Turner B302 & B502 Power Amplifiers
The two amplifiers reviewed here have a long pedigree. When first introduced in 1975, they were way ahead of their time, both in respect of being tightly packed and for their above-average sound quality. And apart from being repackaged in 1977, they've remained much the same since.
The B302 and 502 are twins in almost every respect: ignoring minor internal differences, only the physical size differs, to reflect the higher power rating of the B502. (Table 1 gives the power for both amps into various impedances.) As you can see, the B502 is nearly double the power of the 302, and this fact makes a combination of the 502 and 302 a likely choice for some Bi- and Tri-amped speaker systems, with the B502(s) on mid and/or bass, and a B302 on the top. From the perspective of ordinary, full-range speakers, the 302 will suit small (and micro) monitors, but in most instances, the larger B502 will best realise the power handling capabilities of 8 ohm speakers.
These amplifiers are unusually lightweight, and small. Most pro amps of similar power occupy 30 to 100% more space. To a great extent, compactness has been achieved by allowing the most bulky parts, namely the (fully-shrouded) transformer and reservoir caps to stick out of the rear of the chassis plate, in a fashion rather reminiscent of 60's American amplifiers, like the Crown DC300, say. Sadly, the Turners' volumetric efficiency is largely accounted for by an absent of depth: the (vertical) rack space occupied is average, at 3U (5¼") for the 302, and 4U (7") for the 502. Using these amps outside of racks, though, the shallow profile (both models are under 8' deep, including leads) befits them to many a cramped studio, where the massive depth required for some low-profile amps simply isn't feasible.
The transformer is a conventional, laminate type, but owing to the unique layout a toroidal type wouldn't appreciably reduce the space occupied. In fact, the only niggle here relates to the B502 we tested, which exhibited a loud (acoustic) hum. As the B302's transformer was silent, it's doubtless a rogue, but watch out for this one if it's likely to prove irksome. It all depends on how quiet you want to keep your control room.
Beginning with the facilities on the rear panel for a change, inputs are via standard 'A' gauge jacks. However, Michael Stevens & Partners (UK sole distributors) offer a version suffixed XLR (ie B502-XLR) with male and female Cannon connectors on the front panel. The males offer linking-out facilities for patching, and both XLRs are paralleled with the standard input jack, which is retained.
The output terminals are binding posts. No problems here, except that when rackmounted, these are so deeply recessed (in a 14" deep rack, say) that it becomes next to impossible to clamp down bare wires with the amp in position; you'll need to make these connections before sliding the amp into its rack. On the XLR version, however, the outputs are readily accessible from the front panel - there's a single male XLR for each channel. The inputs are unbalanced, and all the XLRs are wired conventionally, with pin 2 hot, and pin 1 to ground. Mains power is via a detachable IEC lead.
Turning now to the front panel, the on/off rocker switch is on its side, so it's not immediately obvious which side should be depressed to turn on the amp, and in bright sunlight, the neon won't be particularly visible.
Below, there's a neon indicator to tell you the amp has overheated. This indicator relates back to a pair of thermal cut-outs, one on each channel. When activated by excess temperature, either of these will disconnect the mains to the (shared) power supply. By using a neon, capable of working directly across the mains, the circuitry is kept simple (cf. Japanese amplifiers), and the neon's low-ish intensity if compensated for by the pearl-coloured lens, which contrasts well with the neon's natural orange glow.
The gain control knobs are well above average size, and this makes it easy to repeat specific settings with fair accuracy. For example, it's easy to visualise five or even ten incremental positions between each (printed) number. Of course, a 1dB stepped attenuator would offer some advantages, but the technique here, using a large knob, is plainly superior to less refined click-stop attenuators, say those with 2 and 3dB steps, always assuming the legend doesn't wear off. It also keeps the price down.
Sticking with the gain controls, the pots behind the panel are particularly poor types, from AB, and practice proves these to be unduly vulnerable to expiring if any heavy abuse is meted out. This shouldn't (but may well) arise with the amp permanently racked-up in a control room, and in a road context, fitting a mechanically-rugged pot (e.g. Bourns 81 series) is a definite must. A common fault, with AB pots, incidentally, is for the track and backing plate to part company with the shaft. On a positive note, the pots are open, but on the underside only, it's fair to point out that better quality pots won't need cleaning, and are fully-sealed against nasty substances.
