When Gary first purchased a full DIY Acorn MkII kit I must admit to having some mixed feelings about his purchase. I had known of Gary for many years having visited his Quad ESL web site several times and from this realised he was very impressed with the performance of the Quad ESL’s. Would our product stand up to the comparison he would inevitably make?
After first building and then operating the Acorns for some time Gary has very kindly provided a review, but first, a bit of potted history for those that are not aware of Gary’s presence in the world of ESL’s.
About Gary Jacobson.
Gary is the well known and highly regarded owner of The Quad ESL website and has spent many years of rebuilding and assisting others to rebuild their beloved Quad speakers. He has carried out extensive research into all things Quad
As such he is very familiar with what makes ESL’s work and why they reproduce sound so well, making him an excellent candidate to review any ESL, or other speaker for that matter.
Gary Jacobson (B.Eng., Dip.T., Grad.Dip.Comp.Ed. M.N.C., Ph.D ) started The Quad ESL web site in 1999 as a result of utter frustration with the myths, tall tales and outright gibberish flying around the Internet at that time regarding the Original Quad ESL (1957). The site was the result of 8 months of research into patents, and the historical anecdotes of former employees. Time has shown that the site is appreciated by Quad lovers the world over. Gary first heard the Quads in a hotel room around 1979 and was mightily impressed with the accuracy of the mid range and ability to reproduce the “atmosphere” of a good recording. He was blind to its failings because the sound was so realistic. After 15 months of no girls and no booze, (almost), he was able to afford a new pair of original Quads in 1980. He picked up the trumpet when he was five years old, and put it down again, because it was too heavy. He is a sometime (poor) player of the piano, ‘cello and guitar and admires the skill of others playing instruments when it’s live or properly laid down on vinyl. He believes he could write a very good book on music theory, but couldn’t play a piano concerto to save his life. He is also credited, in some quarters, with “re-discovering” the original Quad coating, as outlined in a short non-technical paper, published circa 2000, but insists that anyone with three brain cells who had read enough of the obscure (but available) literature would have come to the same result. He has a professional background in Chemical Engineering, Education and Computer Science. An engaging conversationalist, he can chat non-deterministic data modelling, audio technicalities, music, whisky or wine – probably best approached in that order. As outlined in the following review he admits to an epiphany after extensive listening to the ER Audio Acorn, which is Quad Plus.
A New Universe
...BUT first...the old...
There was a child at the front of the choir, singing the introduction to The Little Drummer Boy. It was Christmas, more years ago than I care to remember, on a warm(ish) summer evening in Smithfield, London, and The Festival Orchestra and Choir of St Bart’s Hospital, was at practice with Professor John Lumley conducting. My niece, Cassandra, was the child. David, my brother-in-law, was singing in the bass section. Cassandra’s tiny voice was the most delicate and pretty sound in that space, until, the body of the small orchestra started to play, Rebecca Lumley (the profs grand daughter) sang the first verse with the main choir backing up, and... all was complete and right in the world. St Bartholomew’s The Less, the venue for the evening, is just a chapel, but the acoustics are so exquisite as to make a strong man weep and I don’t consider myself particularly ‘strong’ in this context.
At the first break, I looked around and I recall thinking... “ ummm, 885 years”. I turned to David, asked about the place, and he said, “Yes, about that, except the windows. Blown out during the war.” The mixing desk I was sitting behind was “nothing flash” as we say, just portable and basic. I find it very hard to concentrate on technicalities in such environments. The history, the place, tends to overwhelm me – emotions not evoked by modern skyscrapers, McDonalds, KFC, et cetera. Oh yes, and sitting, incidentally, about 550 metres from the place where William Wallace was drawn and quartered, quite another story, told less than accurately, by Mel Gibson, in some movie or other. See what I mean? Maybe?
I was recording the event for Cassandra and David mainly for familial posterity. No recording captures the atmosphere in the ways I have attempted to describe above, for example, Diana Krall’s legs under the piano, (hang on, she wasn’t there – another day, hopefully); but it does give me a reference, in part, for assessing audio equipment from time to time. I was there. I have, as I hope you can appreciate, strong memories of the occasion, and when I listen to that recording, I know what I ‘ought’ to be hearing. I think of the glorious sound, the semi-damp stone, even the smell of the place...all triggered by the reproduction of an acoustic event. A distraction, introduced by a defect (let’s not equivocate) in the reproduction chain can jar me back into my listening room, with the proverbial thud, instead of allowing me a moment or two in that tiny church with that mellifluous sound at that particular time.
