Is the amplified acoustic a dead end?

Once you take your acoustic anywhere near a sound system, be it a PA or amps, it ceases to be an acoustic and becomes part of a complex chain of technical and positional variables between your fingers and your audience's ears. What you are trying to achieve is as decent a facsimile of your acoustic as may be allowed by your pickups, the volume at which you need to operate, the quality of the sound system, and the room in which it is all going to happen. You are definitely not going to get 'your original acoustic sound, simply louder,' although there may be traces of it at some levels. There is no single solution, only a range of compromises.

At the bottom of my scale is a stand microphone, which is as close as you'll get to acoustic using a sound system. In fact, in small rooms you won't get the mic much louder than the natural sound of the guitar before it feeds back anyway. Remember that speech is as loud as a strummed guitar and singing is louder. A mouth is focused, very directional, and can be placed right up against a microphone. For live, you need a directional microphone (hypercardioid or cardioid), but the closer you get to it the more you will get proximity effect — a disproportionate and ultimately crippling increase in bass response. A capsule with inherent or switchable bass roll-off is therefore essential.

Between the two extremes are points at which a fluid set of compromises will work, where it becomes necessary to trade lovely qualities of tone for dubiously secure volume in order to be heard! For example, a piezo soundboard transducer, given a nifty bit of parametric tweaking (of which more later) can sound almost acoustic, but will feed back at a low sound level. It will be adequate for strumming and supporting vocals in a small folk club and might be enough for solo fingerpicking in the same venue, but in a large bar it wouldn't be enough for anything but hard strumming behind vocals. For a bar — if you wanted to fingerpick — you'd have to go to a soundhole magnetic, but that's not going to sound close to 'acoustic', except in the sense that people might have become conditioned to accepting it.

Caveats arise immediately. Firstly, nobody listens to an acoustic guitar with their ear pressed against its body, which is very roughly equivalent to the audio quality of the signal the piezo soundboard transducer is producing. Secondly, a magnetic pickup will retain some of the tonal absorptive/reflective qualities of the guitar, as reflected by the string vibration in the magnetic field, and it is certainly going to still be sensitive to the fundamental frequencies of the soundboard and the air box cavity. So, at its upper limit, the 98Hz G is still likely to be the first note to take off unpredictably into roaring feedback. Thirdly, your natural acoustic guitar's sound comes off the guitar in various ways: from the soundhole, from the top, from the sides and from the neck. The sound is full of complex phase and directional information that makes it rich and diffused. And whatever small part of that you succeed in collecting in a pickup will be sent amplified to a speaker, which will send it out, all at once, in a restricted band and in one direction.

It is perfectly possible, and quite common practice, to use two or more of the different kinds of pickup available in combination. For example, one might use a soundhole magnetic to produce the bulk of the low mids and bass (as far as may be possible, given the local feedback threshold), and a piezo soundboard transducer to overlay more woody mids and highs, rolling off the low mids and bass on this to get away from the piezo's extra sensitivity there to feedback. In this instance, it is important to bear in mind that susceptibility to feedback is now not just related to the magnetic, but is governed by the proportion of sound coming from the piezo, and this is a lower threshold than the magnetic on its own. One must remember that the problem is not so much in the pickup's insensitivity, but in the acoustic guitar's inability to generate loud and focused sound.

I have never successfully used a clip-on or clip-in microphone on its own. The closest I ever came to useful was Velcroing an SM98 to the top of an acoustic on the outside, pointing towards the soundhole. Mounting mics inside produced slightly more volume, but only really by increasing uncomfortable cavity resonances that I needed to EQ out. The gain in volume and consistency from the stick-on/in mic's proximity was outweighed by the quality of a more optimally sized diaphragm in a stand mic. I tried a number of different stick-on and lavalier mics, and price never actually equated to guitar sound quality — one then very cheap method of using the mic unit off a cheap Realistic brand PZM barely sounded any different from much more expensive units.

An interesting double-function single unit is the L.R. Baggs M1. At first sight a magnetic soundhole pickup that collects from string vibration, it carries a second coil section that is deliberately made sensitive to body vibration. The area from which it collects body vibration is unusual (one would not naturally site a piezo soundboard transducer by the soundhole, but would put it nearer or on/under the bridge to collect a clearer string sound) and the functional effect in performance is to slide the result back down my scale, away from the level produced by a normal soundhole magnetic and much closer to the lower feedback threshold of a soundboard transducer. It can sound interesting and attractive at low levels — the sound is more complex than a magnetic alone and the attack transient is not as harsh as a piezo — but it won't actually operate well at the levels more usually associated with a soundhole magnetic.

