MIX Magazine - July 1998

SEVEN FOR THE ROAD

On the road, active direct boxes are often taken for granted, with sound engineers and stage techs routinely grabbing the same old DI out of the work box. Musicians and P.A. companies spend thousands of dollars on the finest guitars and monitors with little thought to their connection. We wanted to checkout several new active direct boxes that have come on the market recently and compare them to the leading active DIs used for live sound.

Seven DIs were chosen for survey, representing some of the top products used in touring:
The ARX DI-1, the BSS AR-133, the Countryman Type 85, the Whirlwind HotBox, the Klark-Teknik (K-T) LBB-100, the Radial Engineering (CableTek) JDV and Rapco's ADB+8.
Passive and tube DIs were not included, as we wanted to compare "apples-to apples," although they have their own applications. After these DIs were used for many months on tour with k.d. lang and on regional shows with Relles Sound of Eugene, Ore., bench testing and group listening tests were conducted at the author's Everett Station Labs in Portland, Ore.


Figure 1: THD measurements using a 40Hz sine wave input: At lower levels, the JDV and Countryman were lowest, with ARX and HotBox tied for third. At higher input levels, the ARX and Rapco had the lowest THD. X-axis represents input level.

All DIs tested were pin 2 hot, except the Countryman. Keep this polarity difference in mind when using the Types 85 in combination with other sound sources or other makes of direct boxes: You might consider either permanently labeling hem "pin 3 hot" or resoldering them.

Design Features
The BSS AR-133 is a refined, repackaged and slightly less expensive version of the old AR-116. Gone is the low-pass filter (which only works when the pad is in) and the phase switch. Added are 2 dB of headroom, a .5dB tilt toward the low end and a stylish aquamarinealuminum chassis with rubber end caps to protect it from the rigors of the road and from grounding out to an amp's chassis.

The other DIs are based on the Countryman's mechanical design, with the guts sliding into a rectangular aluminum sleeve. Unlike the Countryman, most of these can be repaired, should they need it.
Actually the klark needn't be opened, as it has no battery, nor do you want to open it, as the ends are mounted to the internal PC board. The Rapco is the largest, with the ARX a close second. All but the Countryman have something to prevent contact with an amp's chassis. The JDV has a neoprene pad across the bottom, while the other four have little (and in the ARX's case, not-so-little) rubber feet.

The ARX, HotBox, JDV and Rapco are transformerless. The BSS, K-T and Countryman all exhibit a rise in distortion below 80 Hz due to their use of a transformer, with the BSS showing the iron between the pickups and the consoles is what's needed to cut RF interference.


Figure 2: The IMD measurement was taken using the SMPTE method. Note: the JDV (lowest response) is a Class-A design.

Power
The Countryman and Rapco lack phantom power LED indicators, which are invaluable in troubleshooting situations. In the cases of the K-T, since it only runs on phantom, an indicator is also indispensable. The ARX has a momentary switch to check its batteries when a jack is plugged in. On the BSS and HotBox. The power LED indicates phantom power.

On the JDV, a momentary switch lights an LED when a jack is plugged in to show an adequate charge on its Ni-Cad battery pack, and it does not run on phantom power. The ARX , the HotBox and the Rapco all require two 9-volt batteries, while the BSS and Countryman use just one. The JDV runs on an internal Ni-Cad battery pack that recharges in a couple of hours from a DC wall wart and discharges over 24 hours. Because it does not "deep-cycle," it won't suffer from memory effect. The BSS and Rapco both have a power switch that must be turned on (and off!) and an LED that winks to confirm the battery's charge (continuously on the BSS, once on the Rapco). The ARX, Countryman, HotBox and JDV are turned on the only when a ¼ -inch jack is plugged in.

Changing batteries is the easiest in the ARX, which slides open, and the BSS, which has a small door that unfortunately is not captive. The Countryman, Rapco and HotBox are all opened with a phillips screwdriver. The ARX can also run on any 7-to 12-volt DC power supply. The JDV never needs changing.

