Taking a compact DAT machine out and about can gain you some unique recordings, but can also present you with some unique problems. Martin Walker packs his shoulder bag and heads for the hills.
One of the most cost‑effective DAT machines launched over the last few years is the Sony TCD D7 portable (now superseded by the similar TCD D8). Although only the size of a standard Walkman, this 'DAT Walkman' is just as capable as many of its larger DAT cousins, but with the advantage that it can run on batteries, making it ideal for roving recording of sounds which will be unique to you and your music. If you're attracted by this idea, there are several things you'll need to sort out before you can begin.
Mic Choices
One of the difficulties for the recordist on the move is sourcing high‑quality microphones, suitable for stereo use, at reasonable prices. Although you can still capture many sounds in mono (spot effects, for instance), the beauty of real‑life sounds in the wild is that they are always in stereo, and already include natural reverb of a much higher quality than even the best Lexicon.
You could take two standard mics, mounted on a suitable stand (two mics can be bolted onto a 'stereo bar' which only costs around £10), but this might be a bit unwieldy on a regular basis for hand‑held work. You would also need to make up a 'Y' conversion cable, to connect the twin XLR mic connectors to the 3.5mm stereo jack input of most mini DAT recorders. For the odd session, you could always take a couple of normal mic stands as well (especially if you have someone else along to help carry things), but not if your chosen location is more than a few hundred yards from the car park! If you want advice on the many ways to mount and position coincident or spaced stereo mic pairs, Hugh Robjohns covered this area in great detail in the February and March '97 issues of SOS.
For regular outside recording, consider buying a stereo mic: suitable budget models costing between £150 and £300 are available from Sony and Sennheiser, amongst others. The choice is largely dependent on the sort of signals you want to record: for hard‑edged industrial and machine sounds, most general‑purpose mics will suffice, but for ambient or wildlife sources, the levels will nearly always be much lower, so low‑noise and more directional polar responses will be important. For distant sounds, a 'shotgun' mic, being particularly directional, will enable you to capture more of the desired sound and less background noise from other directions. If you're likely to do a lot of wildlife recording, it's also possible to buy parabolic reflectors that attach to the mic. These 'focus' the mic response (rather like the end of a torch, with the mic replacing the light bulb), although they can affect the frequency response.
Although there are various low‑cost stereo mics available for under £100, mostly of the back‑electret variety (normally powered from a single AA‑size battery), they are only really suitable for interviewing and other close‑up applications where a reasonable level of input signal is expected. If you're after capturing ambient sounds inside caves, or the call of a circling buzzard, for example, these kinds of signal levels can be extremely low, and in this case noise levels from a budget mic can cause a lot of frustration. I know that many people still use stereo mics of the cheap and cheerful variety, but the result of this can be DAT recordings with the same amount of hiss as a compact cassette. Since your recorder is portable, by far the best solution is to take it along to the dealer and try a few short recordings with several mics: this may save you an expensive mistake.
Headphones
The choice of headphones for location work is very personal, and much the same decision‑making process applies as for studio use. Open headphones tend to be fairly light and comfortable to wear, but if you are hand‑holding the mic, can be prone to feedback. Enclosed types tend to be hot to wear for long periods, especially if you're out in the sun, but have the advantage of letting you hear exactly what is going onto tape.
Sometimes you may want to work largely 'incognito', and cheap earpiece types are more suitable for this — for rough‑and‑ready sound grabbing, monitor quality is sometimes not so important, since most of the serious monitoring will be done back at the studio, when editing. The reason I sometimes use this approach is that holding a mic and wearing obvious headphones can attract interested passers‑by, and if you're trying to capture quiet ambient sounds the last thing you need is people coming up for a chat! I once stood for over an hour in the middle of a wood, with some wonderfully atmospheric wildlife sounds going on, but was unable to get in a single worthwhile take between a string of passers‑by!
Interfering With Nature
Since the human brain and ear combination is remarkably good at resolving wanted details in a morass of clutter (how else could we home in when someone mentions our name in a conversation across a crowded room?), location recordings often contain unwanted background sounds that were unnoticed in the heat of the moment. Back at the studio, it's often surprising how many sources of interference can render an otherwise perfect recording useless (but see 'Forensic Sound Editing' box). One recurring source of interference is light aircraft: these seem to take up to 10 minutes between entering and exiting earshot during an otherwise tranquil day. Other annoyances are distant chainsaws and lawnmowers, whose sound travels for several miles, and groups of people discussing their latest job frustrations whilst walking the dog. People‑related noises can be minimised by recording at unsociable times of day, such as early morning or during the night. This also lets you catch such 'unearthly' sounds as the night calls of animals and birds. It's advisable to carry a torch at night, but you do run the risk of a visit by the local constabulary if anyone spots you creeping about in the woods using it!
