speakers can recommend repairs. I have new capacitors for the crossovers and am prepared to clean the potentiometers. I also ordered the "special" coating for the cloth speaker surrounds.  Beyond that, I am wondering:

  1. If the mid and hi drivers might warrant a tune up by one of the experts available online given the frequency response in that range seems diminished (or just update the crossovers first)?
  2. If the bass performance is as expected - it looks anemic on the frequency response curve compared to the one published in High Fidelity:

Thanks to everyone in advance!

 

-Michael

 

label.thumb.jpg.2de92f1d1575b408254dac5afec39c5a.jpgright.thumb.jpg.809d5e728f8643145408f7373f6694ce.jpgleft.thumb.jpg.66e26e8a97842c61a3ce3742a5e044ed.jpg1110665900_CalNativeResponse.thumb.png.6b388b6ee9c5b607651d707f6b9421f7.png

Link to comment
Share on other sites

Hello Michael,

Congratulations, you've inherited a very early pair of AR-3's (1959) that look like they've been well taken care of. The cabinets look like walnut with a lacquer finish, and the grills have the attractive gold thread, with the logo and '3' pins present. That's as good as it gets for AR-3's, and they are still at home in Boston after all these years, and with the Dynaco amps!

The experts here will respond shortly, but I'll give my advice based on being involved in the refurbishment of a few pairs of AR-3's.

The good news is that all of the drivers are operating and that the pots work! The frequency response graphs you made actually look pretty good, but there is some variance left to right that could be caused by the aged driver suspensions.

If these were mine and I planned to use them frequently, I would have the 'mid's' and the tweeters' suspensions updated. Having the 'mid' rebuilt is the most bang-for-the-buck update in an AR-3, along with properly operating pots. You can send a PM to Roy C here at CSP regarding this. He does an excellent job restoring these drivers.

As for the crossovers, if the capacitors are the wax block type, you may want to replace them. I would suggest using NPE's. If the pots work, a simple cleaning may be all that's needed, along with an application of dielectric grease to the wiper / coil.

AR woofers with cloth surrounds usually hold up well and don't need to be resealed, but you do want to check and see if the surround is still firmly attached to the frame. The same goes for the spider, and you'll need to remove the woofer to check this. New glue might be needed. I've used Aleene's Tacky Glue for this purpose, but an epoxy might be a better choice.

If the woofer has a return time of at least a second when you push it in, you're seal is probably fine. If you decide to remove the woofer, add sealer to the surround after you re-install it, if it's needed. 

 

 

 

Link to comment
Share on other sites

Ooooo, very nice condition, and very cool, being so early. I'm not re-using the wax cap block in my AR-3 speakers, as I don't trust their age to be reliable, and getting to them is not especially easy; I'm putting in new ones while it's apart. I also am getting the mids serviced, as the material/compound around the paper hardens a bit with age, although yours has the 4 blob setup of the tweeters, while mine have a white compound; maybe it's not as necessary for yours.

Link to comment
Share on other sites

On 11/12/2022 at 7:49 AM, mvboland said:

 

 

-Michael

 

left.thumb.jpg.66e26e8a97842c61a3ce3742a5e044ed.jpg

Hi Michael!

Congrats your Dad has been owned and you are still keep the very early AR3 speakers, this is the first time in the NET  I see the  very early AR3 with the real consecutive serial numbers. The lacquer cabinet speaker looks very new up front, if have a time would you please take some more pictures from the side and the back of both speakers. Yup as all of the old ARs speakers rusty dirty pots problems... to me I just used deoxit d5 spray to the pots and turn backward and forth for hundred times..somehow the Mid-Hi will be sound good at one  spot. And are better keep it in all original condition, don't recap the crossover.. in case you want to sell it your speakers are worth over 3.5$K toward the Asian AR-3s collectors that most of them , they want the speakers are in original condition rather than the sounds...

I had collected few ARs speakers pairs.. using for some kind of the old music. Daily I use the JBL C50 and c51 for the modern hifi music... but although the JBL has 15 inch woofers, they sound big and louder then the AR3, 12 inch woofer but the JBL bass is not strong and tight as the AR3s bass.

I saw the very clean , mint Dynakit MarkIII in the background of your speaker pic ..I have the Dynakit ST70 hooked up with the Mcintosh C2300 tube preamp that It sounded different level comparing to the Dyna pas 3 preamp.. the Mid-Hi are sound very very more details ,clear and  clean then the Pas 3... the weakness point of the C2300 is no loudness button so It is less bass then the Pas 3 preamp.

 

315250090_5189948867777696_2754740305733874_n.jpg

Link to comment
Share on other sites

  • 3 weeks later...