Can a speaker system come close to providing the sort of performance that brings the listener close to this effect? In my extensive (I bashfully claim) experience; very rarely. I can count (yes, I can, you know), on the toes of one foot, the number of loudspeakers that can reproduce the above recording, and many others, to a level of accuracy sufficient to involve me to the extent I’ve tried to describe. The point where it begins to stimulate strong, deep recollections or emotions of a moment in time. Like attending a live concert after listening to hi-fi for some time, but, we will draw a quiet veil over the regrettable incident on the top of the Kombi van at Woodstock. Being convinced of the accuracy of any system by such correlations leads us to the possibility that other acoustic events are, likewise, being faithfully and truly reproduced in our listening rooms.
Having alleged that there are such speakers, and that your system is, all things being equal elsewhere, up to the task, then what speakers would I be talking about? Quad – Yes; Soundlabs – Yes; Rogers – Yes … and a few more, like … ER Audio Acorns (who?)… in no particular order for the moment. Some of the speakers listed can be slightly more accurate for certain types of music, not one of them is perfect, or they would all sound perfect, and perfectly the same, wouldn’t they?
Those who have been bored to tears reading my various chortlings on web sites (www.quadesl.org) and in print, in the past, will be aware of my undoubted heavy bias towards Quad ESLs, particularly the original ESL. This bias has remained for decades, simply because the original Quad ESL embodies so many of the optimizations needed to create that elusive device we dub the ‘musical’ loudspeaker. The original Quad doesn’t go terribly low, but low enough for 90% of material to which you would assign the label ‘music’ to be well communicated. It doesn’t go terribly high, but, unless you are a bat-eared professional recording engineer then you won’t notice. It has an almost fabled (and certainly fabulous) mid-range that is very hard to approach, even with our much vaunted ‘modern technologies’ and materials. It doesn’t play terribly loud, although a good reconditioned pair will make 94dB. Finally, it sounds soooo musical, balanced, integrated, smooth, so much like the real thing, that the listener is drawn away from its foibles and is oblivious of same. A good engineering and design trick, when you can pull it off. So few speaker designers and manufacturers do – pull it off – so to speak – others would argue that they are ‘pulling it off’ all the time, but in a different direction. We draw a quiet veil across the latter and move on. So, I have been listening to Quads for 38 years or so, off and on, more on than off, waiting and expecting something that is really better all around and that will put me, however briefly, back in the moment at the Chapel of St Bartholomew The Less all those years ago. Back in the moment, on top of the Kombi van at Woodstock…I promised someone never to mention that again...so I won’t.
...AND now, the New…
ER Audio Acorns, can put you in the moment, they can play low, they can play high and the mid-range is equal to, or (gasp) better than the original Quad
ESL. They are the best electrostatic loudspeakers I have heard, and (yes) in spite of my predilection for Quads, I have heard pretty much all of them and a goodly number of electrostatic headphones as well. You may be thinking this is a little hyperbolic, but it is not. I’ve had the best part of 2 years building, fiddling about, and listening, to form the opinions in this review and the pleasure of receiving a number of useful lessons from Rob Mackinlay, the designer of these and a few other excellent speakers.
Now you can stop reading and go to the ER Audio web site, or talk to Rob Mackinlay on the phone, or email and order a pair of Acorn Electrostatics. You should not bother wasting your time reading the rest of this dissertation, but rather you should be reading the Acorn construction and assembly instructions. I am that confident that anyone with two partially working ears and a brain connected thereto will appreciate the overall superiority of this speaker, and, above all, its musicality. However, if you are my kind of masochist please read on and I will give you the low down, and the high up, of my experience with this really delightful transducer. You have been warned.
