DIRECT FEEDS
                        Why, What, and How.
    A number of secondary techniques used in recording seem to go in and out 
of style every few years, like trench coats and double breasted suits. Some of 
them go out and stay out, and we're well rid of them, but every once in a while 
a really useful gimmick gets shunted aside in the relentless quest for new and 
more perfect sound, when it doesn't seem to work as advertised, or just doesn't 
fit in with "The  way we do it now." Engineers, at least professionally, are 
nothing if not stylish. 
    Direct  feeds are a classic example of this sort of thing. Almost every
studio owns a couple of directs, and most engineers use them here and there, 
but very few take full advantage of these remarkably helpful little boxes. 
    Aside  from being about half ugly, directs are probably under used because 
one of their main functions is somewhat obscure and because they are commonly
used as the only pickup on a given instrument. That's too bad, as it's possible 
to make the housing look decent, the concealed function is critical to both the 
musicians and the studio, and directs don't have to be used as solo acts. In 
addition, they sound better than mikes. 
    That last is a little off. Directs don't sound better than mikes. Directs 
sound better than woofers. 
    Musical instrument amplifiers generally don't have tweeters in them. They 
come with one or two very high quality 12 to 15 inch woofers, and that's it. 
    There are good reasons for not putting in tweeters, including expense and
reliability, but not having them leads to problems. Good woofers produce plenty
of low end and a fair amount of mid-range, but when an amp tries to move those 
big soft cones fast enough to put out high frequencies, they wrinkle up instead 
of producing top end, and fall dead at something around 5 Khz. The cones also 
put out very little in the way of transients as they again develop pleats 
instead of pushing air. Stiff cones work better, but the resonant point of the
speaker rises rapidly as the cone is stiffened, and the speaker runs out of low 
end as it gains top. This is a no-win situation, and there's no such thing as a 
single full range speaker. 
    There are a few amps around with tweeters in them, and they help an audi-
ence no end.  Unfortunately, they don't do a whole lot for the engineer, as 
they are separate from the woofers, and he's got to mike the amp from three 
or four feet away in order to get both the woofer and tweeter. At that distance 
the mike will pick up everything else in the room, and sounds pretty dull to 
boot. 
    Co-axial speakers would solve most of these problems, and a good case can 
be made for spending the money to shoehorn them into dedicated studio ampli-
fiers, but in addition to being big and heavy, they're not made to sustain 
continuous high power use, which makes them poorly suited for general stage 
work. 
    All of  which brings us back to the direct, which avoids the speakers  
altogether, and sometimes the amp as well, while quietly bulletproofing the 
musicians against shock hazard. 
    Bulletproofing first, as it's simpler. 
    A properly made direct with no repeat NO grounding switches (none) will 
nail an amp down to the console ground through it's cable to the console input. 
    This is a very strong ground because the mike cable shield amounts to about
#10 wire, and XLR connectors are rated at 10 amps continuous current per pin. 
    With the direct connected, both the amp and the instrument will be at
microphone ground come what may, and that's good, as what may come is a life 
threatening situation.
    Instrument amps have a .05 mfd local grounding capacitor connected from 
the  A.C. line to the amp chassis. The cap is switched from one side of the line 
to the other by the three position "on" switch, so the cap is in the circuit if the 
amp is on. 
    Capacitors fail. If the local grounder cap fails and shorts out, the amp 
will either be at the A.C. neutral voltage or 120 volts A.C. above it. That 
puts the attached instrument and the musician playing it 120 volts A.C. above 
the nearest mike, and a fella could get killed that way. 
    That sounds like a remote possibility, and it is. On the other hand, the 
writer has seen grounding caps short on two occasions; once before a session 
and once in the middle of one. In both cases the amp was on a direct feed, so 
the result was to blow the amp's A.C. line fuse rather than allow the amp to 
go hot to ground.  
    One musician was pretty upset, thinking we had somehow blown his amp. The
other understood what had happened pretty quickly and was grateful that the 
direct had saved him from a bad if not fatal shock. The gratitude  was welcome, 
but the other guy's people would probably have everybody involved in court 
cases to this day if he'd got killed, and it wouldn't have done a damn thing 
for the studio's reputation. 
    It may seem preposterous that a musician's life can hang on a 70 cent 
capacitor, but it's a fact. The direct's ground insures both him and the 
studio against this threat, and if the pipe ground (the round pin on the power
plug) is lifted with a 3 to 2 pin ground lift adaptor AFTER the direct has 
been connected, the amp will be at absolute console ground. There will be 
none of the little shocks guitarists and such constantly get (and hate) and
the intractable hum problems typically associated with directs will dissapear. 
All that will make the players feel that somebody is really concerned for their
welfare, which they like, and make your ife easier. 