Batteries are the bane of the stage tech's existence. How many times have you asked musicians whether there's a battery in their guitars, only to get a blankn stare? Batteries in DIs serve not only to provide backup to phantom power from the console, but sometimes are a better source of power. The ARX and Rapco improved their performance slightly running on batteries and, in the case of the ARX, on a DC power supply. Consoles with a heavy phantom load, bad mains AC or a poorly regulated supply may produce phantom power that degrades a DI's performance. We used a 9-volt-powered switching supply that operates at a frequency of 10 kHz to simulate noisy "brown-out" phantom power of 36 volts. We found that the Countryman more than doubles distortion for signals below -10dBu, but this impaired performance still beats many other DIs. The Klark's headroom drops by over 10dB, to under +5 dBu. Other DIs tested were mostly unaffected unless they had dead batteries. When did you last check the batteries in your DIs? (Don't give me that blank stare!)

Comparison Methodology
Formal listening tests were conducted first so that subjective opinions would not be colored by technical specs gathered during the measurement phase. Guitar tech Ron Johnson of Carruthers was on hand to assist with fine-tuning the instruments and lend his ears. Several musicians, each with different musical interests and styles, participated. For speakers, we alternated between a Meyer HD-1 studio monitor and a Meyer UM-1P floor monitor. Guitars used included two beautiful Taylor 414 acoustic 6-strings, one with a Baggs Ribbon transducer and another with a Fishman Matrix pickup: other guitars used included a Fender P-Bass Special, which has both Jazz and Precision pickups, and a Carruthers nylon electric. We also used a Spanish custom-made rosewood acoustic with a passive pickup and acoustic with a passive pickup and Gibson Les Paul. Listening tests were conducted using a Whirlwind Multi-Selector FS to select from a bank of four direct boxes hidden from view. Each bank was randomly chosen, matched for gain and the process repeated. Finally, measurements were performed with an Audio Precision (AP) System One, with a 47k resistor used to load each DI's input. Along with gain, signal-to -noise and phase response, we used the AP to examine total harmonic distortion (THD) with first 1k and then 40Hz sine waves, with amplitude swept over 50dB to simulate the range of contemporary stage instruments' signals. To look beyond these results we also tested intermodulation distortion (IMD) to see how cleanly the DIs reproduced musical signals.

The bench tests cleared up several of the seemingly contradictory observations made by several listeners. Comments like "cloudy lows," "covered-up highs," and simply "missing something" became meaningful when examined in the context of the bench test results.

Of course, listening observations are subjective, and playing style and musical context can determine the preference for one DI over another, but there can be no doubt that the choice of direct box affects the sound of instruments. And rather than present the listening results in the form of long -winded comments reminiscent of wine tasting (i.e., "the HF had a distinct effervescence, while the mids were dry with a hint of tolexed birch ply..."), we simply listed the group's overall consensus in each of several categories.

The Results
We began our measurements by examining gain and, to no one's surprise, found the seven DIs varied in gain features. The BSS, K-T and Rapco direct Boxes all produce unity gain, while the HotBox produces -8 dB. The Rapco ADB+8 has a three-position switch to add 8 dB (hence the name ) or sub tract 16 dB. The K-T has a -15 dB attenuator and all four have pads, allowing them to be used with line- or speaker-level in-puts: -20 or -40 dB on the BSS; -40 dB on the Rapco; and -30 dB on the K-T. The HotBox has three-way switch providing -20 dB of attenuation in the "line" position and 55 dB in the "speaker" position.

The ARX, Countryman and JDV direct boxes have output levels of +4, -10 and +2 dB respectively, relative to input. The ARX has a recessed potentiometer, allowing its output to be variably increased by up to +20 dB. The JDV has a 10dB input pad and, like the ARX, is not intended for speaker level input. The Countryman has no line- level pad, but a "speaker" Position has -50 dB of attenuation.