There are tonal variations from maker to maker and, in particular, variations in susceptibility to the string attack transient. These variations almost deserve a little scale of their own, but suffice it to say that the compromises once again trade tone for volume. The most successful unit in terms of volume that I have seen in a 'band and drum kit' scenario is the Fishman Matrix on Dylan's guitar. This pickup is renowned for the harshness of its attack transient when driven hard, but kicks out the most signal of the genre. And perhaps it is the transient itself that enables it to project in a live band mix. On the other hand, in a soloist's guitar this level is not usually necessary, unless the player is going to do a lot of bar gigs. The attack can be musically disruptive and tiring to listen to, and a readily accessible example of a less harsh undersaddle might be the Highlander units installed in Martin Simpson's Sobell guitars.

It can be instructive to listen to notes played through a compressor while varying the attack response from slow to fast, allowing more and then less transient through in order to gain an appreciation of just how significant the transient is. Of course, listening to other instruments closely can yield similarly interesting perceptions — the slow envelope of a fretless bass or a brass instrument; the quicker, but similar start of a steel pan note; the thickness of a harp pluck, the unvarying twang of a harpsichord; and the fuller-and rounder-note body, but equally clicky and dynamically flat attack of a lute-harpsichord. The latter sounds like someone playing guitar using too thin a plectrum, and indeed it is physically closely related — there are just many more plectrums than we can use at once. A crucial part of how we identify an instrument comes from how the note starts, and by and large while a piezo may well give us some wood in the flow of the sound, it does the attack no musical favours at all.

Everything in this article is about compromise, and the TC DPE was the perfect compromise between function and convenience. It fitted in a string box and used very little current from a 9V battery. OK, the battery required you to undo four screws to change it, but the device didn't use much current anyway (1.3mA) and didn't need a little on/off LED (usually around 2mA draw) to burn up even more. The neatness of its construction precluded the use of a space-hogging Bulgin battery drawer or its successor boxes and hatches. It only had a 1MΩ input impedance, when, technically, coming straight off a very high-impedance piezo, we probably should have used 10MΩ. But in practice, that didn't matter, especially at a time when some players weren't bothering with preamps at all and were getting a decent enough sound. If it wasn't in regular use, its compactness also made it the perfect 'plan B' for those gigs that had an unusually troublesome room resonance or dip.

The TC DPE worked very simply. It had two bands of three-control parametric EQ, an overall plus or minus gain, and an overall treble at 10kHz. The parametric worked as usual: you set Function to 0, Bandwidth to 0.1, and turned up the overall Gain control until something started feeding back. You then turned the overall gain down and, taking the first channel, turned Function into the plus area and swept the Center until you hit the feedback. Then, you turned Function down into the minus until the feedback stopped. If the feedback still warbled either side of that as you raised overall gain, then you simply widened the bandwidth until it stopped. You could then raise the gain further until another feedback note appeared and repeat the procedure with the second channel. If you needed to push gain further than usual, there would be a bit of extra room to deepen notches, trading off yet more tonal character for the extra volume.

Fishman made a noble effort to replace the TC DPE, making a Dual Parametric DI, which was just as effective, but bigger, incorporating a balanced output and a battery drawer. It's sad to say that it disappeared quite quickly. If you want a full parametric nowadays, you'll have to get a four-band, mains-powered rack unit. There is a Carl Martin, mains-powered, three-band quasi-parametric around, but personally I have reservations about adding in another mains-powered gain block in front of the amplification. Keep an eye out for the Boss parametric EQ pedals — useful little gadgets, but they're now disappearing.