@Ronan - thank you for the comments!  Most of the surfaces are in good shape:

PXL_20221129_191929916_sm.thumb.jpg.58f329e97b4d4cbf76c9ec595619ca52.jpg

But there are some minor scratches, worse on the top of one which served as a base for a small sculpture for some period of time:

PXL_20221129_191947639_sm.thumb.jpg.f3fe6afdcc8467eac334aa53db71ea03.jpgPXL_20221129_191844641_sm.thumb.jpg.2a65fee54b096b154fb24acb5ee2096d.jpg

 

@RoyC - as a highly regarded expert on these early AR speakers, any thoughts on whether the hi and mid drivers warrant service based on the frequency response curves above?

Thanks again to everyone!

-Michael

 

Link to comment
Share on other sites

9 hours ago, mvboland said:

-as a highly regarded expert on these early AR speakers, any thoughts on whether the hi and mid drivers warrant service based on the frequency response curves above?

Thanks again to everyone!

-Michael

 

Without a doubt, the mids would benefit from refurbishment. Since the AR-3 mid substantially affects high frequencies as well, and is often in need of significantly more work, I recommend dealing with both of those first. You should also make sure you can hear a difference as you turn the level controls. It is very common for the tweeters and mids to produce little to no sound at all until the controls are properly serviced or replaced. (In my experience, cleaning solutions such as Deoxit are seldom effective to adequately restore proper functionality to typically crusty AR pots.)

You have very early specimens. Your version of the AR woofer is actually more typical of the AR-1, though I'm not sure if that is the cause of your concern regarding bass performance.

Roy

PS Send me a personal message if you would like me to be of assistance. I see you are in Boston. I'm in upstate NY not too far from you.

 

Link to comment
Share on other sites

Since I have not seen some of these components (capacitors really) in other posts, I thought I would post images here from opening the cabinet:PXL_20221202_012643471_sm.thumb.jpg.dbb114e7090435b8e477c8380fceee75.jpgPXL_20221202_014515515.MP_exported_1214_sm.thumb.jpg.97267c94506dd6c785dcf53ff4c7f99e.jpgPXL_20221202_021320339_sm.thumb.jpg.740076c25f0e37bf0d88e1151e5f5610.jpgPXL_20221202_021907140_sm.thumb.jpg.7aad0572a0fa9f344ee529ea445e52dd.jpg

 

The potentiometers are actually not bad. One of them goes smoothly from ~1 ohm to almost 16 ohms and the other looks like it needs some mild cleaning (readings are glitchy over rotation):

PXL_20221202_020849651.MP_exported_1111_sm.thumb.jpg.ec8682de6c8539dd3d7b737610c7b905.jpg

Link to comment
Share on other sites

Beauties! Listen to Roy for all aspects of the refurb. I’ll just add that you have oil filled capacitors that should probably be retained. Check the capacitance with an LCR meter but they are probably good. 
Kent

Link to comment
Share on other sites

13 hours ago, mvboland said:

The potentiometers are actually not bad. One of them goes smoothly from ~1 ohm to almost 16 ohms and the other looks like it needs some mild cleaning (readings are glitchy over rotation):

What you have there are an older version of the A-P pot. The wipers are formed brass with coil springs rather than the spring steel wipers that are more prone to corrosion in later versions. If you open them up to clean, the insides will look like the ones on the left in this photo (the newer pots are the ones on the right).

image.jpeg

Link to comment
Share on other sites

8 hours ago, genek said:

What you have there are an older version of the A-P pot. The wipers are formed brass with coil springs rather than the spring steel wipers that are more prone to corrosion in later versions. If you open them up to clean, the insides will look like the ones on the left in this photo (the newer pots are the ones on the right).

Good point, Gene. Coincidentally, I just refurbished some of these for Vintage_AR. Something I found interesting is they all measured just under 14 ohms. I recall that being the case with others of this generation I measured a few years ago.

Roy

Link to comment
Share on other sites

I have, Gene. Ended up sacrificing that one because they don’t go back together well 😠

I’m away from home and can’t get pictures from my computer but if you look at Ebay item 255803762838 there are pictures of the innerds there. 
Kent

Link to comment
Share on other sites

As I am disassembling things, I was able to measure some of the component parameters and have included them below.  Question for the group at this point - I have the woofer out but the hi and mid drivers are REALLY stuck to the cabinet. I tried pushing them out from behind, tapping the mid with a rubber mallet, and prying from the front (only succeeded in marring the cabinet!).  Any tips for that step so I can send them off for refurbishment?