Still here? OK. As I said before, I am a long time Quad lover and a small time aficionado of all things Quad. I think the line of electrostatics from Quad has been the State of the Art in hi-fi speakers, until now. What’s new? Starting in the nether regions, the bass end of the ER Audio Acorn is around 35 Hz and it plays music down to that frequency with authority, not just a sense of the sound ‘being’ at that frequency. The sense of micro-pressurisation that you get in the auditorium when the audience stops mumbling, the conductor raises his baton and the orchestra fires up is well served by this marvellous capability. The Quads hint at it, but this is much closer to the real thing. My original Quads (even stacked) cannot reproduce the depth of ‘bass notes in space’ that the Acorn does, nor can most of the other electrostatics I have heard, with the exception of the big Soundlabs units at 20 times the price. In comparison, the Quads present a lovely, accurate miniature representation of an acoustic event. Moving upwards, say between 100 and 300 Hz, the Acorn produces the most natural male voice you will hear from any loudspeaker, Rogers probably coming closest in the contest, with Quads neck and neck with Rogers. I’m being very specific here, sorry if you haven’t heard either speaker. If you haven’t heard them, then you reach for the ultimate reference, of course, and you go and listen to a male choir (if you can, boyo) or any choir, but pay attention to the male bass singers – OK? Something familiar, of which you have a good recording – and compare. A very hard, but interesting job.
The mid range is where original Quads excel. It is why they often referred to as being the most natural sounding speaker you could ever hear. True, but the Acorns will float your boat in this respect as well, being at least as equally pure and natural, but playing at scale, without difficulty, and without (of course) noticable distortion, or any other solecisms that we come to expect in other loudspeakers. I am talking Window Rattling William Tell Overture here – OK? All the way through to 35-37 kHz there is none of that shrill noise that arises from a lot of modern tweeters.
...BUT… can the Acorn play a touch 4th ?
More to the point, can
you hear the artist (especially solo guitar players) using this harmonic effect in the piece? Or, does the speaker screw up the harmonic structures through phase error, or (gasp) pitch errors, however slight? A touch 4th is
an ‘artificial harmonic’ often called for by composers and intrinsically understood (it seems) by virtuoso players of the guitar. Those are the folks who make mind-buggeringly difficult playing look ho-hum-ish.
“What? Of course I can play under water – can’t everyone?” Then, there are the ‘pinch’ harmonics, which are even harder to execute, but for the blessed few to whom it comes naturally. I am not
All of these increasingly subtle techniques are completely lost if the loudspeaker is not sensitive enough to support the correct superposition of waves on the moving diaphragm that makes the sound we hear. Electrostatics, in general, are rather good at this and it is why they are referred to as ‘pure’ or ‘uncoloured’ and ‘accurate’ by most listeners. The speaker communicates subtlety, seemingly without effort, as does the master musician. If we were to anthropomorphize the Acorn, then it would be saying, “Is that all you want me to do?” Many listeners will also, rightly, refer to this as ‘detail’ and though there is a bit more to ‘detailed sound’ than this; without it, the cause of high fidelity is lost.
The Acorn is particularly accurate and you will be easily able to visualise the solo guitarist moving his fingers to the expected positions on the frets (or not, in case of artificial harmonics, heh heh). If you so wish. If not, just become immersed in this glorious sound and the performance. Either course of action is easily possible. There are many such properties this speaker demonstrates, all of which add up to a musical performance that is very close to the original recording. I cannot smell the semi-damp stonework in the Chapel of St Bartholomew The Less, but I can visualise it (the Chapel, not the smell) by association with this realistic sonic simulacrum – wow – I don’t say that much.
Sources and Amplfiers
What do I use to push these along? An Aura turntable with an Origin Live Conqueror arm and Kiseki purple cartridge,
ZuAudio Denone 103 cartridge, or a Rega Saturn CD player. A pair of Manley Snapper (100W) mono EL34 amps. An ME50 (rebuilt with a bigger power supply and more current delivery). An ME850, ditto as the ME50. A factory
The discerning reader will see a definite leaning towards high current delivery and I recommend to the propsective user that they use an amplifier with a very capable output stage in this regard, lest you hear clipping distortion. Electrostatic speakers of any species are somewhat demanding in this respect, even though Rob has done his best to eliminate every skerrick of ‘stray capacitance’ in the panel design. I waffle, there is none.