Next we looked at signal-to-noise. The Countryman and JDV topped out at better than 110 dB. Next were the BSS and the HotBox at better than 100dB. The K-T and ARX had better than 100dB. The K-T and ARX had better than 90dB, while the Rapco measured 87dB. The lower noise floor provided by the Countryman and the JDV make them a better choice for acoustic guitar pickups without pre-amps (no 9-volt battery in the axe), and both maxed out at about +10 dBu.

THD measurements were performed within a 22 to 22k Hz band. THD testing was first performed at 1 kHz while sweeping the level from -30 up to +20 dBu. At 1 kHz, the BSS, ARX and the Rapco exhibited the lowest THD for the widest range of inputs. At low input levels, the Countryman, JDV and BSS were the quietest. At the highest levels, the K-T, Rapco, ARX and BSS performed the best, with the Rapco having the most headroom, followed by the K-T, with the ARX and BSS in third. However, simply testing with a 1kHz sine wave can be misleading.

Using a 40Hz sine wave is a little more strenuous and simulates the lowest notes from a bass guitar, as seen in Fig.1. At levels below -20 dBu, the JDV and Countryman had the lowest THD, with the ARX and HotBox tied for third, making them good choices for guitars with passive pickups. On the right side of the graph we can see that the ARX starts to "brick-wall" at +14 dBu and the K-T at +15, while the Rapco is good until +18 dBu. The JDV and BSS are in trouble after +10 dBu, while the HotBox and Countryman are out of headroom after +6 dBu, but each of these has a line-level pad (except the Countryman, which jumps to speaker level).

At the higher signal levels that would be produced by synths and electric basses with active electronics, the ARX and the Rapco are clean for hot signals while the JDV's THD begins climbing. All three consistently placed as favorites, yet the JDV was the hands-down winner on most listening tests. Could the JDV be exhibiting "good" distortion?

This prompted us to take a different approach and look an intermodulation distortion (IMD). One common IMD measurement is the SMPTE method, which uses dual-sine wave stimuli consisting of a low-frequency (LF) tone (60 Hz @0 dBu) combined with a high-frequency (HF) tone at one-quarter the amplitude (7kHz @ -12dB). After filtering, IMD products appear near the HF tone spaced at multiples of the LF tone and are measured as a percentage of the HF tone. In our version, we swept the upper signal's frequency from 2.5 to 20 kHz. The JDV is the clear winner, with the Rapco ARX+8 a distant second the ARX DI-1 in third and the others tied for fourth close behind. It should be noted that the JDV is a Class A design. The highest spikes on the graph belong to the Countryman.

In phase and polarity testing, the JDV exhibited the smoothest phase response, with the Rapco placing second, which again was how these two ran in many listening tests. Above 10kHz, the Countryman exhibited nearly twice as much phase shift, with the others falling halfway between.

It would be great if one direct box were best for every application, but since we apply this criteria to our microphone inventory, it follows that some DIs have application-specific advantages. Musicians will find it interesting to make their guitar pickup, playing style and taste. Many commented that a particular model made them feel better about their playing. As with microphones, making a comparison is as easy as plugging into two channels on your console and matching the gain-you can even use an AB box to go back and forth.

I know that many readers use articles like this to confirm choices they have already made. Let those of you with aging DI inventories be warned that the performance of your oldest boxes will deteriorate long before they fail. For example, we compared the specs of some really old Countryman Type 85s with the new one and found quite a bit of variation. The time to replace your old DI is before it breaks. Don't take our word, though: Get a few musician friends together and have your own listening test. Don't worry; you can still use your old DIs to prop up stage monitors!

Mix sound reinforcement editor Mark Frink predicts a direct box category for the TEC Awards will come about sometime in the next millennium. He wishes to thank several technical gurus for their help: Fred Bijeh, Jim Kelsey, Lloyd Baggs, and Bob Taylor.

(-) zurück