 

Updates from the project since last post:

The parameters for the woofer are below. This design seems unusual (not seen in photos I have found online) but I was able to find the same style at https://www.radiomuseum.org/r/acoustic_ar3.html# where they have the serial number in one image and it is of a similar vintage (mine in 900's, images from one in the 2000's):

1567856872_Rwoofer.thumb.png.81af3d62675cf710a13683eaa4455c60.png

The domes on the mid and hi drivers are hand-labeled with numbers:

PXL_20221208_155258926_sm.thumb.jpg.ab6f23ba10e43be11e88232a27ca3551.jpg

A couple of the foam mount points on the mid driver dome have come loose (one intact, one broken below):

PXL_20221208_155138357_sm.thumb.jpg.13e064ca4c907d840cef7d46f5c751e3.jpgPXL_20221208_155114221_sm.thumb.jpg.338ed2d7b7381168cbb36335961e25d7.jpg

And thanks to those who commented on the longevity of those oil-filled capacitors!  They seem to be close enough to the design values to leave in place:

1533902620_R6uF.png.9911263fbafb210ea7501344696185e8.png1960515816_R24uF.png.297f0bdd93b320e66a75dfd3d6aaa511.png

 

 

 

 

Link to comment
Share on other sites

That's pretty impressive cap performance. My wax-block caps were roughly 4 times out of spec, using a multi-meter. I used a chunk of wood, and another 12'' long, one by one board, to leverage out a mid after using a plastic scraper(carefully) to remove the sealant around the outside gap. I need to do that a few more times this weekend, to send out another mid and tweeter for service.

Link to comment
Share on other sites

Interesting evidence of some build quality issues in these early AR-3s:

- One of the screws mounting the tweeter to the enclosure was missing (so only 2 of 3 present). The screw hole was filled with the sealing putty and so probably looked OK to the builder.

- One of the woofers was likely marked and installed incorrectly and then fixed before shipping. Note the red X marking the positive terminal that has been crossed off in orange. When de-soldering the woofer, I found two sets of solder connections on each terminal, one where the bare wire portion remained but had been cut off and the other (correct) wire soldered further up the terminal.

PXL_20221210_191813320_sm.thumb.jpg.b44c31690e126328cfa7d0b3dee22fed.jpg

Link to comment
Share on other sites

  • 1 month later...

After refurbishing the mid and hi drivers, the speakers are now back together!  Thanks as always to @RoyC for his work and for collaborating on getting a displaced pole piece re-centered so these could stay all-original!

In terms of other typical repairs:

  • The potentiometers had minimal corrosion (see images above) and worked great after a little Deoxit and some dielectric grease.
  • The original capacitors still measured as good and were left in place.

In terms of how they sound, at least one of the mid-range dome glue dots had popped loose so it is not surprising that the mid-range was better after some TLC (green = pre, red = post)

59664262_Lfrequencyprepost.thumb.png.b19347807fcdb3bee3e3c8890a0e1164.png

Not knowing how these should sound, I remain disappointed with the early fall off in bass seen above.  I measured the woofer Thiele-Small parameters while it was out but have not found other measures of these very early models for comparison (sounds like really AR-1 woofers here):

1307493574_RwooferT-S.thumb.png.1aa066994d2ea3028260bb9fb6e57ee0.png

Similar issue with no clear comparison data for the paramaters of the assenbled speaker (measures essentially the same as before restoration):

611145915_Lspeakerparameters.thumb.png.87e9d05b9b5936a744f09e84f79e3f76.png

I did seal the cloth surround with the solution @RoyC devised and it made things much more airtight based on the time for the displaced woofer to return to rest (went from maybe 1 second to at least 2 seconds) but the low end response remains the same.  All testing above has been done with a single speaker at a time on the floor (so they are not cancelling each other out) and a calibrated microphone connected to a DATS v3. The speakers are currently being driven by Dynaco Mark IIIs.  I don't have any other full-range speakers to try with the Dynacos and also only have an AV receiver as another amp which I am not sure is up to the task for these low-impedance speakers.

Hopefully an update after I test the Dynacos some more to see if they are part of the problem but any other input is welcome!

PXL_20230116_153617971_sm.jpg

Link to comment
Share on other sites

I am posting separately with some lessons learned about these early AR-3s in case the summary is useful to someone else later:

  • The components used in these early models seem much better quality than what came later:
    • The potentiometers looked great after 60+ years with minimal oxidation. They were also not 16 ohms as advertised but ranged from 0 to about 15 after being cleaned up.
    • The oil-filled capacitors are still working great as well and measured favorably compared to new polypropylene
  • These were stuffed with squares of pink fiberglass which seems unusual.  I looked up pink fiberglass to make sure it was not a later addition and it was indeed available in the 1950's - Owens Corning added red eye to differentiate their product - https://www.sciencehistory.org/distillations/in-the-pink
  • The internal wiring does not match what is shown for the early serial numbers elsewhere on CSP (http://www.classicspeakerpages.net/_Media/ar-3_crossover_schematics_0.pdf).  The photo above shows what I found and I have included a schematic below with the slightly different wire colors and connections.