All mains connections are sleeved, and even colour-coded; are you making notes on all this in Japan? Along the bottom, there's independent (+) and (-) supply rail fusing for each channel. The fuseholders are open types, with above-average, but nonetheless still skimpy spring contacts. All four are tightly packed together, which makes it difficult to remove the fuses until the amp is switched off and (note) fully discharged - say after 2 minutes have elapsed. DC rail fuses are unwelcome in that if one fuse drops out or dies spontaneously DC can appear across the speakers. With these particular amplifiers, only a failure in the +ve rail fuse will cause the full rail voltage (50 to 65V) to appear at the output; if the -ve fuse goes down, the output will only swing about 13V positive, which isn't too harmful, except to a micro-monitor perhaps. Lightly soldering the +ve rail fuses in place then seems a sensible safeguard, and as with other Bi-polar amplifiers, some form of DC protection is advisable (See HSR, March 84, p. 54).
The majority of the amplifier circuitry is on a pair of plug-in cards, one per channel. These are firmly retained by a quick release plastic clip. It is hard to imagine an arrangement which is quicker and easier to undo, whilst improving on the near 100% impossibility of the cards becoming detached by accident. The one small snag is that unless you just want to inspect the card, you'll have to unsolder the input lead, which is wired direct. On the other hand, it does prevent casual theft!
The circuitry is related to several top UK Hi-Fi amplifier designs from the mid 70s. At this time, a handful of designers made spectacular advances in high-power amplifier technology, using triples (the output current is built up gradually, over three interconnected current-amplifying stages), lots of current sources, matched input devices, 'fast' output transistors and convoluted compensation techniques. These latter two made it possible to achieve decent top-end performance from Bi-polar amplifiers for the first time, but there are trade-offs in ruggedness and stability, so all this comes at a price. The upshot of all this is that the Turner circuitry, whilst dated, can still outperform many of the more recent Bipolar designs, most notably in the region of top-end sound quality. It's all rather like a Volkswagen Golf, which also dates from the mid 70s, and still outclasses most of the competition vis-a-vis subtle refinements.
Turning now to the remaining components, mounted on the chassis, every single screw is well locked with varnish, not by accident, but by dint of experience; designer Mike Turner has obviously seen what happens to some rock 'n' roll amplifiers... This attention to detail extends to the reservoir caps on the B502, which have screw terminals. But, unlike some, these ones won't come undone in a hurry. In fact, the only screws without varnish are those securing the lid and base-plates; some shakeproof washers are definitely called for here.
Wiring is tightly loomed, military-style, and it's hard to believe this can contribute to low distortion and stability. From an exterior viewpoint, and ignoring the pots, the amp is only vulnerable insofar as being dropped onto a concrete floor could dent or dislodge the transformer and capacitors - rather than smashing off the binding posts, as is wont to happen with other amps! But given the protection of a rack enclosure, the space saved by having external components overrides this. Moreover, these parts benefit from a free airflow which keeps down temperatures and thereby boosts longevity.
In common with certain other UK amplifiers, Turner's are particularly easy to service. After unscrewing the side bolts, the front panel can be swung aside. Then, with the cards unplugged, the entire guts of the amplifier are accessible; there's no complex dismantling, and only a handful of screws need be removed. Michael Stevens and Partners can supply kits of spare output devices and cards to professional users, and by keeping a kit at hand, downtime can be as little as a couple of hours if anything goes drastically wrong. One benefit here is that you don't have to tie up funds in spare amplifiers - or race out and hire a spare at great cost - in order to keep a busy studio double-covered.
As to routine maintenance, it'd be a good idea to give the pots a clean, and also to unplug the cards and clean the edge connectors about once a year. These onerous tasks should only take 15 minutes or so; contrary to some designs, you only need remove four screws to gain access.
The B302/502's heatsinking is unusual in that it avoids a macho display of fins, and instead, utilises the chunky aluminium chassis members as a literal sink. This does, however, mean that these amps will run at a higher temperature than some, when driven hard, and are less adept at handling gross overload or continuous tones. In the control room environment that is unlikely to be a nuisance although in a rack, some fan cooling could be advisable, especially when driving bass-end exclusively. Putting all this into perspective, these amplifiers lack the massive thermal capacity of some American amplifiers for the sake of more prudent cost and weight; but there are plenty of much worse designs in this respect.
Table 2 shows the power output into various impedances, and the corresponding input level just before clipping. As you can see, the sensitivity of both amplifiers is below average. This isn't necessarily wrong, although it may be inconvenient in PA work, where a variety of amps may be linked in tandem. It'd be handy if, as a standard, all amplifiers went into clipping just above 775mV (0dBu), impedance regardless, but this is simply impossible, at least without the over-the-top extravagance of having fully-regulated supply rails. And as we've seen, sensitivities, even when they're nominally aimed at 0dBu, vary considerably. So the main observation is that you'll perhaps need to wind up the monitor gain on your desk by 5 to 10dB over the usual, for any given level. If you envisage any difficulties, you can always ask the dealer tb increase the sensitivity of the amp, by 6dB, say; this is a standard modification, and shouldn't present any difficulties.