There is also the possibility that you could damp your guitar. Soundhole plugs are available or can be made, and simply disconnecting the inside of the guitar from the outside air can reduce feedback significantly. Damping the top of the guitar is getting pretty desperate, but some heavier soundhole pickups, such as the Sunrise, will have a damping effect by virtue of the inertia that they add to the instrument's top. If you find yourself regularly stuffing soft material into the cavity, consider whether or not you've bought the right instrument. If this happens only occasionally, at an unusually demanding gig, then it's just another viable 'plan B', so long as you are careful you're not damaging internal cables and connections

Another little trick uses the phase (polarity) Invert button on the Baggs and several Fishman preamps I know of. Without getting too far into nerdic lore, every note generates a sound wave. If you and your instrument are standing at the anti-node point (the widest swing of the sound wave) of a note that has a tendency to feed back and you can't simply move away from it, then inverting the signal polarity will turn that anti-node into a node (where the sound wave doesn't swing) and the feedback may be stopped. Of course, it's possible that this could cause problems with other notes by turning their anti-nodes into nodes, but it's always worth a try because it tends to be notes that are already most sensitive that cause the most problems.

The venue is the variable over which you have the least control, so it's the one you have to be equipped to cope with. This doesn't necessarily mean carting truckloads of gear for any eventuality, but means understanding the limitations of your guitar and its various pickups, and how it all operates in different circumstances. The ideal situation is a big concert hall with an open stage and a good classical acoustic — not too dry, not too wet, but just nicely live. In these circumstances, you need very little volume and might even get away with a stand microphone if the house system is spread around. Most likely, a well EQ'd soundboard transducer into a flat PA system will work fine. If you need clear, single-note lines in an acoustic band, then a softer undersaddle will be plenty. If you find you have to use a magnetic then something is wrong, because all the feedback thresholds are higher naturally and you may be playing too loud for the hall.

Where dinner gigs and background music are concerned, they don't want you loud anyway. An undersaddle and a small, flat response amp will be fine. A soundboard transducer could be tricky, because you might be set up in an awkward corner. For dinner-concert gigs, you need a bit more level if you're featured, and the audience will all turn their chairs around and pay attention. The successful venues tend to have a well-damped room acoustic so that diners can hear one conversation rather than everyone else's, so the sound you put out could well be soaked up right away. But by and large, it's still undersaddle territory without too many feedback issues.

For bar gigs in a wine bar or bigger, you need everything you've got. Ambient noise levels are usually high and room acoustics can be nightmarish and unpredictable, so you need a magnetic and EQ that will take down your own major feedback frequencies and some spare for the room. So, that could be a parametric for the issues you brought with you and a graphic for the architectural issues. Signal-to-noise ratios here are a venue issue, not an electrical one, so processing units don't need to be expensive and sophisticated — ambient noise is likely to be well over any little hisses in your or their gear.

The biggest problem with soundhole magnetics is that so much significant general harmonic content is lost, and it seems to me that if you have to compromise tone to that extent in order to be heard then perhaps you're playing the wrong instrument. However, if a new gig turns out to be more problematic than anticipated, something in the string box to get you out of trouble is a good idea. As an alternative to the Sunrise, I'd suggest looking at something like the DiMarzio Reference Acoustic, which you can pop in for a 'plan B' and then stash again afterwards. Like most soundhole magnetics, it is balanced for acoustic strings. It assumes non-ferrous string windings on third, fourth, fifth and sixth strings, and is more sensitive to detect ferrous core movement. Thus string-to-string balance with any of the genre is tricky, and reducing the level of the famously coarse second string is always a problem. You can fit nickel-wound strings and rebalance pole pieces accordingly, but at that point we would once again seem to be moving far enough away from our original acoustic guitar to raise questions about our musical purpose and choice of instrument.

There is another way to look at the whole thing, however, and that is to work downwards from the top of my scale. The first thing to wonder is how much acoustic do you actually need? If you can get a guitar that is built primarily from an electric perspective, then the issues reverse. They become questions about how one might add sufficient harmonic character for the electric to function as an unaccompanied solo instrument, or as a quasi-acoustic ensemble instrument. That can be done by looking at magnetic pickups that have more character, quite often with less feedback resistance than the really loud rock machines.

