Almost as if the AR1ms’ unexpected appearance on eBay sparked a global upheaval, the local market suddenly came alive shortly after the New York transaction, and within a three week window two different AR sets breezed through the used items scene. The first was a lovely pair of AR48’s, advertised by a guy living a mere 2km from my work in Stellenbosch! Cosmetically they were both only slightly marked, but one was missing its mid-range driver, and on the other mid-range the surround (paper suspension) was in a final state of crumbling decay. The seller, Diederick, also an AR lover, bought them from the original owner, the late old Mister Frowein, a well-known jeweler of Stellenbosch. Diederick could not recall exactly when he took ownership of the set, but we know that the 48's were manufactured from 1981 to 1983, so if we assume that the set hit South African shores at round about that time or just thereafter, Diederick probably bought them in the late Eighties or early Nineties. Be it as it may, after removing one of the mid-ranges to have it repaired, it unfortunately went missing in the process, so the set ended up in storage in his garage. These drivers were manufactured with a rubbery kind of suspension which did not last very long, so it was the first and most common repair on all 48's. Okay, I was facing one or two challenges, but still, at $50 I was very happy with the purchase.
I ordered a set of surrounds for $19 ($9 for shipping) from Vintage AR in Boston, and a new replacement driver from AB Tech in Hopedale Massachusetts to bring the 48’s back up to 1984 specs. Only when the driver arrived did I realize the mistake; the unit was not new-in-box old stock, but a close-to-original-spec substitute. That did not fit in with my restoration plans, so the ABT unit was shelved, and I got the proper old driver from Klotz Audio in Hereford Arizona for $20 (and another $28 for shipping). They were taking apart a set of AR48’s, so I counted myself extremely lucky to get hold of that one.
Replacing the surrounds initially sounded like a task beyond my capabilities, so I started looking for South African speaker repairers, but from the very, very short shortlist I compiled, I could not verify anyone as capable to lovingly restore my precious drivers. No doubt there are guys out there who have successfully worked on these vintage items, but I finally settled on giving it a try myself. A kind of come-hell-or-high-water situation, because I had to obtain the know-how, seeing that the AR48’s most probably were not going to be the last set of speakers coming through my door requiring restoration. I downloaded a video clip from the Internet, and closely followed Vintage AR's excellent and most comprehensive documentation which accompanied the replacement surrounds. Ok, so the first operation resulted in glue all over the driver frame, but I was quite proud of my workmanship on the second one. But more importantly, when I finally hooked up the speakers, the sound coming from both mid-ranges sounded just excellent.
Despite this early stage of restoration, was this the perfect sound I heard in 1975? Probably not, but for my singed little memory lobes it came extremely close - it was as good as or better than anything else I have ever heard. I was happy and I was satisfied that I have after three decades finally started closing the loop. Contend? Ha! I have reached my original objective, but as I have by now came to accept, the ideological replacement of the AR3a’s was simply the symbolic opening of Pandora’s Box.
More work though was required, because the one woofer was emitting a worrying scratch when pumping the extremely low frequencies. The only possible reason I could attribute to this was that the voice coil was not properly aligned. Did it go out of true because of the ravages of time, or was it a bad restoration job on the surrounds? From what I could make out, the surrounds appeared original, but then it may also been a very neat re-surround job. The cure for this was to carefully loosen the surround and realign it, being careful not to damage anything. It required a blunt knife, a steady hand and lots of patience. I was hoping that by loosening only the one half of the surround the release in tension would make it re-align by itself, but that was not to be, so I had to spend the better part of an hour carefully prying it loose from the frame. Even when it was completely free, the voice coil was still rubbing, so by now I started believing that the speakers indeed had the suspensions replaced, and that the voice coil on this one got damaged in the process. This of course was rectified by forcibly aligning the voice coil using a set of shims, and at the same time gluing the surrounds back down with the proper cement. Voila! Scratch gone.A second problem was that on starting up the first few times, the speakers had to “warm” up before any sound could be heard. This could have been an indication that the cross-overs were shot, which was to be expected after all these years, but seeing that it eventually got better, I did not attend to their replacement immediately. At this point I must also admit to a somewhat embarrassing discovery – I have been playing the 48’s for a day or two before I discovered that the high tones were only coming from one speaker.
The tweeter in the other one was completely dead, and I could only pray that it was because of the condition of the crossover and not the driver itself. Unfortunately my prayer was not answered, as tests would finally reveal that the tweeter was out of order. A handwritten sticker attached to the back of the magnet indicated that the unit underwent repairs in 1992 for voice coil and wiring repairs, and looking at the remnants of the repair work I was fairly convinced that the old problem was either not properly dealt with or that it was repeating itself. The handwork and soldering required here was way too fine for my finger stumps and senior citizen eyes, so I had to take this to the pro's.
