Behringer DCX2496 digital X-over

Jump to Latest
#462
Happy happy happy eBay!

Needing 6 more ch's of amplification...Just using the DCX's EQ full range (no xo) and the transparency is remarkable for such cheap products.

Having way too much fun, and just scratching the surface,
-Andy
 

Attachments

  • hhebay.jpg
    hhebay.jpg
    58.6 KB · Views: 1,285
#465
Very nice job, Peter, they look beautiful. Any chance of a peek inside?

I got up and running on 8ch. recently. Though not DIY, independent volume controls are present. Everything is working very well, better than I dared dream of, actually.
 

Attachments

  • 8ch rack.jpg
    8ch rack.jpg
    56.9 KB · Views: 1,060
#467
Thmartin said:
Hi gents,

I drastically modified mine and I'm really happy with it.


DCX2496 tweaking for dumb

I'm now listening music through one of the best D/A converters around, honnest. Even compares with an excellent Goldmund DAC.

Sorry the website is in french, may be some day I'll have time for translation

Thierry


DCX2496 tweaking for dumb
Click to expand...

Could you tell us what was your mod. I was able to figure out from your site that you changed caps to black gates.
Unfortunately I do not sepak french.


:bawling:
 
#468
Basically I removed the active output stage and replaced it with 1st order low-pass filter.

The DACs are now feeding my amps directly withouth passing through the crazy amount of opamps and caps and resistors Behringer stuffed there.

Look at the new signa path in RED
An externally hosted image should be here but it was not working when we last tested it.


You can see the work it takes here

I'll ad a translation tool as soon as possible.

Merry Xmas
 
#470
"The DACs are now feeding my amps directly withouth passing through the crazy amount of opamps and caps and resistors Behringer stuffed there."

Are there any potential side effects to bypassing all that circuitry?

Behringer must have had a reason for paying for those parts and putting them in.

Thanks
 
#471
Besides the potential phase shift of the capacitor, it wouldn't handle long runs of interconnect nearly as well, and I hope the input impedance of your amplifier is rather high.

Note: These are my uninformed, stupid, assinie thoughts. I am dumb, and am probably wrong 😉
 
#472
noah katz said:

Are there any potential side effects to bypassing all that circuitry?
Click to expand...


Yes, the thing now plays MUSIC 😀




Behringer must have had a reason for paying for those parts and putting them in.

Thanks
Click to expand...


Yes :

  • 22dBU output level
  • output balancing
  • output impedance lowering to 10ohms
  • 24db/oct low-pass filtering

Now consider the side effects of Behringer's approach :
  • 22dBU output level : way too much for our domestic systems
  • output balancing : so what ? we all have unbalanced amps
  • output impedance lowering to 10ohms : the AK4393 has 600ohms output impedance. Low enough to feed a an amp with 47Kohm input.
  • 24db/oct low-pass filtering : the AK4393 has a built-in low-pass filter. What's the point ?
  • last but not the least : the bloody lot of IC's makes a lot of NOISE (hiss) and NO MUSIC 😡

I agree on the fact that you'll have to keep interconnects short. That may be an issue in some cases. I use 1ft 1/2 cables between the DCX and the amps.

Once you have tried this configuration you'll never ever get back to Behringer's solution. We compared it against a Goldmund DAC and the tiny DCX was damn close to it 😱
Before the mod, highs sounded vulgar and dynamics were almost non existent, typical PA sound.

Merry Christmas

Thierry
 
#473
Hi Thmartin,
how is the 600r output impedance isolated from the 33r passive filter?
The 3.3uS RC time constant would end up in the mid audio band if not isolated.
The 33r will permit many metres of interconnect cable.
Even 633r allows 2 or 3 metres of cable.
 
#475
Thmartin thanks for the explanation.
That is very similar what I did. I installed Lundahl transformers that are passing ballanced signals directly from AKM Dacs. Signal at the Lundahl's input is ballanced as well as the output. In my case it works great. Just like Thamrtin explained, you would nevoer go back.
 
#476
Konnichiwa,

noah katz said:
Are there any potential side effects to bypassing all that circuitry?
Click to expand...

You are replacing a 3rd order lowpass with a 1st order one, though it would be easy to use an LC Filter to get a 2nd order lowpass. This might result in some amplifiers reacting badly, as they have poor supersonic noise tolerance, but those are just plain bad amplifiers, full stop.

Another issue is that using an SE Output from the Balanced Chip output might cause some interresting low level problems, these are best avoided by actually loading both outputs of the DAC equally, I explained this in my original article on Valve Output Stages for modern DAC's published in the old "CyberValve" Mag.

The best solution might be a suitable 1:1 Transformer, wound to have a defined and well controlled HF Rolloff above 48KHz, which would give a 2nd order rolloff, DC blocking and an outout that has suitable levels for consumer gear and the option to be used balanced or unbalanced as needed.

noah katz said:
Behringer must have had a reason for paying for those parts and putting them in.
Click to expand...

Behringer has reasons for putting them in, I feel there are good reasons exist to nix them.

Even if one where to have an active stage, I would like to note that the kind of circuit used in the erlier Behringer Digital EQ's and other DSP units from them is much preferrable to that cludge in place now.

This other circuit could also be applied to the DCX/DEQ 2496 as mod with ease, as it is actually present in the second part of the original circuit, just omiting the first two Op-Amp's, but I still think a Transformer is preferrable.

Sayonara
 
#479
Kuei Yang Wang said:
Konnichiwa,

Another issue is that using an SE Output from the Balanced Chip output might cause some interresting low level problems, these are best avoided by actually loading both outputs of the DAC equally
Click to expand...

Hi,

This is due in the next website release, watch this space.

Even with the SE output the DCX2496 is light-years ahead from the crappy standard version, that's enough for me. The thing litterally smokes a BSS 366. It has been tested in 4 different systems including one with Goldmund amps and with Mosfet amps "à la Pass" ones.

Of course and as always : each on his own. Just wanted to tell people : been there, done that and liked it.

Merry Xmas

Thierry
 
#480
Thmartin said:
AKM4393 datasheet
Click to expand...

Thierry,

I think you better re-read the datasheet. The output impedance of the SCF is not specified and if it really were 600 ohms your output filter scheme (as correctly noted by Andrew T.) would be totally incorrect and bring the -3db cutoff frequency of the filter well down into the audio range.

The 600 ohms referred to in the specifications is a minimum load requirement for the device to meet the quoted specs.

Andrew T.....good catch on that.

Cheers,

Davey.
 
You must log in or register to reply here.
Top Bottom