A good example is Bareknuckle Pickups' Mississippi Queen, which is ostensibly aimed at the blues market. Remember what was going on in the early days of, say, Chicago street blues, where players were using instruments and amps that were not designed specifically for high volume and distortion, but were pressing into service kit that was designed in the early days of Les Paul, Merle Travis, Charlie Christian — all essentially clean players. High fidelity rather than power handling was then the (not always achieved) priority in speaker manufacture, and pickups were simpler, sometimes more microphonic, and the harmonically uncomplicated (relative to humbuckers) P90 was around. The Mississippi Queen has the openness and simple harmonic structure of the old P90s, is shielded very well indeed, comes in a standard humbucker-size fitting, and at the lower volumes at which a fingerpicker operates, sounds wonderful — fuller and fatter than a standard Strat or Tele-style single coil, but still with the single coil's breathy top end. I put one in a 'bitsa' Strat at the neck position and was so impressed by the clarity and response detail I've been wondering about having a guitar built around it ever since. It seems to have a natural limiting that is kind to ragged nails or mis-hits, and a tonal character that is simple and flexible.

Consider the modern electric guitar. Functionally, everything about it is designed to produce an electrical signal that will be sent down a wire to electronic shaping and amplifying devices. Since the 1930s, focus has been on electromagnetic transduction. And George Beauchamp, designer of the famous 'Frying Pan' guitar, certainly believed that acoustic properties were actually undesirable in an electric instrument. In fact, we can also go back to the same time in England and a G. Woodward, Electronic Engineer, of Rickmansworth, who handmade the first magnetic pickup in Emile Grimshaw's 1934 Electrawaiian lap steel. Clearly, much more than the potting of seedlings was happening in garden sheds in the western world!

The acoustic guitar, by comparison, is a self-contained producer of volume and tone, and we hope to capture the tone part and deliver it to an electronic system that will give us the extra volume we need in a live gig. Many of us have spent a long time discovering that this is difficult and unsatisfactory at best, and embarrassingly unstable at worst, largely because the instrument's onboard physical amplifying process, the sound box, gets in the way. It seems to me, just as it seemed to George Beauchamp, that if we want to work with electronic amplification we need to move some of the tone-generating process, and certainly all of the amplifying process, out of the instrument's physical zone and into the electronic. Thus an amplified acoustic becomes, all over again, more honestly an electric.

We could pay attention to what has worked for the electric so far, modifying it to suit our specific needs, and the most significant point in the guitar's electric evolution was the abandoning of any attempt at physical amplification. In fact, quite a lot of this is already happening to the electro-acoustic, but much of it looks as tacky as a reproduction steam engine with petrol engine parts sticking out from under the faux boiler, and tonally it is frequently hobbled by attempts to avoid feedback. Even today's sophisticated digital processors, such as Fishman's Aura, do not make a plugged-in guitar sound exactly like an acoustic. Nothing does. Acoustic is one thing; plugged-in is another. Pardon what may seem like pedantry, but there is a point to it. It's a shame that the notion of a purist carries with a slightly pejorative sense, because an un-fiddled-with and well-made acoustic instrument has a character to it that is an utter joy, and there is happiness and sociability in the idea that if you want more acoustic-instrument volume you simply add more humans playing more acoustic instruments. But, alas, there is also a budget issue, and our employers, our audiences, want their music cheap or free.

Of course, many of today's audiences may never have heard an acoustic instrument in performance without some degree of electronic reinforcement. What they have heard is an acoustic instrument adapted to function as an electric, and this lacks the diffusion and harmonic richness that distinguishes a natural acoustic sound from one that is collected selectively by transducers and fed into a system that will distribute summed sounds directionally. For many of us, in fact, 'plugging in' has become the norm to such an extent that we may be close to losing our original, genuinely acoustic reference point! Do we actually need an 'acoustic' sound, or are we now simply looking for something cleaner, harmonically fuller and richer than the 'rock weapon' into which the standard electric has evolved? Having for many years personally journeyed back and forth over the technicalia and hiccups of making fingerstyle guitar louder, it seems to me that in fact we're looking at another lateral evolution of the electric, and further extension is now becoming more feasible because of the development of either factory-fitted or retrofit piezo saddles that match physically standard electric guitar saddles. For us musos, the decline in acuity of punter perception of what constitutes acousticness has turned out to be quite a convenient thing.

