AudioRail® Technologies:   It simply works.

Behringer ADA8000 analysis

Click for image of front of unit
Click for image of back of unit

This is an 8 line/mic input + 8 line output 24-bit A/D + D/A (16 channels total) to ADAT Lightpipe with a street price of $230. This is currently the most inexpensive product of its class on the market.

Because we cite this product on our company website as being a cost effective companion to ours, it is well within the call to duty to make sure that we aren't citing a substandard product, which a $230 piece of equipment could potentially be.

The following is our analysis of the product.

Note: AudioRail Technologies is not affiliated with Behringer in any way, nor do we have any business relationship with them. We do not endorse any particular company's products as better than another's. Our multiple citings of Behringer products on our website is entirely due to the price of their products, which yield the lowest price points for end user configurations.

Functionality:

Inputs are separate XLR + TRS connectors. Outputs are XLR connectors. There is an input trim adjustment on each input channel, but not on output channels. Phantom power is on all channels or no channels on XLR inputs, using a single push button switch. Inputs have a single green LED for signal present and red LED for clip. Word clock is internal 44.1K, internal 48K, ADAT input, or BNC connector, selected by a slide switch. There is no word clock output BNC connector.

Internal design:

Alesis Semiconductor AL1101 A/D converters
Alesis Semiconductor AL1201 D/A converters
Alesis Semiconductor AL1402 ADAT optical decoder
Alesis Semiconductor AL1401A ADAT optical encoder
Toshiba TORX176 Toslink receiver
Toshiba TOTX176 Toslink transmitter
ST Microelectronics TL074C op amp and LM339 "comparator"
other discrete components

"The mic preamps in the ADA8000 do not use op-amps - they are based on a low noise conjugate-pair transistor pre-amp circuit similar to that used in our mixing consoles. We do utilise TL074 op-amps in the line-level circuitry, and the LM339 is used in its intended role as a comparator" -Andy Cullen, Behringer

Power supply is their design, a 3" toroid transformer driving a simple linear regulator configuration, using 3-terminal regulators: Several 7805, 7815, and 7915 parts, all in TO-220 packages, some on heat sinks, others not, which obviously provide +5V, +15V, -15V. The regulators are all very hot to the touch, too hot to hold your finger on, which is a concern that they do not have a lot of margin in their power supply. The transformer is completely cool to the touch.

Test configuration:

Echo Gina sound card driving ADA8000 with 12 foot Toslink cable.

8 XLR outputs looped back to 8 XLR inputs using short (6 feet or less) standard mic. cables. (Line inputs were not tested.)

Event EZ8 sound card receiving ADA8000 output using 12 foot Toslink cable.

Adobe Audition 1.0 and RightMark Audio Analyzer (RMAA) 5.3

8 channels are driven simultaneously with inputs that have the following 10-second sequence:

Silence: 2 seconds
Sine wave: 0.1 seconds (for transient response)
Silence: 0.9 seconds
Sine wave: 7 seconds (for FFT analysis)
Silence: 1 second

Sine waves were as follows:

Channel 1: 20 Hz
Channel 2: 100 Hz
Channel 3: 400 Hz
Channel 4: 800 Hz
Channel 5: 2000 Hz
Channel 6: 4000 Hz
Channel 7: 8000 Hz
Channel 8: 20000 Hz

Tests were then conducted with the RightMark Audio Analyzer (RMAA) using the loopback configuration with the "+4 dB" "9:00" trim setting, as before. Special thanks to Arny Krueger for help in getting RMAA to work for us.

Test results:

Screen shots were taken as follows: Note: These are high resolution images, some as large as 2048x1200. View at 100% scale for optimum readability.

ADA8000_FFT_ref.gif: This shows ADA8000 completely removed and replaced with a single optical Toslink coupler that connects the ends of the two fiber cables together.

ADA8000_FFT.gif: Input trim knobs set at +4 dB ("9:00" position). FFT scan was 5 seconds long starting at just under 4 seconds to just under 9 seconds.

ADA8000_FFT_mic.gif: Input trim knobs set at maximum, digital output driven at -57 dBFS. FFT scan was 5 seconds long starting at just under 4 seconds to just under 9 seconds.

When outputs were muted and 10 second recording was made with input trims at maximum, noise was white with peaks of -36 dBFS and RMS power of -47 dB referenced to RMS full scale sine wave.

ADA8000_FFT_mic_nc.gif: Input trims at max. No cables connected to inputs, digital zero sent to unit on ADAT cable, 10 second recording.

ADA8000_transient_1.gif: 20 Hz and 100 Hz waveforms shown.

ADA8000_transient_2_begin.gif: 400 Hz and 800 Hz waveforms shown.
ADA8000_transient_2_end.gif: 400 Hz and 800 Hz waveforms shown.

ADA8000_transient_3_begin.gif: 2000 Hz and 4000 Hz waveforms shown.
ADA8000_transient_3_end.gif: 2000 Hz and 4000 Hz waveforms shown.

ADA8000_transient_4_begin.gif: 8000 Hz and 20000 Hz waveforms shown.
ADA8000_transient_4_end.gif: 8000 Hz and 20000 Hz waveforms shown.

ADA8000_RMAA_crosstalk.gif: RMAA crosstalk test result
ADA8000_RMAA_IMD.gif: RMAA intermodulation distortion test result
ADA8000_RMAA_noise.gi: RMAA noise floor test result
ADA8000_RMAA_range.gif: RMAA dynamic range test result
ADA8000_RMAA_response.gif: RMAA frequency response test result
ADA8000_RMAA_THD.gif: RMAA total harmonic distortion test result
ADA8000_RMAA_summary.gif: RMAA summary window screen capture

ADA8000_RMAA_12_RMAA.sav: RMAA binary results file, channels 1&2
ADA8000_RMAA_34_RMAA.sav: RMAA binary results file, channels 3&4
ADA8000_RMAA_56_RMAA.sav: RMAA binary results file, channels 5&6
ADA8000_RMAA_78_RMAA.sav: RMAA binary results file, channels 7&8

Conclusions:

Waveforms show that the analog output signal is inverted going through the unit, with respect to the digital signal provided into the unit. (The analog input signal is not inverted in the other direction.)

Steady state FFT scans show roughly 70 dB S/N ratio at the +4 dB setting, reducing to roughly 40 dB S/N ratio at maximum trim sensitivity. Note: The 40 dB S/N ratio is due to line outputs driving mic. inputs at maximum sensitivity. The noise floor here is being driven by the line outputs being amplified some 60 dB, not by the noise floor of the microphone preamps. Do not be misled by this!

This unit should be fine for most live sound applications, but may or may not be suitable for the most critical studio recording, depending on the nature of the material, quality, and accuracy required.

We have experience using this product regularly in a live sound application and have encountered no problems in either reliability or sonic fidelity.