Simply put, driving a 2 metre high capacitor is a ‘difficult load’ and in the upper frequencies, the impedance is, shall we say, ‘not 8 ohms’, but considerably lower. The ME50 copes 95% of the time at high levels on louder orchestral works or rock material and it has a 375VA power transformer! If you listen to mostly chamber music or solo singers, then a good 50 Watts may well suffice. The ME850 manages nicely and its power transformer is 2500 VA, the ME1500 is 5000VA. No, I didn’t cook the speakers or even activate the protection devices on the crossover board, but then, I don’t go ape $h@#! with the gain control, usually leaving it around 9 o’clock on the ME25 preamp. Yes, I like the older Aussie designed amps. A lot of bang for the buck and they outdo most modern designs anyway. Mr Dagostino has learned a lot from Peter Stein. OK, let’s get back to the Acorns, shall we?
How is the above performance achieved? Having seen this speaker in parts – it is a kit after all – I do not have to guess. Lightweight 3.2 and 4.2 micron diaphragms are a large part of the story as far as superposition of waveforms goes, and the quite light, but robust, bass panel material is very likely a large contributor to the terrific articulation of the speaker in the bass and mid-bass; keeping in mind that high order multiples of low frequencies will affect the higher part of the audio spectrum. The crossover is invisible in this performance, by the bye, appearing to be a 3-way design, but there is a set of filtering resistors in the mid-treble panel that makes it technically a 4-way design. No dips or bumps of an audible nature. While we’re here, the crossover is all quality components with names like Mundorf and Dale being prominent.
Also, a couple of large inductors that you will definitely not find on-the-shelf in Jaycar, or anywhere else. It is a ‘bespoke’ crossover where ‘bespoke’ is needed, to achieve performance and the highest quality ‘off the shelf’ where that meets the designers requirements. Well engineered, in other words, spending the customers money where it really counts and saving money otherwise. Gold plated resistor leads are not required, nor is silver wiring. All the electronics are conveniently on the one board: Crossover, EHT ladder, regulators, inverter and easy to assemble thanks to clear diagrams and manual.
The approach to obtaining high voltages in electrostatic speakers is, usually, a transformer providing an A.C. output well above mains
voltage to a Cockcroft - Walton ladder of capacitors and diodes yielding the bass bias voltage(s) at the top of the ladder and tapping the ladder lower down for the mid/treble bias.
In the Acorn the mid/treble bias is around 2.5 kV to 2.7 kV and the bass around
4.5 kv to 5.0 kV. Selection of these values is governed by a number of factors, the major ones being the output desired from the speaker, the sensitivity and the type of coating in use (to some degree). Rob Mackinlay has taken an unusual approach in that there is no mains transformer (as such) and the high tension is derived from a 12Vdc plug pack! The 12 Vdc is fed to an inverter/regulator circuit and the output of this small sub-system is adjustable via a multi-turn resistor. The a/c output feeds the base of the high voltage dc ladder/multiplier. So, mains a/c to 12Vdc plug pack to dc-ac inverter to dc HT ladder – Why; I hear you ask?
I can’t speak with authority on this point, but my semi-educated guess would be that the panels, being part of a kit designed for DIY assembly, in spite of having high quality components all around, are, nevertheless, subject to final assembly by human beings. We human beings are so good at stuffing things up and coating of electrostatic speaker diaphragms is one of those areas where you can stuff up. Again, Rob has gone to great lengths to make the process as foolproof as possible, but, as Dykstra put it...’fools are so ingenious’. I am one such fool and although I have done a lot of Quad panel restorations I fell foul of the Acorn panel coating process because I (perhaps subconsciously) kept treating it like a Quad diaphragm, which it ain’t! The emphasis here is squarely on the perpendicular pronoun, ‘I’.
There is no particular difficulty in the coating process, I just made it difficult for myself. Fortunately, Rob has apparently infinite patience, but, if I were approaching this project again, I would buy the panels pre-built. The gentle reader being, almost certainly, more DIY competent than I am, can probably approach the job as a full DIY build quite easily.
Now, back to that interesting high voltage bias supply design. It does have something to do with all that ramble about coatings! No matter how well we apply the coating, we are not machines, with the accuracy and precision of machines. So, none of us will produce a coating on the diaphragm that is totally uniform across the surface, nor, indeed, uniform from panel to panel. We can get close if we are careful, but uniformity to a micron, or so, is not happening any time soon on my bench. Hence, every diaphragm will have a little bit different resistance to every other and will charge a little differently, especially when newly produced. This is where the variable output from the inverter comes into play. It is a very neat way to adjust the sensitivity of the speaker in general, but also allows for slight variations in bias that may be needed if the resistance of the coating varies slightly from panel to panel and speaker to speaker, (which it will). Clever.