Initial feedback from the “pro’s” prompted me to just forget the repair and find a replacement, which I duly did in the form of a nice working pair for $15 out of Wellington, Florida, USA. I popped one into the gaping hole, plugged in the cables and pushed through some sound. The first few seconds was brilliant, and then it just died! Closer examination revealed another busted voice coil! Ok, so we had a leak somewhere, and removing all the drivers showed that the woofer had no seal. This either went missing in a previous life and I never noticed when I worked on it, or yours truly lost it himself. We're not talking about being airtight, which would be a bit more complicated to achieve due to the very nature of both building materials and construction, but simply an acoustic seal good enough to give us the required acoustic cushion. Common double-sided foam tape solved that issue, and now with a proper seal, everything ran very well. At the back the old connecting posts had started disintegrating, so it was the ideal opportunity to also replace those with proper RCA binding posts.
I am very impressed with your blog. And I can see the speakers are back in good hands again! Well done.
Regards,
Diederick
(2) Acoustic Research AR3a
This set is the one from Sybe Bakker's collection, having been in storage for approximately ten years up until early October 2009, and ultimately replaced the AR48's. I vaguely remembered the pair of AR3a's I saw in the Cape Town shop thirty-four years earlier, but so much of the finer detail was escaping me, that this in a sense was my first really close-up examination of the model. Even unrestored, the wood looked very impressive, but then again I'm easily drawn to any old plank which are standing up against the ravages of time. I could hardly wait to put a hand to them and bring out the deep and rich lustre of the American walnut. Only the front part of the frame was solid wood, and the rest of the panels the normal laminated MDF combination of the time; nonetheless these speakers were extremely heavy. Picking up one felt like double the weight of the AR48 which was about the same dimensions, and certainly also heavier than the much bigger AR94's. Specifications reveal a hefty 24kg, compared to the AR48's 17kg and the AR94's 19½kg.
The grills had no physical damage, but as could be expected they were showing the normal signs of discolouring associated with the ageing of the original cream material. I'll therefore try a wash before discarding them for something new, but for the sake of budgetary constraints, the replacements will have to be a locally sourced version of the suggested style of Irish linen, rather than the replacement sets available out of the US. More on this later.
The cabinets only had a few nicks and scratches and one or two loose splinters, so the overall cosmetic condition was very good. That made wood restoration relatively straightforward. The rough patches and deeper scratches were "smoothed out" with some wood filler and the deft use of a mouse sander, and all sides were then lightly sanded down. The classic oiled finished treatment took this phase to its conclusion. The backs of these speakers were closed up with plywood, so there was no crumbling pieces to be locked in like some of the other AR's I worked on, but I'll nevertheless still put down a coat of sealer as soon as possible. Both cabinets took just about half a working day to finish. The picture below shows the one speaker restored (left) and the other still requiring work.
With the grills off and following the Champagne/Kantor/Luong/O'Hanlon guide, I was able to identify the period of manufacture as between 1967 and 1969, making this one of the earlier AR3'a sets. My first visual inspection revealed no signs of damage as I couldn't detect any deterioration of the surrounds, and a manual pressure test for cabinet sealing also checked out fine. It was time to take the next step - a proper audio test with power fed to the drivers. Ever put off something because you were fearing bad results? I wanted to, but the urgency to relive the past was just so much greater, so I hooked up the Onkyo amp, hoping that it's claimed 60 watts will indeed be the 60 required to drive the notoriously hungry AR3a's. Starting with the recommended low-level test, my heart dropped like a bag of wet cement, knees going lame with that sickly sensation of disappointment. On both the woofers were reassuringly flexing in and out, but as for the rest, on the one speaker the mid-range was out, and on the other the tweeter made no sound whatsoever. I listened, and listened again, fiddling with the volume and tone settings to make certain. What was dead, remained dead.
I had to swallow my disappointment; after all, I couldn't really expect any piece of audio equipment that old to come out of the wars unscathed. Taking out the multimeter however brought the first signs of good news - both conductivity and resistance measured okay, so electrically the suspect drivers were all functional. Crossovers shot?
As I was about to start reading up on AR3a crossover rebuilds, I remembered the potentiometer issue. Both tweeter and mid range had old rheostat controls on the back panel with which tonal quality could be adjusted, so I started fiddling around with the one attached to the dead tweeter. At first nothing happened, but then a faint crackle sneaked out. Soon I found a spot where I had full electron flow! Repeating the process with the second pot brought similar results. Okay, so I had six working drivers, but how well were they performing? The pots most definitely required surgery, and hopefully a careful cleaning process would do the job in restoring full control. Disappointingly, and it just about broke my heart then, but at that stage the AR3a's were found wanting. The one speaker was way down on dB's, and on the whole the sound was muddled and without life - I had the AR94's doing duty in the hi-fi setup at that stage, and they were light years ahead.