When it comes to handling noise, I suspect that most people will be surprised just how much noise they can make themselves, even when trying to be particularly quiet. If you attempt to hand‑hold the mic, the slightest movement of your fingers, hand and arm can create alarming clicks and rustling noises. The cheapest solution is to balance the mic in the fork of a convenient tree, and many people have used this impromptu approach. Alternatively, get hold of a cheap mic desk stand, as used for interviews, and just place this on the ground, or other convenient object, such as a large rock. This will remove handling noise altogether. If you need to 'aim' a bit more, try dismantling a studio mic stand, and take the boom arm with you (but be aware that, after a few hours, most will seem surprisingly heavy!) The ultimate solution to handling noise is one of those suspended cradles normally used with delicate high‑end mics. As long as the mic is suspended, handling noise should become a thing of the past. Cradles can be quite inexpensive (I spotted one in the Studiospares catalogue for about £30), and a convenient way to support one is with a pistol grip or extendable fishpole. Again, these can be obtained from most studio suppliers, such as Studiospares, HHB, and Canford Audio.
Mini DAT recorders like the Sony TCD D7 and D8 have a combined unbalanced mic/line input on a 3.5mm stereo jack socket. Most 'budget' stereo mics (under about £300) will probably be supplied with a lead already terminated in the required 3.5mm stereo jack plug, but otherwise you will need to make one up yourself. This is not such a pain as it sounds, since another potential noise problem area is the mic cable itself, and it may be worth making up a special cable to minimise this. Try holding up a length of cable very close to your ear and flexing it. Any creaking, rustling or crackles may be the braid moving internally, or the cores moving against the sheath, and these noises are more likely to cause problems with the unbalanced mic inputs found on typical low‑cost DAT machines than the balanced inputs of most mixing desks. It's worth replacing a cheaper mic cable with one that incorporates a cotton filler (which is always a pain to deal with when terminating cables, but really earns its keep, stopping internal movement). I made up a special cable for hand‑held location use; it's only a metre in length, but this is quite sufficient for most 'point and click' recordings. For wildlife sounds, much longer cables will allow you to get the mic close to a likely spot, while you and your recorder can remain hidden behind a convenient rock.
Getting The Wind Up
Of all sources of interference when location recording, the most frustrating must be wind (from the surroundings rather than the recordist). The slightest breeze can cause wild excursions on the input level meter, and give rise to overloading unless you're very careful. Many mics incorporate switchable bass‑cut filters, and this is a good way to remove the occasional low‑frequency rumble caused by a light breeze, without removing a significant portion of most desired signals. If nothing else, it will help prevent what used to be known in the days of analogue moving‑coil meters as 'banging against the endstop'.
A straightforward foam pop shield is rather ineffective for outdoor recording, although those with an air cavity between the foam and the head of the mic can be slightly more effective. The 'tapestry hoop and nylon stocking' shields used for vocals are not at all suitable for reducing outdoor wind noise, unless you experience one of those rare winds that always blows in the same direction! Professional wind shields do seem surprisingly expensive (starting at about £100), but once you've tried to work without one a few times you might be tempted to raid the piggy bank. (The expense of windshields is, by the way, another reason why using a pair of mono mics for field recording tends to become even less desirable.) The classic professional design is the BBC 'hairy sausage', which uses long, hairy, fleece‑like material to present a high‑impedance path to wind movement, while compromising the frequency response at audio frequencies as little as possible.
It is possible to improvise, if you don't mind getting odd looks from passers‑by. I have, on a few occasions, used an umbrella as a portable windshield. With the umbrella handle supported underneath your arm and the 'business' part resting on your head, you can divert most light gusts of wind sufficiently to prevent popping sounds, while still leaving both hands free. You can also angle your body to optimise the results, depending on wind direction. Because the 'windshield' is so far from the mic, this approach has the added advantage of not compromising the frequency response of the mic, which may happen when placing any object close to a mic.
Running Out Of Juice
Anyone who has ever tried using a portable DAT recorder on location will have come across the frustration of using rechargeable NiCad (Nickel Cadmium) batteries in the field. Seemingly, each time you buy a fresh set they last for three or four recharges, and the next time you need to use them a month later, they only hold enough charge to power the DAT machine for about five minutes. This is very frustrating when doing location recording, and it always seems to happen when you're trying to capture one of those unrepeatable sounds.
The beauty of real‑life sounds in the wild is that they are always in stereo, and already include reverb of a much higher quality than even the best Lexicon.
