Thursday, September 19, 2013

Ls3/5a is meant to be used with its grille

There are many modern myths around Ls3/5a:

1. Speakers placement is critical

2. Ls3/5a needs to remove its grille when listening...

This article suggest that The LS3/5a is meant to be used with its grille in place...instead of removing...

The BBC-licensed LS3/5A help make Rogers into a global brand. Its current Far East owners have worked with the UK designers an manufacturers to create this BBC-licensed 11 Ohm 'revival'.


Rogers LS3/5 Speaker  We won't labour the LS3/5a story, which began in the BBC in early 1970s, as this was covered in depth in Vol1 No3 (May-June 2007). The fit and finish of this official £1499/pair Rogers revival was excellent. It certainly looks the part with its fine matched rosewood veneer, though the repro Tygan grille material might be a little softer and more open than the original. As expected, these speakers use a 5-inch (138mm) polymer cone bass/mid drive unit partnered by a nominally 19mm dome tweeter, mounted in a sealed-box enclosure. Build detail includes thick felt absorption around the tweeter, to ameliorate reflection effects from the integral grille frame. (The LS3/5a is meant to be used with its grille in place.)

This new example has a glued on back, as were most LS3/5as. The bass/mid driver has a 110mm pressed steel chassis with a flared, talc-filled polypropylene cone, powered by a double-wound voice-coil (the windings series connected), and a large ferrite magnet. A fine 19mm tweeter with recessed soft dome and sophisticated rear loading is fitted, while that unnecessary perforated metal cover completes the original look. Built of birch plywood according to the original recipe, the panels are well damped by traditional bitumen loaded pads, and the interior is lined with thick, absorbent polyurethane foam. This model is bi-wired (unlike the original) with average quality gold plate brass binding posts and brass link pins (preferably replaced by copper wire on installation). The crossover is built on the usual epoxy glass reinforced PCB, securely bolted to the panel area around the tweeter. It uses normal film capacitors, ferrite core inductors, and an array of resistors forming a calibrated attenuator to achieve the essential close tolerance matching of mid and treble levels. (The original 15ohm model used mu-metal cored transformer inductors and ratio matching of driver sensitivity levels.)


Sound Quality
With references assembled, and memory not too dimmed from the pleasant experience of that other, Derek Hughes guided 11 Ohm LS3/5a for Stirling Broadcast (Vol1 No3), I set to. Location as ever, was free space with little toe-in, and copper links fitted to the terminals.

From the very start that genuine LS3/5a magic was there, though inevitably at modest loudness levels. Shut your eyes and coloration is really low, and the soundstage is spacious, deep and focused. The natural timbre and the elegant, surprisingly deep image perspective reminds one of much larger and more costly designs. If a listener doesn't demand higher volume levels or need cracking dynamics, it might not be necessary to go further than this in the quest for genuine high fidelity.

In my opinion this is the best of the modern LS3/5a clones I've heard to date. It has quite good coloration in the low midrange -- a weakness of many -- though it is a little 'boxier' than the best 15 Ohm examples. It is also a little 'sweet', not quite as crisp and open as is possible, though one would not want it as bright as some later original Rogers 15 Ohm examples. This new contender is self-effacing, clean, even, smooth and well balanced, with a particularly seamless transition from midrange to treble, albeit a tad dynamically reticent in the piano's middle register. Spacious, transparent and detailed, if a touch 'darker' (and therefore not quite a dynamic, expressive, transparent or 'powerful', this new version of that old trouper just sounds bigger than the original, with a touch more realism.

This new Rogers certainly has a good percentage of the better drive and timing of the earliest models. We should remember that middle period 15 Ohm models often sounded hard and forward in the midband, only just meeting the BBC spec. This new version actually bests a huge number of the less good 15 Ohm models, never mind the more boring 11ohm examples and most of the 3/5 clones (which were never proper 3/5as anyway). In reality this Rogers is quite close to the standard, comparing well with the equally well mannered and similarly neutral Stirling Broadcast version. I think that it may be just a little livelier musically than the Stirling Broadcast, but reckon you would have to do an A/B to find out.