    Onward to what: 
    There are lots of directs on the market, and they all work. Some work 
better than others, mostly because of the designer's background. We all know
that most professional recording equipment seems to be laid out by people who 
know a great deal about electronics and such, but very little about actual 
recording. It's as if nobody in Detroit knew how to drive, and the car 
companies depended on letters from their buyers for design input.  Directs
are a prime example of this sort of thing. Because of power transformer hum
fields around amps, a direct is  almost always placed on the floor beside 
the amp. In fact directs live on floors, getting kicked, stepped on, and dirty.
Despite this, they are commonly built into lightweight boxes with three colors
of paint and a few tiny  switches sticking out here and there. Give one of 
those things a few weeks on the floor and it'll self destruct. 
    Admittedly, a handy pad switch on a feed saves a bit of time in setup, but
it is the writer's opinion that a direct box should be capable not only of 
sustaining abuse, but  inflicting it, and I know of no switches that can be 
used to drive nails. A second output XLR at minus 20 will do the job without
sissying up the box and is more reliable anyway. As to the box as such, 
electrical boxes work pretty well, provide considerable shielding, and cost
almost nothing. They are also ugly, and if looks are important, Bud and 
Hammond make some very nice little metal boxes in steel and aluminum. Newark
Electronics (branches everywhere) carries both. 
    As to design, the simpler the better. As with any tool, a direct should 
not be interesting. It should just work. 
    The simplest designs use transformers, which are as reliable as lead 
balls, as they have no moving parts. Figure one is such a design, and is a
little more sophisticated than it looks. 
    The trick here is that the transformer is a bridging input at about ten
times it's marked impedance. For that reason, it's a decent match for a guitar
pickup when, for some reason, there's no amp for the instrument. It's a poorer
match for a bass, which is normally fed into one jack with the amp fed from
the other. 
    That works out OK because a bass player sets all his instrument controls
at wide open, while guitarists usually play at about half volume to allow 
turning up for solos. Lastly, the transformer is a terrible match for a power
amp at about 1/10 ohm. The result of these varying matches is that everything
comes out at about the same level. Handy.
    The transformer drawn is a TRW/UTC series "A". These and similar mike input
transformers are great for directs as they are moderately priced, well shielded
units, with a plus 8 Dbm rating, (strength of ox) and they sound terrific. 
    The UTC's are off the market, but they can sometimes be found in a studio 
junk box, which makes the price really attractive. 
    If you can't find any A's or equivalents, Jensen Transformers in Holly-
wood makes a dandy direct transformer designed specifically for the purpose. 
It's a first rate unit, and much smaller, as it's not shielded.
    Again, good stuff. The bad stuff is little half ounce transformers. Big 
fat transformers turn out big fat sound. Little ones sound tinny even when
they don't distort, and probably account for some of the common  misgivings
about transformer directs. They also generate more noise; the cheaper the 
noisier. 
    The how to part of all this is real easy. Put a direct on anything that
comes out of a speaker. As an example, a direct on the B-3 Hammond will get
you some actual bass from the instrument, although the key clicks will need
equalizing out. On guitars, synths, and the like, the direct will supply a
warmth and presence that cannot be picked up any other way. In fact, it's so
good a musician may comment that it doesn't sound like his instrument. A fast
playback on the Auratones will reassure him, as they're pretty close to the
amp speakers, so he'll hear what he expects on them. 
    Directs sound wonderful, but they also transfer (not generate) some prob-
lems. A direct transfers everything that comes out of an amp into the console, 
including some things you don't want to hear. 
    In the case of guitar amps, these include amplifier noise, key clicks, 
fingering noise, a lot of trash from cheap toys if the artist is using them
and, occasionally, too much presence. Most of the trash can be either removed
with eq or masked by using a mike a couple of feet off the amp. Don't get too
close to the speaker with a mike or you'll run into phase cancellation 
problems, but the combination of direct and mike is so flexible that you can
get about anything you want from the amp. 
    Guitar players use their amp tone controls as part of their sound, so it's
best to pick up the amp output rather than using the direct between the axe 
and it's amp, which is optimum for other instruments. Using a direct on the
preamp output of a guitar amp is easy, but it's the writer's experience that
it never sounds quite right. That probably has something to do with going
through a chain of inductors, and starting with one (the speaker) makes some 
sense from that standpoint. 
    A couple of final notes. 
    One, a direct will be flat to about 10 cycles, so if the feed seems to
be muddying up everything, particularly on bass, try cutting it off at about
60 Hz. 
    Two, if you want to know what directs sound like, listen to a country
record. They use 'em. 

       Summary: Aside from the obvious isolation advantages, direct feeds 
   prevent electrical shock hazards to musicians and deliver far better sound 
   that can be gotton from music amp loudspeakers. They have been a part of the
   recording scene for more than 40 years, and are worth listening to on any 
   instrument that will permit their use. 
                              -30-   
                           Words 2068