NiCad battery packaging always states that these batteries can last up to 1000 recharges, so we must be doing something wrong. If you look into it a little further you will find that there are various methods of prolonging the life of NiCad cells, and, more importantly, ensuring their reliability in the field. Before we get onto these, pause a moment to consider one alternative: alkaline cells (such as Duracell). Although non‑rechargeable, they do have double the capacity (and therefore operational life) of equivalent‑sized NiCads — with the Sony TDC D7, Duracells should last up to four hours in continuous operation, while a set of NiCads only lasts about two hours. If you have an important recording to make, always have a set of Duracells handy for emergencies. If your NiCads go belly‑up, you can swap them for the Duracells, which have a shelf life of up to five years.
Charging Cells
The so‑called 'memory effect' used to occur when repeatedly topping up partly discharged cells, leaving them in a state where only the 'topped up' part of the charge was effective, and the charge only lasted a very short time. This is technically known as a shallow‑discharge/full‑charge cycle, and used to happen particularly with devices left in a charger 24 hours a day, such as pagers and cordless phones. Modern cell design has largely removed this problem, but, as always, the myths continue that this is the primary cause of NiCad failure.
In fact, capacity loss problems are normally due to several other factors. If cells are stored for a long time without being used, they will slowly lose any remaining charge (at a rate of about 1% per day). When cells have run down by themselves, the initial recharge never lasts as long as it should, due to internal chemical processes. If you haven't used particular cells for several months, you'll need to 'wake them up' by going through two or three charge/discharge cycles. After the first charge, you might get 40% of normal capacity, and after the second charge 70‑80%. Finally, after the third recharge, the capacity should be up to 95% or more of its original value.
Long‑term over‑charging (normally when people continually leave the cells on permanent charge until needed) can result in the same lowered initial capacity. Again, a few charge/discharge cycles will normally restore cells to their full usable capacity. Long‑term 'trickle' charging can also create problems, although it does make using portable units with mains adaptors much easier, as you can recharge the cells every time you plug your unit in on return to the studio.
If you want to get reliable use from your NiCads, adopt a sensible charging regime. When you first buy a set of cells, look at the packaging to determine their capacity (measured in Ampere Hours or Ah. A cell with a rating of 1Ah will supply a current of 1 amp for an hour, on a single charge). Since portable DATs like the Sony DAT Walkmans take a fair amount of juice, it's worth looking out for NiCads with larger capacities. Always use the same cells in a set — mixing cells from different manufacturers is a recipe for disaster, as they will tend to run down at different rates.
My Sony TCD D7 claims a power consumption of 1.2 watts. Using four NiCad cells (each of 1.25 volts, giving a 5 volt supply) this works out to a current drain of 0.24 amps, or 240mA (milliamps). So if I use cells rated at 500mAh, they should last a little over two hours. However, if I choose some like the Ever Ready types I'm currently using, rated at 0.65Ah, a single charge of these should last about 2.7 hours — a useful increase of around 45 minutes per charge.
Before each recharge, the cells should ideally be discharged to a fairly low level. Don't leave your DAT player running with the Low Battery indicator flashing until it grinds to a halt (you may damage the recorder), or try shorting out the cells with a piece of wire to completely discharge them (you'll damage the cells). Often, a Low Battery indicator shows a suitable time for a recharge, but occasionally (once every few months) you could try the following procedure. Most battery chargers use a small light bulb to test the state of one of the cells in the unit. Once you know your cells are on the last legs of a charge, you can put each of them in turn into the test position, and leave them until the bulb goes out.
What causes more problems than anything else is charging cells for an inappropriate length of time. Contrary to popular opinion, you should not just leave them in the charger all day, and certainly not continuously until you need to use them. For the longest life, each cell will have printed on it the recommended charge rate, for example '50mA for 15 hours', or '200mA for 5‑7 hours'. However, unless you buy the same make of cells and charger, the chances are that the fixed charge rate provided by the typical universal NiCad battery charger will be different from the recommendations for your particular cells.
Charges are normally carried out at either of two standard rates: C/3 and C/10, where C is the capacity of the cell. For my current Ever Ready cells, with a stated capacity of 650mAh, C/3 is about 200mA, and sure enough, marked on the casing is 'Fast charge rate 200mA for 5‑7 hours'. Since most of us want rapid results, many modern chargers work at the Fast rate (C/3), and so should be used for a maximum of about seven hours. At the Fast rate, over‑charging can cause damage, so you could attach the charger via one of those mains timers used to persuade burglars that you are switching lights on and off. Set it up so that it will switch off after seven hours, and then you don't need to remember to switch off the charger at the correct time. Once your cells are completely charged, any further time in the charger ends up generating heat, and if you find that your cells are warm, this shows that they are already fully charged.
