Behringer DCX2496 digital X-over

6 channel digital amp

Goskers, Steve (SFDoddsy here) is now using a Panasonic XR45 digital receiver (under $200 with 6x100watts) according to a recent post of his--and he claims it's fine with his tri-amped dipole array (Bob5). This certainly piques my interest, too,as I'm building similar dipoles now--and although I have three stereo amps to use with them I might want to downsize to one 6 channel amp in the future. Sure seems the ticket for someone with a small listening space or tight budget but wanting to go the tri-amp route. I wonder if there is/are any turn-on thumps in those 6 channels or if there's a delay circuit in them, though? Worth further investigation, I'd say...
Cheers,
Jeff in Hawaii:cool:
 
dcx2496 digital input

I have got one question about the Behringer DCX2496: As far as I understand the digital input accepts up to 96kHz, 24bit in AES/EBU format (up to 10V). Which high quality DVD-audio or SACD players are available which actually output this format? Otherwise, which adaptors from RCA to AES/EBU do you recommend?
I understood from some players I looked at that due to copyright requirements DVD-A and SACD data is downsampled to 44kHz before reaching the digital out. Is this correct or are there players which really output high resolution digital data? Do you otherwise have to use the analog outputs and then digitalize again when you enter the dcx2496?
 
Digital-in Behringer

@ rasputin

here you can download any Info available about the digital input:

http://www.behringer.com/DCX2496/index.cfm?lang=ENG

As far as I have understood (got the Ultradrive and Ultracurve just some weeks ago) and did not have the time to get familiar with them) the behringers will accept ANY current digital format of whatever you use as CDP/DVD/DAT -digital-out.
The 24/96-format is only the maximum of resolution and the internal format.

regards
Dirk
 
Improve the digital in on DCX2496

I'm working on tweaking/rebuilding DCX and I tought I share one of the tweaks.

The digital input in original form is far from great. Starting from balanced input connector the signal passes relay contacts (switching between analog/digital in) -> then some 10cm on PCB -> then cheap ribbon cable -> again PCB trace to digital transformer on DSP board. The transformer secondary is terminated by 110 ohm resistor.

So It is optimised for AES/EBU digital and even that is not so good. If one runs a SPDIF signal to BNC connector it will work but is far from optimal digital interface solution.

What I did is to run a 75 ohm cable straight to DSP board to leave out the problem of the connector alltogether. Then replaced the 110 terminating resistor (R15 underside DSP board near digital transformer) with a 75 ohm resistor.

Actually I went a bit further and also replaced the digital transformer with Schott unit which is of a known good quality. I do not know how good or bad the original is though.

All this makes the digital interface a lot better and listening test confirm it. There are clear differences between transports and after the tweak they can be heard.

If anyone is interested I can post some pics and more specific instructions.... I have replaced the SPDIF out circuit on my transport side also.

Ergo

PS. aah, if you do the tweak do not forget to cut the traces going back to ribbon connector way. Otherwise the whole chain is still mocking things up
 
Ergo'smods on DCX

Ok here we go :)

This is the schematic

An externally hosted image should be here but it was not working when we last tested it.



This is the TX side - inside Sony CDP-XB930

An externally hosted image should be here but it was not working when we last tested it.


This is the RX side - on DCX2496 DSP board

Notice the red markings - these are the places I cut the traces on PCB

An externally hosted image should be here but it was not working when we last tested it.


.....remove the 110 ohm resistor.....

An externally hosted image should be here but it was not working when we last tested it.


I removed the analog in C connector to make a hole for the cable

An externally hosted image should be here but it was not working when we last tested it.
 
Modding The DCX2496

Hello out there,
right now I did a very impressing mod I like to report here. I use the DCX2496 with an Usher six-Channel amp since a little time now. I like this unit very much for it's really good sound.

See http://freerider.dyndns.org/anlage/ueberblick.htm

Today I changed the caps in the output stages of the DCX2496. There are six 47uF caps and I changed them against Black-Gate-Caps BG-NX High-Quality. Not chaep! I payed 30Euros for the six caps!!