But no despair yet. It simply meant long nights and more applied AR science for a full blown AR3a restoration.
September 2010
If you have ever dreaded opening a piece of very valuable equipment for repairs or restoration, fearing the unknown, or perhaps more correctly fearing the possibility that you will not be able to properly repair said item, then you will know why it took me eleven months to finally gather enough courage to tackle the AR3a's The fact that in the meantime the AR48's were doing a mighty fine job of delivering pure unadulterated sound, naturally helped to prolong this state of affairs. During this period I collected and read through many papers on the AR3a, the most authoritative one being the Champagne/Kantor/Luong/O'Hanlon guide from The Classic Speaker Pages; read it many many times, trying to work out how I was going to meet and overcome setbacks, but as it turned out, good fortune was never far off.
The tweeter in the other one was completely dead, and I could only pray that it was because of the condition of the crossover and not the driver itself. Unfortunately my prayer was not answered, as tests would finally reveal that the tweeter was out of order. A handwritten sticker attached to the back of the magnet indicated that the unit underwent repairs in 1992 for voice coil and wiring repairs, and looking at the remnants of the repair work I was fairly convinced that the old problem was either not properly dealt with or that it was repeating itself. The handwork and soldering required here was way too fine for my finger stumps and senior citizen eyes, so I had to take this to the pro's.
Initial feedback from the “pro’s” prompted me to just forget the repair and find a replacement, which I duly did in the form of a nice working pair for $15 out of Wellington, Florida, USA. I popped one into the gaping hole, plugged in the cables and pushed through some sound. The first few seconds was brilliant, and then it just died! Closer examination revealed another busted voice coil! Ok, so we had a leak somewhere, and removing all the drivers showed that the woofer had no seal. This either went missing in a previous life and I never noticed when I worked on it, or yours truly lost it himself. We're not talking about being airtight, which would be a bit more complicated to achieve due to the very nature of both building materials and construction, but simply an acoustic seal good enough to give us the required acoustic cushion. Common double-sided foam tape solved that issue, and now with a proper seal, everything ran very well. At the back the old connecting posts had started disintegrating, so it was the ideal opportunity to also replace those with proper RCA binding posts.
It was time to start using the 48’s as full time hi-fi units, and during the summer of 2009 I split the hi-fi section from the home theatre section (no more sharing of components). With the AR amp supplying the juice, I for the first time was able to really push and evaluate the 48's. Gone was all earlier suggestions of aged cross-overs, and true to the traditional clear and no-BS sound of AR, they now took a very definite step onto the same level of excellence as the AR94's. As for my ears, the biggest difference was in what I shall call the greater musicality of the AR48's - I suspect the main contributor here was the AR amplifier, even though I have always considered the AR94 to display a smoother and rounder sound than the 48. The latter I now in tonality placed firmly in the AR3 and 3a genre, perfect for music and a worthy middle-of-the-road AR speaker for the early Eighties.
FEEDBACK
I am very impressed with your blog. And I can see the speakers are back in good hands again! Well done.
Regards,
Diederick
(2) Acoustic Research AR3a
This set is the one from Sybe Bakker's collection, having been in storage for approximately ten years up until early October 2009, and ultimately replaced the AR48's. I vaguely remembered the pair of AR3a's I saw in the Cape Town shop thirty-four years earlier, but so much of the finer detail was escaping me, that this in a sense was my first really close-up examination of the model. Even unrestored, the wood looked very impressive, but then again I'm easily drawn to any old plank which are standing up against the ravages of time. I could hardly wait to put a hand to them and bring out the deep and rich lustre of the American walnut. Only the front part of the frame was solid wood, and the rest of the panels the normal laminated MDF combination of the time; nonetheless these speakers were extremely heavy. Picking up one felt like double the weight of the AR48 which was about the same dimensions, and certainly also heavier than the much bigger AR94's. Specifications reveal a hefty 24kg, compared to the AR48's 17kg and the AR94's 19½kg.
The cabinets only had a few nicks and scratches and one or two loose splinters, so the overall cosmetic condition was very good. That made wood restoration relatively straightforward. The rough patches and deeper scratches were "smoothed out" with some wood filler and the deft use of a mouse sander, and all sides were then lightly sanded down. The classic oiled finished treatment took this phase to its conclusion. The backs of these speakers were closed up with plywood, so there was no crumbling pieces to be locked in like some of the other AR's I worked on, but I'll nevertheless still put down a coat of sealer as soon as possible. Both cabinets took just about half a working day to finish. The picture below shows the one speaker restored (left) and the other still requiring work.