Note from Rob.
Because this is a DIY project (well, most of the time) we don’t exactly know what the build variations or tolerances are going to be. Providing some adjustment to the bias voltages was considered essential to enable the builder to optimise his construction. The bits and pieces are glued together and the conductive coating is applied by hand, if the builder follows the instructions the tolerances will be within the design tolerance – great!! But what happens if they are not? To address this, we made the high voltage supply adjustable to compensate for thick or thin glue lines between components, heavy or light coating application. This makes it easy to optimise the performance of the completed speaker.
Panel to Panel
Connections to various panels are clearly marked on the main circuit board and wiring them up is simplicity itself with the coloured wire supplied. The only thing I did not like (a very personal thing I admit) was the ‘chunky’ look of the plugs used to connect the top and bottom panels to each other. I replaced these with individual 2mm bullet plugs used in radio controlled drone building. This left a bit of exposed metal plug carrying up to 6kV to insulate, so I solved that minor issue with some silicone tubing. This is easy to slide on and very effective insulation at 45kV / mm. It also allows a connection not much larger in diameter than the wire and enable the wire to be hidden in the frame just a bit more easily while not electrifying same (!) I used a metal frame, so extra care was needed, but even a wooden frame will conduct a little at these voltages.
Note from Rob.
Gary’s call on this was far better than mine. We have discarded the plug / socket arrangement and moved over to some slender male / female plugs with a silicone sleeve over them. Nowhere near as bulky and more reliable. Thanks Gary!
As this is a DIY kit, to keep costs within reason, you must build your own ‘enclosure’. Rob has a couple of perfectly
practical and relatively inexpensive suggestions, with diagrams, in the construction manual. However, masochist that I am, I had to try something a little more adventurous (and of course, too
I thought a separate box for the power supply/crossover component of the speaker would ‘look cool’, and I set about making something that appeared to have been extracted from Dr Who’s control room. This, unfortunately was not considered in keeping with the décor of the listening room by the ‘indoors authorities’ and was eventually discarded in favour of building the more classical housing at the base of the speaker.
Not to be utterly thwarted, I mounted a couple of LED meters which are monitoring the 12V line and the regulated line from the inverter. The latter can be quite practical as it saves having to attach a meter to the test points on the board while adjusting the regulator to between the 7 and 7.5V Rob suggests. It might be overkill, but it is surprising how many times I have had to adjust that voltage during the run-in period. I also used a bit of red acrylic to mount these on the front of the frame and expose the gubbins to public view in a muted fashion. A retro, round dial voltmeter completes the cosmetics – which is all it is intended to be; connected to a dropping resistor on the 12V line which powers the red LEDs that provide the muted internal lighting. All of which is nothing to do with the sound, but thought I would mention it anyway.
Now, let’s talk framing details. This is still a loudspeaker ‘enclosure’. The builder cannot, if they know what is good for them, simply mount the panels in a frame of some kind and let rip. Well, they can, but this will lead to sound which in no way resembles anything I described at the start of this review. In fact, bare mounted panels will yield PPS – Piss Poor Sound – especially in the bass spectrum. It is also unsafe leaving a few high voltage bits and pieces uncovered. Don’t do it! Forget about ‘see through’ panels. Looks cool, sounds shit.
After building the frame, you should fit a grille cloth to the front of the panels. It will work perfectly well, just stretched enough to stay in place and in contact with the stators at the front of the speaker. It will provide some protection against the ‘finger poken in das machinen’ phenomenon that is known to occur in domestic installations from time to time and a tiny amount of front stator damping.
The frame I built (and use) is carbon-fibre-vinyl-wrapped aluminium, in the main, which looks very nice to my aesthetic sense and will revolt those who prefer the warmth of hardwood. To each his own. Vinyl wrap is easier to change over to a different ‘look’ than re-paint any day. The entire frame with panels mounted is roughly 32mm deep, except for the base where the electronics live. This was a sort of ‘design aim’ of mine and meant that the frame needed a brace at the rear. This is something Rob suggests and is indeed very beneficial to the sound. I used a 25mm circular section length of stainless steel boat railing with appropriate pivot mounts top and bottom to stiffen the entire structure. The bottom 30-40% of the tube is filled with number 7 lead shot. No ringing happening there!