279789692_EarlyAR-3Schematic.thumb.png.a7fb902bc340278031101e2ccef2b4ed.png

 

  • Finally, it is great to have all of the accumulated wisdom available here - thanks to everyone who helped so far both on and off line!
Link to comment
Share on other sites

  • 2 weeks later...

>I did seal the cloth surround with the solution devised and it made things much more airtight based on the time for the displaced woofer to return to rest (went from maybe 1 second to at least 2 seconds) but the low end response remains the same.  All testing above has been done with a single speaker at a time on the floor (so they are not cancelling each other out) and a calibrated microphone connected to a DATS v3. The speakers are currently being driven by Dynaco Mark IIIs.  I don't have any other full-range speakers to try with the Dynacos and also only have an AV receiver as another amp which I am not sure is up to the task for these low-impedance speakers."

"One of the screws mounting the tweeter to the enclosure was missing (so only 2 of 3 present). The screw hole was filled with the sealing putty and so probably looked OK to the builder."

Finding a screw hole patched with putty clearly suggests that someone has worked on the speaker in the past and probably stripped-out a T-nut and just simply caulked the hole.  T-nuts can be temperamental after many years.  I am confident that the speaker never left the factory in 1959 (early serial number) that way originally, as there were too many QC checks and inspections to have allowed that to happen.  It's possible, of course, but highly unlikely.

There are several things that can affect the roll-off in deep-bass measurements, and it could involve your microphone, test setup, amplifier/preamp and so forth, but more importantly, the speakers should be measured facing into a 2-Pi environment.  Your measurement appears to resemble a 4-Pi measurement as if in an anechoic chamber, if the response begins to fall off long before resonance of around 43 Hz  In a sense the floor measurement is close but probably not ideal, but its likely that something else is at play.     

Another possible issue may be the damping of the speaker, and you mentioned that it had "squares of pink fiberglass" in it, which suggests that someone has definitely been inside the speakers in the past.  Many of the early AR-3s used a rock-wool insulation, and it was in ramdom pieces of a specific weight.  I don't remember AR using squares of fiberglass (KLH did at one point), but some may have had "squares."  The important thing, however, is the correct weight (amount) of fiberglass/rock wool in the cabinet, as this will directly affect the damping ("Q") of the acoustic-suspension system and thus the output down close to resonance.  An air leak will also affect damping, so there other issues at hand here.  The roll-off suggests more a problem with the measurement solid angle.  

The response of a properly working AR-3 (or AR-1 or AR-3a) would be very close to this (below) if measured correctly into 180 solid angle:

AR_12-inch_FR_1959_ARHPG_001.thumb.jpg.0da304eb20427ca36c11bd3e31386fdc.jpg

This is what happens when the same loudspeaker is measured when facing into full space, looking into a 360 degree solid angle:

AR-3_Woofer-Anechoic_Response(02).thumb.jpg.3b206ea82fc06432e14e892a9b3d7c95.jpg

This measurement was taken of an AR-3 measured in Harvard University's large anechoic chamber, with the speaker measured into full space (360 degrees).  This would be akin to attaching a rope to your AR-3 and lowering it into the center of your listening room so that it was away from all surfaces.  The bass response would be thin.  This anechoic-chamber measurement shows the reduced output, starting at the frequency of ultimate air-load resistance (Lambda) down to resonance, where the 12 dB/octave begins to occur.  The speaker will have flat output down to resonance when it is faces a 180 degree solid angle, as against a wall and up off the floor a foot or two, so that it looks into the proper angle.  

Tom Tyson

 

Link to comment
Share on other sites

Thanks as always to @tysontom for sharing AR wisdom.  I was led astray after going down a YouTube rabbit hole on speaker placement as modern speakers seem to be best away from walls. Instead, I should have paid more attention to item #4 on the decal on the back of the speaker:

Quote

4. One excellent location for the AR speaker is cater-cornered (touching each wall), at least 3 or 4 feet above the floor

Perhaps as expected, the bass response is much improved with the speaker in that position.   The frequency plot below is with a calibrated microphone at 1 meter (red line - still showing some room acoustic artifacts) and right in front of the woofer (orange line).  Still lacking a home anechoic chamber but both are much flatter than the ones above.

In terms of the missing screw, I do still believe that was a factory boo-boo since the sealing putty that had filled the hole was painted over (just like the screws) and the t-nut underneath worked great fine once I got a new screw.  I can see how it might have been missed as it was the screw that is closest to the top of the case, almost under the overhang where the grill mounts. These have also been in the family since they were new and have not been opened or taken for repair in my memory (40+ years). I checked with my mom and she doesn't remember any service either (her memory goes back somewhat further 😉)

Thanks again to everyone for contributing!

1953349318_RAR3Wall.thumb.jpg.15813712cf6569c3ff75e8f4517a974c.jpg

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...