The spec says 83dB/W/m sensitivity, some 1.5dB higher than the original 15ohm model, and truthful. Frequency responses were very tidy, and with superb pair matching. The on-axis frequency response gives  70Hz to 29kHz +/-3dB, with -6dB points at 60Hz and 30kHz. Partnering stands should not be too high because the response dips sharply by 10dB at 10 degrees below the axis in the crossover zone, around 3.3kHz, while remaining perfect at 10 degrees above. The speaker should therefore be sited at or a little below seated ear level. The off-axis lateral results were good too, but that noted dip was beginning to show itself in the 45 and 60 degree off-axis results. The small tweeter keeps the output going nicely up to 11kHz off-axis, so it should stay open and clear. The bass is understandably limited with some characteristic 'plumpness' 80 to 150Hz, though this is part of the 'voice' of this little box. The in-room averaged response (not shown) is very good: it's typical for the size, with good overall power integration and with in-room bass down to about 50Hz.

The impedance graph shows an average value of about 7 Ohms. It's easy to drive, with a minimum  of 5ohms and moderate variation. The box bass resonance is seen at 91Hz. The waterfall response is quite tidy, well integrated in the early response, with only moderate decay clutter thereafter. Enclosure coloration is low thanks to the good construction and small acoustic footprint.


This LS3/5a contender can be recommended. It's beautifully made with great care and fine finish, and while I still await the equal of my ancient Robin Marshall made BBC originals perhaps it simply cannot be done, as some of the materials used to make that speaker are no longer available.


Tel: +44  01628 820134
Price: £1499/pair

---------------another nice article------------

BBC LS3/5a loudspeaker

These diminutive little sleepers have been available in the US for quite some time but have attracted little attention because (1) they have never really been promoted and (2) they are just too small to look as if they could be worth $430 a pair.

Originally designed by the BBC for monitoring of on-location broadcast and recording pickups (footnote 1), they hide most of their cost—a complex equalizer and phase-corrected crossover network—inside a cabinet only slightly larger than a shoe box. They were intended for "close-in" listening in a small control room rather than to fill a large room, and they will definitely notput the kind of levels beloved of rock nuts without woofer-bottoming or ultimately permanent damage.

Despite what must be a rather large amount of built-in bass boost (to compensate for the small size of the woofer), they are fairly efficient: We would estimate around 1¾%, which is comparable to an average acoustic-suspension system. Maximum safe output level is around 95dB SPL (sound pressure level) at a listening distance of up to 15', which is about as loud as a symphonic crescendo from 10' behind the conductor. This is with full-range program material; the limiting factor on power input is the "woofer" (because of its bass boost), so when the speakers are used with a subwoofer (crossing at 60 to 80Hz), they are capable of a clean 100 to 105dB, which is enough to give any masochist a most gratifying case of permanent ear damage.

Judging by their size, one's first thought is likely to be that these will work just dandy up on the wall, right below the ceiling and toward the room corners, where standing-wave resonances in the room will help augment the speakers' thin bass. But their size is very deceptive. These arenot thin-sounding. In fact, they produce an overall balance similar to that of a pair of large systems when they (the Rogers) are located on 30"-high stands, right out near the center of the floor. (This bass-balance design is consistent with the BBC's recent research findings which showed that the smoothest bass response is obtained when a speaker is as far as possible from room boundaries.)

It is because these speakers are so well-balanced when they are out in the room that they may well produce too much bass when placed against a wall, particularly when located near the junction between three room surfaces. In corners, they are (in most rooms) intolerably boomy because they are designed for out-of-corner placement, and because that location excites the maximum number and amplitude of standing-wave resonances (footnote 2) in the room.

The close proximity of room surfaces (or, worse yet, of a box or shelf under the speakers) also causes diffraction interference—the chopping of deep holes in the frequency response due to selective cancellation of certain frequencies. The smaller the speaker enclosure, the less audible are these diffraction effects and the smoother the system sounds. But nearby corners and surfaces can spoil the advantage of the small enclosure.

Another advantage of a small sound source is that it tends to radiate sound waves as expanding spheres rather than as a planar wave (as from large screens). Human ears react in a seemingly paradoxical manner to a spherical sound field: The reproduced sound seems, much bigger than its source, yet the angular localization of sounds across the "stage" between the speakers (ie, the imaging) is dramatically improved. In fact, the apparent audible size of these tiny speakers is almost laughable; we had the feeling that it just could not possibly be.

Adding to the illusion of a large speaker system, is the remarkable low-end performance, which is not really all that deep (subjectively flat to a bit below 57Hz in our rooms) but sounds deeper than it is because the response is actually pretty flat down to there (rather than drooping), and the bass detail is astonishing from 5" woofers. The speakers gave such a startling account of themselves at the low end that we were not acutely aware of the lack of deep bottom until deeper notes (as from bass drum or the bottom range of the string bass) that we knew were on the recording failed to come through.