What a surprise! Now I know that the Black-Gates are no voodoo. Instruments got off the speakers right into my room and the stage wided a lot. I got so much more precision and a richer sounding bass. The hights become bright and clear I never heared before. This is a mod I really will recommend.

Next I will change the caps on the power suppy of the D/A converters.

Oh another thing to mention. The idle noise of the DCX2496 (I only hear at full volume) is audible decreased using the Black-Gates. The image shows which caps to change

http://freerider.dyndns.org/anlage/images-DCX2496/k_pic0008.html

Have fun ;-))
 
Hallo Oehlrich !

Would you mind helping me to translate this email I received from Quadral a few months ago ? It is about crossover parameters when using an active setup.
I am using Quadral Titan MkIII with a Behringer DCX2496 and I thought you might be interested.

Look here for the story


Thanks for the tip about the black gate caps, I have ordered 6 of them and I will put them soon in the crossover ;)

I would greatly appreciate your help.

Thanks

Thierry

Hallo Thierry,
Die unterschiedlichen Filter (Butterworth, Linkwitz, Bessel)
unterscheiden sich in der Form der Filterkurven.
Die Grundtendenz eines Butterworth-filters ist sehr gerade, ohne
Ein-und Ausschwingvorgänge. Ebenfalls ist der Phasengang
(Zeitverzögerung in Abhängigkeit der Frequenz) bei Butterworth
gutmütig bzw. ohne Welligkeiten.
Linkwitz und Bessel erzeugen steilere Filterkurven, wenn die
gleiche Rechnerleistung wie für einen Butterworth-Filter verwendet
wird. Umgekehrt bedeuted das, dass für Butterworth-Filter der DSP
viel mehr Rechnerleistung benötigt, wenn die gleiche Filtersteilheit (12dB/Oktave) benötigt wird.
Bessel und Linkwitz-Riley-Filter verlaufen fast gleich, der
Unterschied ist sehr gering. Klanglich sehr gut sind Bessel-Filter
mit den von Ihnen erwähnten 12 dB/Oktave. Damit sollten Sie gute Ergebnisse erziehlen.
Eine entscheidende Funktion ist noch wichtig: Können Sie den
Behringer zwischen FIR und IIR-Filtern umschalten? Die FIR-Filter
verdrehen die Phasenlage der Musik nicht, sind daher klanglich
sehr vorteilhaft. Die FIR-Filter (Finite-Impuls-Response, Filter ohne
Rückkopplung) benötigen sehr viel Rechnerpower. Sofern der
Behringer mit IIR-Filtern (Infinite-Impulse-Response, Filter mit
Rückkopplung) rechnet, besteht die Gefahr der Phasendrehung.
Sofern IIR-Filter verwendet werden, könnten Sie mit Butterworth-
Filtern das beste Ergebnis erziehlen.
Da in einer solche Anlage immer verschiedene Einflussfaktoren
eine große Rolle spielen, sollten Sie aber in jedem Fall
ausprobieren, wie die Musik am besten klingt ( verschiedene
Filtercharakteristika verwenden). Da auch die Elektronik
(Vorverstärker, Endverstärker) eine große Rolle spielt, kann meine Antwort nur eine erste Hilfe sein.
Freundliche Grüße nach Frankreich,
G. Stracke

Quadral GmbH&Co.KG
Am Herrenhäuser Bahnhof 26-28
30419 Hannover
Tel.-Nr.: 0511-7904-151
Fax: 0511-7904-116
 
Titan MK III

Hi Thierry,
the translation I read in the posted link is not so bad. So please tell me which point it is not clear an I will try to translate it in a better way.

Sorry for having not answered your fist posting. I am not that often in the forum and just did not read it.

The values of -9 and -6 db you got by Quadral nearly match the measurements I took Look here: http://freerider.dyndns.org/anlage/images-Messungen/00-Sweep.jpg

If you are interested in my configuration you can download it here http://freerider.dyndns.org/anlage/MyDCX.dcx

There are eight configurations included. They use four different sets of adjustments combined with the feaure for analog or digital input (4x2=8). The first configuration ist the simulation of the original passive filter, the second uses the same configuration with my digital room correction (DRC) for the bass. The third just uses Butterworth with 48db steep filters and DRC. The fourth (my absolute favourite) is also 48db Butterwoth combined with the result of the exact measurement also with DRC. In the third and fourth configuration the filter frequencies were 180Hz and 5Khz. I need the higher 5Khz frequency because the tweeter of the Vulkan is worse than the tweeter in the Titan. I guess 4Khz should be good for the Titan.