The most successful acoustic facsimiles I have produced have been with Graph Tech saddles into a Roland VG-88, including a recording that got a reaction of, "Which acoustic is this?" The body modelling aspect might appear to be something of a blunt instrument, lacking the scratchily fine detail of the Aura's narrow EQ spikes and dips. Nonetheless it is more versatile, has different body shapes/styles and constantly variable sizes, and produces the major cavity and top resonances with no problem and in enough quantity to get you into serious sonic difficulty! Its ability to add in octaves allows one to add some artificial second harmonic or to fatten in the bass, and if the ersatz quasi-acoustic sound gets as tiring as it often does in the real plugged-in thing, then one can add in a virtual magnetic alongside a piezo model, and tweak everything from input to a range of speaker choices and miking positions to produce a harmonically rich electric sound.

Further still, it's a simple enough matter to rearrange processors, amp and the point at which the hex model enters the chain, and then inject a mono line from the guitar's real pickups and process that tone separately for level and tonal shading. You can, for instance, leave the hex EQ tweaks to the unit that sits in the guitar section, move the effects EQ to before the hex joins the chain and use it solely on the mono input. Alternatively, the wah can be used at a fixed pitch there to introduce a different colour. It is feasible to leave a hex EQ quasi-acoustic model uncompressed for dynamic range mixed later in the chain, and to move the compressor to before that point onto a mono magnetic input for a constant level of warmth at the back of the tone. Balancing between hex and mono for particular sounds can be tricky in some chains (the virtual mixer doesn't always have enough swing), but there are plenty of gain increase or decrease opportunities in some of the effects, such as the compressor or the EQs.

However, scattered throughout musical instrument production are some excellent-quality devices that have wider applications than their logos and promos might suggest. We just need to read behind the advert to understand the musical possibilities of a particular gadget's potential function. The tricks are to listen and select carefully, to learn to recognise aural fatigue, to clear one's head of the shinies and the blurb, to break with dogma and habit. By trusting instinct, ears, feel, and learning from an occasional bad guess, we can find more organic musical paths outside the bombast of corporate rock and the herd panics of the McMusic Biz! 0

Fishman's Aura: digital, but not modelling

If a gig is more problematic than expected, the DiMarzio Reference Acoustic soundhole pickup might get you out of trouble.

The summed output from my Graph Tech piezos running through the complex EQ characteristics sampled from acoustic guitars offered by Fishman's Aura creates quite a decent facsimile of a plugged-in acoustic. I picked the Graph Techs because they had the least offensive attack transient of the sets available a few years ago. Whatever attack transient you have, you're still going to have it through the Aura, and some of the Aura's EQ characteristics might well even accentuate a harsh one. The tone it adds has all the unevenness — lumps, bumps and scratchy bits — we might normally expect. It sounds to me like a mid-price-upwards acoustic with a well-sited soundboard transducer that has had simple subtractive EQ applied to take out the more dangerously unstable resonances. My soloist instinct, refined by more than two decades of nasty accidents, is to treat it exactly as I would a plugged-in acoustic — that is, to EQ out the scratchier mids, apply a little limiting and a dab of reverb. In fact, un-EQ'd and simply pushed through reverb to take out the dryness, it functions extremely well, and the note-to-note unevenness is organically redolent of thin steel stretched across a soundboard. Where it really seriously scores is when one or two whole-tone bends' harmonics move through the very narrow EQ peaks so that an unexpected and authentically wiry accent happens in passing. Which raises a question: how far or how near do we want to stay to an original acoustic character? Whole-tone bends are quite rare on a heavier-strung real acoustic, but with lighter, more electric stringing, we can take a texture from an acoustic slide guitar context and put it into an alternate picking style. Is that liberation or anathema?

One immediately obvious function for the Aura is as simple one-stop shaper for a band electric player who already has piezo saddles fitted and who needs to switch to a quasi-acoustic sound. Assuming a 'Y'-lead output with magnetic pickups on one line going to the usual effects and guitar amp, the other line could take saddle piezos through the Aura and into a DI so that any tweaks could be applied at the mixing desk. There is enough change for it to work convincingly as a 'less electric' rhythm sound for a background part in a couple of numbers. The most important element of that kind of part is the 'ching' that works along with the drum kit, and you might well be able to achieve that with the Aura, or merely with an A/B switch that simply reroutes your guitar away from the regular electric setup into a clean rig, or the PA, with the electric-ish mids scooped out. How much money do you actually need to spend on that? Your own ears will answer that one.

Is the amplified acoustic a dead end?

Acoustic guitars on stage