With the grills off and following the Champagne/Kantor/Luong/O'Hanlon guide, I was able to identify the period of manufacture as between 1967 and 1969, making this one of the earlier AR3'a sets. My first visual inspection revealed no signs of damage as I couldn't detect any deterioration of the surrounds, and a manual pressure test for cabinet sealing also checked out fine. It was time to take the next step - a proper audio test with power fed to the drivers. Ever put off something because you were fearing bad results? I wanted to, but the urgency to relive the past was just so much greater, so I hooked up the Onkyo amp, hoping that it's claimed 60 watts will indeed be the 60 required to drive the notoriously hungry AR3a's. Starting with the recommended low-level test, my heart dropped like a bag of wet cement, knees going lame with that sickly sensation of disappointment. On both the woofers were reassuringly flexing in and out, but as for the rest, on the one speaker the mid-range was out, and on the other the tweeter made no sound whatsoever. I listened, and listened again, fiddling with the volume and tone settings to make certain. What was dead, remained dead.
I had to swallow my disappointment; after all, I couldn't really expect any piece of audio equipment that old to come out of the wars unscathed. Taking out the multimeter however brought the first signs of good news - both conductivity and resistance measured okay, so electrically the suspect drivers were all functional. Crossovers shot?
As I was about to start reading up on AR3a crossover rebuilds, I remembered the potentiometer issue. Both tweeter and mid range had old rheostat controls on the back panel with which tonal quality could be adjusted, so I started fiddling around with the one attached to the dead tweeter. At first nothing happened, but then a faint crackle sneaked out. Soon I found a spot where I had full electron flow! Repeating the process with the second pot brought similar results. Okay, so I had six working drivers, but how well were they performing? The pots most definitely required surgery, and hopefully a careful cleaning process would do the job in restoring full control. Disappointingly, and it just about broke my heart then, but at that stage the AR3a's were found wanting. The one speaker was way down on dB's, and on the whole the sound was muddled and without life - I had the AR94's doing duty in the hi-fi setup at that stage, and they were light years ahead.
But no despair yet. It simply meant long nights and more applied AR science for a full blown AR3a restoration.
September 2010
If you have ever dreaded opening a piece of very valuable equipment for repairs or restoration, fearing the unknown, or perhaps more correctly fearing the possibility that you will not be able to properly repair said item, then you will know why it took me eleven months to finally gather enough courage to tackle the AR3a's The fact that in the meantime the AR48's were doing a mighty fine job of delivering pure unadulterated sound, naturally helped to prolong this state of affairs. During this period I collected and read through many papers on the AR3a, the most authoritative one being the Champagne/Kantor/Luong/O'Hanlon guide from The Classic Speaker Pages; read it many many times, trying to work out how I was going to meet and overcome setbacks, but as it turned out, good fortune was never far off.
First stop after opening up the speakers was the potentiometers. Each pot was showing a fair amount of corrosion, even on the outside, but that was to be expected after so many years. I cleaned up as best I could with a fine brush on a Dremel tool, but half the contacts were so bad that I could not establish complete contact over the full range. Nonetheless, once firmly settled in a position, performance increased tenfold over initial switch-on almost a year ago, and I decided to leave it at that. Phase 2 will see a revisit to the pots for home-made replacement contacts, and quite possibly the replacement of capacitors, while Phase 1 was concluded by refurbishing the speaker cloth and badges.
Classic AR speakers came with grills covered in 18-count (threads per inch) Irish linen, which of course is no longer being manufactured, so the closest match I could find was a 16-count local version known as hessian and produced from, I believe, sisal plants. Coloured in a very mild off-white, visual appearance was absolutely spot-on as far as matching the AR feel was concerned. Going for about $5 per running meter, I left the shop with whatever was left on the roll (enough I found out later to cover the grills of 40 speakers!). The AR badges were re glued to their fitting plates (a steel plate with a screw soldered to the middle), and after a mild scrub and polishing they were good to go.
So, sound wise, where did Phase 1 take me? In the December 1970 issue of Hi-Fi News one Ralph West reviewed both AR3a and AR4x, and came to the conclusion that the 3a was only marginally better than the 4x, after initially having placed the smaller speaker ahead of the AR3a! Ralph obviously was suffering from severe inner and outer ear disturbances, whereas for me it was fairly simple and easy to after a week declare that the AR3a was way ahead of any AR speaker in my possession, and that included the AR4x. It was just stunningly different in any aspect of tonal comparison, and for the first time ever I had to turn down the amplifier's bass control by 50%. For the first time in my life I came "face-to-face" with the awesome bass capacity of this groundbreaking speaker, a proud legacy to the principles of sound advocated by Edgar Villchur. Now, sitting there amongst the weird collection of both vintage and modern home theatre pieces, their presence is so commanding, that even switched off and when walking through the room, they draw one's attention like two ominous giants of past battles waiting on the heights for the next encounter.
My thirty year wait had come to end.