The other charging rate, C/10, is often known as Standard rate, and is slower (the same cell states 'Standard charge rate 65mA for 14‑16 hours'). At this rate you can normally leave cells in the charger for far longer than necessary with no damage, since the slow self‑discharge mechanism will help to ensure that over‑charging doesn't occur. The main thing to remember with NiCads is that if you find your current set running down quickly, try them for two more discharge/recharge cycles before throwing them away.
Country Pursuits
One thing that often surprises people, when they first start to record in the wild, is that certain sounds end up far 'bigger' or 'smaller' than you would expect. The classic one that catches most people out is the bubbling brook or mountain stream — the real thing always sounds like Niagara Falls, and what most people would consider a suitable 'stream' sound is actually a tiny trickle. Conversely, bird‑song is often embarassingly quiet compared to the background noises that we normally don't notice, such as the sound of other people's footsteps (and particularly other people's children!) Once you've wandered about 'under headphones' in the countryside, you'll always hear things in a completely different light. I once spent 10 minutes grumbling about the amount of background hiss on a woodland recording, before it finally dawned on me that it was the leaves rustling in the trees!
Forensic Sound Editing
The Noise Reduction plug‑in of Sonic Foundry's Sound Forge.If you make some great recordings that are let down by background noises, it may still be possible to extract usable sounds with the aid of computer noise reduction. Even the shareware WAV editor Cool Edit 95 has this facility, although it is unlikely to compete with more expensive specialist packages.
The main proviso when attempting to clean up audio of any description is the regularity of the offending noise. Continuous, unchanging sounds, such as hiss and hum, can often be reduced quite significantly using something like the Noise Reduction plug‑in of Sonic Foundry's Sound Forge, especially if there is a second or two of the noise by itself somewhere on your recording, from which you can generate a 'noiseprint'. (Always try to capture a snatch of any background noise by itself — if you are aware of it while recording — as this will make the process easier). A noiseprint is formed by spectrum‑analysing the audio signal, to build up a 'frequency picture' of the offending sound. Accurately subtracting these frequencies from the waveform then partially removes the noise, whilst largely leaving the desired signal intact. Noise from cheap mics and DAT machines can often be significantly improved using these methods.
If you don't have access to noise‑reduction software, all is not lost, especially if you have a parametric equaliser, or failing that, even one of the typical 'swept mid'‑style EQs on your mixer. You can often trim away significant parts of the spectrum, especially the extremes — 100Hz and below, and 6kHz and above — without totally destroying your sound. Using this in conjunction with a good enhancer or dynamic filter, you can still get rid of steady hisses without 'dulling' the sound too much in the process.
The Professional Approach
HHB PortaDAT professional digital recorder.
The professional portable recorder market was for many years dominated by the name of Nagra. These rugged analogue machines were used (and still are) by many TV, radio and film recordists when high quality and long‑term reliability were needed. Nowadays many professionals have turned to DAT, using machines like HHB's PortaDAT. This provides a rugged four‑motor transport for reliability, as well as four heads, for essential off‑tape monitoring. In addition, the rechargeable battery uses Metal Hydride cells, which don't suffer from the so‑called 'memory effect' problems of NiCads (see main text).
Other big plus points are the standard 48V phantom supply, balanced XLR mic/line inputs, and a built‑in limiter, which could be a life‑saver with the unforgiving nature of digital overload. Timecode becomes more important when recording in sync with timecode film cameras, and HHB have a model (the PortaDAT PDR1000TC) with this built in. The basic model costs around £3000, and is used widely by professionals around the world.
Alternatives To DAT: MiniDisc
All is not lost if you don't have a portable DAT recorder, as there are several alternative approaches. The reputation of Sony's MiniDisc format has gone from strength to strength, now that the latest version of the ATRAC compression system can sometimes even fool the experts in a side‑by‑side comparison with DAT. There are several portable models starting from about £350, but don't get caught out buying one of the much cheaper playback‑only models.
Compact cassette recorders are not renowned for quality sound, but can still be acceptable for grabbing the odd sound effect, especially if you have access to some computer‑based cleanup facilities (see 'Forensic Sound Editing' box). The main problem is likely to be high‑frequency noise, so if your desired sound is not particularly wide‑band, you can roll off the top end without compromising the sound too much.
There is, however, a distinct advantage to buying a second portable DAT recorder, in addition to a main rackmount studio model — you can make digital copies of your precious DAT masters, assuming that both machines have compatible digital I/O sockets.