My ‘design aim’ of 32mm thick means, as you have probably guessed, more pain
and suffering. Why didn’t I just use the 50mm PVC pipe and alumnium frame...
why,why, why(?) … because I am contrary, that’s why. So, when it came to adjusting the ‘Q’ of the speaker my work was cut out for me. Even luminaries like Peter Baxandall have long ago commented on speaker ‘Q’ as something best adjusted on a ‘cut and try’ basis rather than raw calculation. Plenty of theory has been developed, of course, and he was writing back in the 1960’s when the calculations would have been impractically restricted by the time available to perform them and surely there should be software for doing this in the 21st century, but, I could find no such thing, at least not for electrostatic speakers. So, ‘cut and try’ it was then. You may think ‘trial and error’ but ‘cut and try’ is a more literal description, since what is being described is the process of placing acoustic resistance in the speaker enclosure, or indeed, in the panel itself to prevent untoward diaphragm travel. Think ‘damping’ and you pretty much have the idea for an electrostatic speaker. The situation is a good deal more complex when we start to involve boxes which tend to have a mind and resonance of their own.
How hard can that be? How long can it take? Pretty hard to do properly, and about 18 months (off and on). No, I’m really not that stupid, it just takes a good while to add a material, listen closely to various recorded material, add or remove, et cetera. It is an iterative (and irritating) process.
The ‘Q’ of a Quad ESL, for instance, is around 2, meaning that it is under-damped with a rising bass characteristic, in an effort to moderately compensate for the lack of output in the lower registers. Those who own the Quad ‘63 and later incarnations may wonder where this damping material is, and the answer to that is that it is inside the panels in the form of a polyester screen-like material, glued to the inside of the rear stator. I tried this approach with the Acorn, using several different grades of ‘silk’ screening material (which is polyester these days, not silk) and generally thought it was a bit over-damped. The ‘ideal’ thread density was not available for the ‘ideal’ acoustic results and I don’t have the many thousands of dollars needed to produce a run of something suitable, that I might have to throw out anyhow. I did end up using polyester, ‘silk screen’ fabric though. It was placed about 1 cm behind the rear stators, trapping about a centimetre of air behind the panels. Not quite enough damping, so I resorted to the old standby of ‘mover’s felt’. That’s the greyish looking felt that house movers wrap around your furniture to prevent scratches that they always end up getting on the woodwork anyway – you know the stuff?
After many, many weeks of to and fro, try, replace, add try again the (more or less) final combination was the aforementioned polyester screen (200M) with one layer of 3mm felt behind the bass panel and two layers of felt behind the mid/treble panel. The felt is affixed to the screening material with lightweight double-sided tape that you can obtain at Officeworks. The polyester material was attached to ribs in the aluminium frame with ‘pinch weld’, which is the tough rubber material mounted on a spring metal backbone that car manufacturers use to cover weld seams. The ‘sandwich’ at the rear was completed with a 32M layer (very open and coarse as silk screen material goes), also held by the pinch weld strip. This gives just a little bit less than critical ‘Q’ (about 0.7), which is, to my ears, about right for this speaker. No need to artificially boost the already prodigious bass end on this one! It also looks presentable with no wires showing and no finger openings to be explored!
Was a very quick rundown indeed, of 18 months of adventures with this exceptional speaker. It is the equivalent of speakers costing $30,000 plus in the retail space and although I know Rob Mackinlay is not prepared to have his door metaphorically knocked down by a stampede of speaker buyers, I would recommend trying to join the queue before he does something like retire from the field. You will not be alone in your adventure, anymore than me. Rob is understanding to a fault and provides support that is far beyond ‘the extra mile’. Now I will withdraw to my listening room, draw the curtains, and, as the sun fades early from the winter sky await the appearance of anything from a string quartet to a full symphony orchestra or a rock band running full blast at one end of the room. How they got in here I don’t know but it may be to do with those two tall, thin things I have installed.
One of Gary’s Acorn MkII Speakers
E R Audio Pty Ltd. ACN: 120 797 775, ABN 48 120 797 775
© Copyright E R Audio 2002-2018 Legal Website designed by RNM