High-end performance is quite remark able. The speakers have a very slightly rising response above about 5kHz (fig.1), but because there is no audible peak at the top, the rise does not cause any sizzling or spitting, but tends rather to exaggerate slightly the extreme high-end energy in the program, adding a bit more sibilance to voices, a bit more shimmer to cymbals, and a bit more overall airiness to the sound than is actually in the program material (footnote 3).

We were not able to determine how far out the high end goes, because our oscillator's upper limit is 20kHz, but the smoothness and the ability to reproduce hard transient information (triangles, castanets, etc.) suggested that 27kHz might not be a bad guess for the upper rolloff point. Overall, the monitors have an extraordinary feeling of depth and breadth to their sound. Inner details were excellently rendered, and there were no audible vowel-like colorations that we could detect. The only thing we could the Monitors on (other than the low-end range) was a subtle tendency to lighten or "stretch upwards" the vowel structure of some instrumental sounds. The effect could be likened to the difference between the lower range of an oboe and the upper range of an English horn, although the extent of the difference was less.

Qur comments about the sound of these speakers have thus far been based on its performance with what we feel (as of this writing) to be the best available power amplifier: the Audio Research D-150. (The new Audio Research equipment has just been announced, but will not have been auditioned by the time we return our Monitors.) We freely acknowledge the idiocy of driving a $430 pair of speakers with a $2700 power amplifier, but we can report that these Rogers/BBC speakers did just as well (if not slightly better in one one respect: a less "hot" high end) when driven by a Dual 76A. It is both a tribute to, and a liability of, these speakers that the lesser their driving equipment, the worse they sound. They reproduce the high-end roughness of solid-state amplifiers mercilessly, and most such amplifiers tend also to dry up their bass end a bit so that they lose some of their nice bottom fatness. (Placing them nearer to room boundaries then helps to compensate).

As for room placement: Except for the possible detrimental effects of standing waves, these are refreshingly uncritical of room placement. They are, as we said, best used on stands, and any carpenter can make up a pair. The speakers should be about 6' apart and 10-12' from the listening area, and each should be toed inwards, by about 5 degrees.

Optimally placed, and driven by suitably high-quality electronics and signal sources, these are a perfect way of getting sound that is comparable to that from Quad electrostatics, at far lower cost and with added bonuses of slightly smoother high end, better stereo imaging, a broader listening area, and considerably greater apparent (that is, audible) size. Their low-end output is, in most rooms, deeper and fuller than that of the Quads, but like the Quads, their major weaknesses are limited sound output and lack of extreme bottom. Above 60Hz, the Rogers BBC minimonitors outperform the vast majority of systems costing upwards of $500 per channel. And with the addition of subwoofering, via M&Ks or Janises, these could provide perfoorance comparable to that of some of the best systems com mercially available, regardless of cost.

The only real handicap that we can see to owning a pair of these is that your tineared friends won't be as impressed as they would be if your speakers took up the whole East Wall of your living room.

Footnote 3: Stereophile's subjective frequency response curves show how the transducer under test sounded to us, rather than how it measured. The vertical scale on each curve is the same, and is scaled so that a barely perceptible audible deviation from flat frequency response is reflected in a barely perceptible visual deviation of the response curve.

Fig.1 Rogers LS3/5a, "subjective" frequency response.

A Followup appeared in December 1977 (Vol.4 No.1):

Subsequent experience with these remarkable little speakers has strengthened our feeling that that they should not be used with solid-state power amplifiers. The speakers need a slight high-end softening, and are more likely to overload (the woofers bottom, making an alarmingly loudbang) from the average solid-state amp. An Audio Research Dual 51A is ideal (with a solid-state amplifier for the woofers if you choose to bi-amp), while an upgraded Dynaco Stereo 70 does very well. The Audio Research D76A is fine but has more power (and a higher price) than is necessary.

Output level is limited to around 85dB with a high-powered, wideband solidstate amp, 95dB with tubed electronics and no subwoofer. With a subwoofer and the Rogers rolled off below 70 or 100Hz, listening levels of over 100dB can be obtained without stress.

If you missed our full report on these in the last issue, a summation: Superbly balanced sound overall; very subtly nasal in some rooms; slightly rising (above 5kHz) but very smooth and extended high end; no deep bass but deficiency not noticeable on most program material; very good detail; extremely large apparent soundsource; very good stereo imaging; limited output level.—J. Gordon Holt


No comments: