inside BEHRINGER - DDM 4000 - 32bit Digital Mixer + new test results.

[Artesia]

A little while back behringer made a pioneer djm 600 mixer clone.. it sucked really badly on many levels, it was wrong and poorly built. they obviously got their fingers burnt, as their new flagship ddm 4000 digital dj mixer is quite the opposite ;)

(NOTE: new test results added afew postings down this thread)

http://www.behringer.com/EN/Products/DDM4000.aspx

(their new, rubbish website)

Dispite being lighter than you might expect, it is built as well as any mixer youd pay top dollar for; it packs more features & sound fidelity than its 'rivals' and rolls in at an obscenely cheap £250 / $300 :)

and honestly, after using it.. i personally couldnt justify buying the priceier alternatives ;)

hmm, its based on 2x 400mhz analog devices blackfin dsps & cirrus logic 114db adc/dacs.. under specced it is not ;)

certain widgets like bpm syncronised effects are increadibly fun & useful, also features eq and dynamics processing on all channels. the mic section also has its own effects section for adding reverb etc ..and supports channel ducking for announcements/radio talkover.

digital out also, however no digi ins :/ ..would be nice to see, as would remove one more adc/dac stage when using the cdj-1000's

the mixer was probably designed with being paired up with the cdj's in mind, as dispite the odd orange and blue dominated lighting scheme it actually fits in very nicely with the decks :)

also, took the time to test the players and mixer with midi control from the mixer to use it as a control surface for traktor ! ..and used the time code cd's in the cdj 1000's to control the scratching section of traktor to manipulate music on the pc also. i can confirm all of this works flawlessly :) ..supprised at that ! ..was expecting a niggle.

unfortunately, this does mean ill never build a mios traktor controller ;) :) ..ah well, more time for other projects :)

right.. down to business:

..and lights off, ooh ! pretty ;)

nice speckled silver/black finish :)

also modded the usual problem of overly bright blue leds, 4.7k added worked a treat.. tho maybe the next value up for some peoples tastes, only this doesnt burn the retinas out now ;)

EL backlit screen.

sliders are standard resistive tracks, not optical as reported some places !

nice, slightly resistive feel.. marked with a B in a triangle

new power opamps for headphones, supprisingly good.

manages to run even relatively insensitive headphones like my other set, the akg k702 reference cans quite loudly.

the sony mdr's are dangerous !

dsp card with two 400mhz blackfin dsps !

a supprising ammount of shielding & rfi filtering for a 'budget' product ;)

very simple smps ..not many linear psus in berry stuff these days.

the plastic screen guard needed wiping from the inside to clear some residue from manifacturing ..not bad, but a noticeable streek on the plastic.

main board:

* 2x xtal - 16mhz

* 2x headphone amps - cool audio V4580L high current opamps.

* 2X AMTEL - ATMEGA8 RISC PIC

8-Kbyte self-programming Flash Program Memory, 1-Kbyte SRAM, 512 Byte EEPROM, 6 or 8 Channel 10-bit A/D-converter. Up to 16 MIPS throughput at 16 Mhz. 2.7 - 5.5 Volt operation.

(lots of:)

* 74HCT165 - 8 bit parrallel in serial out shift registers

* 74HCT595 - 8-bit serial-in, serial or parallel-out shift register

* 74HC4051 - 8-Channel Analog Multiplexer

* 74HCT244 - Octal buffer, line driver; 3-state (afew)

main board, analog i/o section:

* jrc - 4580 - dual bipolar opamp, 0.0005% dist, 15mhz, 5v/us slew, 0.8uv/vn (standard behringer faire)

* st - 074c - quad jfet opamps, 0.01% dist, 3mhz, 16v/us slew, 15nv/vn (old opamps, very odd !)

dsp card:

* 2x analog devices - black fin, adsp-bf532 - 400mhz !! (800 MMACS) 32 bit dsps.

http://www.analog.com/en/embedded-processing-dsp/blackfin/adsp-bf532/processors/product.html

* samsung k4s281632i-uc75 - 128mb sdram, 3.3v 133mhz/cl3

* hynix - hy57v161610ftp-7 - sdram, 3.3v, 16m, 64ms, 133mhz

* xtal - 25mhz

* xtal - 11.289

* 74LVT244 - Low Voltage Octal Buffer/Line Driver

* 1x cirrus logic - cs4351 - 24-Bit, 192kHz, 2ch dac, 112 dB, -100db thd, with Line Driver

* 2x cirrus logic - cs4345 - 24-Bit, 192kHz, 2ch dac, 105 dB, -90db thd (marked 345c071)

* 4x cirrus logic - cs4272 - 24 bit, 192khz, 2ch adc & dac codec, 114db snr, -100db thd/n.

* 1x cirrus logic - cs4271 - 24 bit, 192khz, 2ch adc & dac codec, 108(adc)/114db snr, -98/100db thd/n.

[Artesia]

hmm.. no response :) ..they went nuts for this one on speaker plans ;)

[abcmann]

I tried this this mixer too. Its REALLY not feeling like Behringer...... and thats meant a good way ;P

The only thing what could be better are the sliders.

If it doesnt disambles itself and if the pots wont suck after 2 month this is really a very reasonable mixer ! 

[nebula]

Nice to see maybe sometimes you do get more than you pay for.

I don't like most Behringer stuff, although I do like the build quality of the BCF/BCR 2000 controllers.  I often find their stuff has underspecced power supplies that run too hot, which can lead to other problems.

[Artesia]

Hi again,

finally had the time to run some diagnostics on this little unit to determine how far it can be pushed signal wise & heres the results gleaned with passable, but less than optimal equipment.

when discounting what anomalous distortion the measuring equipment adds to the data, id have to say this mixer certainly should perform pretty much to its spec sheet breakdown :)

the behringer blampf:

AUDIO INPUTS

Mic 1/2 (XLR, electronically balanced)

Max. input level            -14 dBu

Input impedance             2 k?

Phono (RCA)

Max. input level            -18 dBu

Input impedance             47 k?

Line (RCA)

Max. input level            +30 dBu

Input impedance             15 k?

AUDIO OUTPUTS

OUT A (XLR, balanced)

Max. output level           +21 dBu

Output impedance            200 ?

OUT A / OUT B / TAPE (RCA)

Max. output level           +21 dBu

Output impedance            100 ?

Phones Out                   max. 260 mW @ 100 ? / 1 % THD

S/PDIF                       (coaxial, 16 bit, 44.1 kHz)

EQUALIZER

Stereo Low                  -? dB/+12 dB

Stereo Mid                  -? dB/+12 dB

Stereo High                 -? dB/+12 dB

Mic Low                     -12 dB/+12 dB, Shelving

Mic Mid                     -12 dB/+12 dB, Peak

Mic High                    -12 dB/+12 dB, Shelving

DIGITAL EFFECTS PROCESSOR

DSP                         2x Analog Devices Black Fin

AD/DA converter             24-bit Sigma-Delta/128-times

                              oversampling/Cirrus

Sampling rate               44.1 kHz

LCD display                 320 x 40 pixel

MIDI interface               5-pin DIN jacks In/Out/Thru

SYSTEM SPECIFICATIONS

Frequency response

Mic                         25 Hz – 20 kHz, +0/-3 dB

Phono                       20 Hz – 20 kHz, +0/-3 dB

Line                        20 Hz – 20 kHz, +0/-3 dB

Signal-to-noise ratio

Mic                         > 87 dB

Phono                       > 83 dB

Line                        > 102 dB

Distortion (THD)            < 0,009 % (Line – OUT A)

Crosstalk                   < -80 dB / 1 kHz (Line)

POWER SUPPLY

Mains voltage / fuse

100 – 240 V~, 50/60 Hz      T 1 A H 250 V

Power consumption            max. 20 W

Mains connector              standard IEC receptacle

DIMENSIONS/WEIGHT

Dimensions (H x W x D)      approx. 4 1/3" x 12 3/5" x 15 1/2" (110 mm x 320 mm x 392 mm)

Weight                      approx. 9.3 lbs. (4.2 kg)

MY Blamf:

the simple, general rule of thumb gleaned from the bench testing below:

* keep input signals on or below the yellow 0db mark to maximise resolution/quality of incoming signal, whilst allowing sufficent internal overhead to prevent effects and eqs clipping.

* keep output below +8db for cleanest sound, try setting the gain of your pa to match this (with limiting applyed beyond this).

* definately do not exceed 12db if you want your sound to remain clean.

* itll sound crap & youll probably break things attached to this, if you push it waay into the clip region ;)

The full breakdown:

NOTE: measured with ARTA & Saffire le soundcard, performance of unit may well be higher on accounts of distortion & noise floor.. as the figures quoted will include what marginal hash it generates too.

disregard the 156hz peak, this seems to have nothing to do with the mixer.

Also noise & 1k sine sources used as an arbatary standard of sorts for these tests, no notable increase/decrease in distortion was noted with changing frequencys.

dig out:

* noise floor - -105db 100hz, 4.1k, -110db 20k, -83.5db

* 1.15v pulse train, well defined edges.

Line IN (ch1):

line modes:

* full gain without raised output noise floor.

* With max gain - +14dbu/(-12dbfs)/1.81v (at 1k) without clipping.

  (shows +9 clipping on ddm)

* gain trim range - 40db

* freq resp appears to be ruler flat

phono mode:

* max gain noise floor - -75db 100hz, -94db 7k, -85db 20k

output a:

* +21 dbu before distortion.

* +8db on ddm meter (11v) cleanest sound setting. - (THD 0.011%, 21 dbu)

* +12db on ddm meter (14.5v) max 'clean' setting. - (THD 0.028%, 23.5 dbu)

* clip starts to flicker on ddm meter (23v). - (THD 0.83%, 27.5 dbu)   

* beyond clip on ddm (27v) power rail limit. - (THD 1.63%, 29 dbu)

* 27v peak, for 11v out (knob at 12 o clock) before marginal distortion rise.

* opamp lower power rail seems to collapse before upper.

* any output level, neglegible noise floor does not rise.

* scope measured hf noise, mostly out of audio band: 22mv/rms aprox, 45mv peak.

output b:

* +21 dbu before distortion.

* same breakdown as above.

* any output level, neglegible noise floor does not rise.

* scope measured hf noise, mostly out of audio band:

1500 NS cycle semi random sine noise (666.66KHZ)

115mv rms, 190mv peak

tape output:

as with b, but gain fixed to 1:1.

headphones:

15.5v out - clean output (+20.5dbu, thd 0.025)

18.5v clip limit (+25.5dbu, thd 0.11)

(no load)

Digital out, idle all inputs set to minimum level.

Phono in mode, with gain set to maximum.

Standard eq settings, low and high set to max gain.

Standard eq settings, mid set to max gain.

Standard eq settings, low and high set to min gain.

interestingly, the cuts slope rather than shelf as with gain.

after a little thought i concluded this is probably a better idea ;)

Standard eq settings, mid set to min gain.

around about -70db dip.

output a, at 11.5v.

output a, at 14.5v.

output a, at 23v.

output a, at 27v. rail limit.

headphone out 15.5v (no load)

headphone out 18.5v (no load)