Regarding impedances, both Turner amplifiers are rated down to 3 ohms, below which the protection circuits will cut in. As you can see, it's possible to squeeze some useful power out of the amplifier at 2.5 ohms, but this is not recommended. In fact, speakers with nominal impedances below 6 ohms are best avoided, as the protection circuitry is quite aggressive, and any spurious operation will doubtless prove rather irritating... (remember that speaker impedances can dip to around half the nominal value at some frequencies).
Both the strengths - and the weaknesses - of these amplifiers lie at high frequencies. Their renowned treble sound apparently stems from such alleged subtleties as a small phase shift at 20kHz, and a corresponding -3dB roll-off as high as 110kHz, which means less than 0.1 dB level is lost at 20kHz.
Perhaps more important though, open-loop gain is high, and with careful compensation, the top-end is improved through being less distorted than most. The problems arise when we come to clip the output. To begin with, clipping occurs 1 to 1.5dB earlier at high audio frequencies (above 6kHz). On full range material, there should be no problems, because even the peak HF content will tend to lie below the bass and midband material. So to have clipping above 6kHz would be to tempt 100% overload at lower frequencies. This premature top-end clipping can present problems, though, if these amplifiers are used exclusively on top-end material, as in an active set-up. Going a fraction of a dB beyond clipping at 6kHz - or higher frequencies, there's wideband oscillation, accompanied by a rapid rise in heatsink temperature. This sort of behaviour can prove lethal to tweeters (see p.58, May HSR), and unless there's adequate cooling, prolonged HF clipping may also terminate the amplifier.
The way around this is to calibrate your metering (for top-end amplification purposes only) to hit the 0dB point about 1.5dB below the sensitivities given in Table 2. It should be added that you're highly unlikely to drive the top-end of the monitoring system into clipping in a studio; this is really a problem for PA users. Along these lines, it's also sobering to note that the Turners' otherwise excellent HF roll-off (-0.1 dB @ 20kHz) is an open door for any ultrasonic/RF garbage seeking trouble, and the lack of balanced inputs aggravates this. To be fair, RF burnouts are again, mainly a PA problem, but do take care with cable lengths, and perhaps use paired input cable, with a conductive screen - it's all good practice with any pro-audio set-up.
A surfet of RF nasties are suggested if the heatsinks warm up excessively when system levels are wound up, but without any signal present. If this happens, take advice - or use a oscilloscope to identify the source of RF. Aside from actual damage, RF/ultrasonics will create intermodulation products, which comes across as unpleasant and audible distortion in the treble.
At a cost of around £1/watt (into 4 ohms), the B302 is above average value; the B502 betters this at 85p/watt, and whilst of course the overall cost is higher, the 502 does offer realistic headroom for driving 8 ohm monitors without fear of clipping.
The Turners' plus points are simplicity, light weight, small size (provided you're not bothered about vertical space) and elegant design. These amplifiers have been established nearly a decade, and you can be sure that they've passed most of the exams set by the rock 'n' roll business.
In contrast to the multinationals, designer Mike Turner is never complacent - in fact he's renowned (perhaps uniquely) for actually taking note of what reviews say. So doubtless the points we've raised will be looked into con brio.
Recognising that building any product involves compromises that might not align with personal tastes, Michael Stevens & Partners (who are Turner's sole distributors and also know the practical ins and outs of audio installation) are keen to meet any special requirements you may have. For example, they'll happily fit more rugged pots, or even stepped attenuators on request. B302s and 502s with metering and EQ are also available. And this customisation facility isn't the only advantage of going to a small UK manufacturer; there's no planned obsolescence in Turner products, and just as with Quad, your investment will remain serviceable for many years (cruise missiles allowing).
Back in 1976, the Turner brochure read (on the B302) "it combines technology, art and usefulness to a standard... in an age when it is most required". In 1984 these amplifiers may not be the latest in technology, but when other qualities come in, they score well above the average.
|4 ohms||150W (140W)||340W (300W)|
|8 ohms||100W (95W)||190W (180W)|
|16 ohms||60W||105W (100W)|
|Input sensitivity (at clipping)||Power output (at clipping)||Load Impedance|
Review by Ben Duncan
Previous article in this issue:
Next article in this issue:
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!