I am also interested in configuration. Please mail if possible. Thanks

Feel free to mail if i could give further assistance
Charly
 
Re: signature?

barfind said:
Hi everyone, new to this forum. ABout the behringer Xover being discussed, how would you rate its signature, I.E. Flat, nuetral, masked, etc. Thanks in advance.

Nigel


oehlrich said:
Hi Nigel,
what would you mean with "signature". I did not know this word in the HiFi context. Please explain.

Charly

Oehlrich, I am sure Barfind is wondering if the DCX2496 colours/taints the sound...

Barfind here are a couple of posts...

Oehlrich evaluation

And Vadim in the same thread

If you do a search on the DCX2496 there are plenty of people’s experiences and mods on this forum.

Good luck...
 
If I may add - I am also owner of 3 behringer units - src 2496;deq 2496 and dcx 2496, all connected via aes/ebu.

I ran my tri amped speakers through a marchand xover before - 48dB LR.

There is no difference in sound, except that the behinger is so much more flexible and allows one added very important correction measure - time delay adjustment besides a host of other things. Makes all the difference.
For the price nobody can beat those - and i use an ugly word in some high end circles - very neutral sounding units.

They allow you to really fine tune - via remote control through a laptop - the xoverpoints and various filte/dynamic eq settings.

I recommend downloading the manual before purchasing. This will allow you to somewhat familiarize yourself with the unit.

IMO - highly recommended.
 
Hi All

I've read that the DCX2496 has 10-15dB of attenuation available on the outputs.
Is the attentuation method used in doing this compromising sound quality ? -are you better off using resistor network etc. to reduce levels to those required instead and leave the outputs at maximum?

thanks in advance
 
Hi,
yes, you can increase gain or attenuate to +- 15db at the input as well as at the output.
As the level gets adjusted after/before the ADC/DAC for each in/output I suspect they work digitally.
I have talked to several people about bitloss through digital gaincontrols, but they are of the opinion that the quality loss is negligible.
That may be then a matter of taste, but as I am using the deq 2496 as well in the chain and that unit - all gains/attenuaition for each frequency band being adjusted in the digital domain - does not have any by me perceived negative affect on the sound.
Quite the opposite - the adjustment of the speakers to the room have been positive and I have used that unit and ist predecessor for several years now.
 
Bit Loss

The DCX 2496 offers two attenuation of 15 db one at the input one at the output. This sums up to 30db.

I did some tests if this attenuation is suitable for a volume control. My statement: It is NOT!

There are two reasons:

1.) All the volume control inside the DCX 2496 is IN FRONT OF the D/A converter. This means reducing the output signal by decreasing the input values of the D/A converter definitely leads to a loss of bits. Let’s calculate: 6db means half the signal. Half the signal is a division by two (a shift of one bit to the right) and so a loss of one bit. The maximum attenuation of the DCX2496 is 30db. This means 5bit. The unit uses 24bit so with an attenuation of -30db you only have 19bit left for the signal.
This bit loss is audible especially in higher frequency ranges. The sound gets a bit harsher.

2.) The second reason is noise. No unit is noise free. If you directly connect the DCX2496 to a power amp you will have some noise in your speakers even if the power amp for itself produces no audible noise. Decreasing the output of the DCX2496 by attenuation means decreasing the THD of the DCX2496 by max 30db. So from the >100db THD of the DCX2496 there will be left only >70db…. Not really good!

These facts lead to my recommendation to insert a 6-gang potentiometer between the DCX2496 and the power amp. In this scenario the DCX2496 can be attenuated to cover the full range. This means no bit loss and no decrease of THD.

I use a 8-CP2500 potentiometer from Thel-Audio ( http://www.thel-audio.de ). Not a cheap solution (350Euro) but the best you can get.