Behringer DCX2496 digital X-over

Jump to Latest
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
modmix said:


The in stock situation also has some Cs in the output section (e.g. 220p || 2k32 at IC1B). I didn't do a simulation - just a guess: could be a kind of preemphasis (just do have some more bits & less noise in a range with usually lower levels).

This theory doesn't fit well to the obeserved balancing problems... hmmm...

Ulli
Click to expand...


Ulli,

I did a sim of the output stuff, it doesn't have the complement of the input. The caps you mention roll of the response way over audio country and are more like compensation caps for the opamps. 220pF and 2k32 have a cut-over freq of about 300kHz.

The bass balance could be caused by the removal of coupling caps and going to DC coupling which of course extends the freq response down to 0 Hz. No proof, just a possibility, I haven't measured it.

Jan Didden
 
M

modmix

Member
Joined 2006
Last year I suspended my trial to use a dcx2496 in stead of a passive crossover (which was more difficult to design than I have expected ,-). I got two replies I didn't answer at that time (the short answer would have been: I just wanted to change the tweeter - not my entire equipment):

Frank wrote (Post #1349)
Isn’t it already there digital radio via e.g. Astra?
Click to expand...

Jan wrote (Post #1350)
Why don't you run it with digital inputs? And if you have an analog means to set your level in your power amps, even a simple pot to set it one time, you can run the DCX at 0dB levels overall, and most of the things you mention would go away.
Click to expand...

Jan's 6 channel mod makes some of my wishes come true, I guess.
Today, the Pilgham kit did start to run (rs232 isn't working yet for reason unknown; just wires... next weeked...).

The only other (-ive) points were:
* Slight tonal change (esp on the analog inputs) which took a bit of getting used to.
* No input buffer for the analog C input.
* not ideal for the first time builder, making the IDC cables and the high component density made the legend difficult and hence pick correct component placement.

But all in all a positive move and worth the effort.
Click to expand...

Yes, I did have to ask for some more IDC stuff (fast delivery from Ward Mass. Thanx!)
Yes, I didn't feel weel without a circuit diagram - just colours to numbers...

But I fully agree: worth the effort.
And: I did enjoy those "take a break" hints a lot.

Up to know I used just one remote control - never tried one of these
multi-learning-stuff... I still don't have to: the learning mode is simply great - no reason to worry about.

Sorry, no comments on tonal changes as too much time is gone since my last use of the slightly modified box (caps, level, opamp). But I definitely can say that I like what I've heard in the last couple of hours with just one channel using my passive crossover. Looking forward to run it as an active crossover.

Good time to say a big THANK YOU ALL!
If learned a lot and I got much improvement.

Ulli

PS:
All that in addition opens to me a path to more digital sources - not to bad at all.
 
L

Lucifix

Member
Joined 2004
luckit.biz
DAC4392 mods for Behringer

    Hello,
  last year I made this board as an improvement to my existed DAC but one friend of mine asked me if I can adjust this module for DCX2496. So, I changed the firmwares accordingly and now the '4392 I2S Output Port can accept external bit / word-clock signals (from 16 to 192 KHz sample rate) from DCX2496 or other external source. In addition, I added the option to manually change the caracteristics of the ASRC filter (Decimation Filter with or without Direct Down-Sampling option enabled) and the direction of the I2S output port (Master / Slave option) using only 2 jumpers. The momentary switch toggles between all the inputs available (USB, Optical, RCA, BNC and External I2S).

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

Here's one example where this board is used with ESS Sabre DAC eval. module:
http://www.diyaudio.com/forums/showthread.php?postid=1485660#post1485660 and
http://www.diyaudio.com/forums/showthread.php?postid=1485726#post1485726

Best Regards!
Lucifix
 
AR2

AR2

Master Burner
Joined 2002
burningamp.com
Re: DAC4392 mods for Behringer

Lucifix said:
    Hello,
  last year I made this board as an improvement to my existed DAC but one friend of mine asked me if I can adjust this module for DCX2496. So, I changed the firmwares accordingly and now the '4392 I2S Output Port can accept external bit / word-clock signals (from 16 to 192 KHz sample rate) from DCX2496 or other external source. In addition, I added the option to manually change the caracteristics of the ASRC filter (Decimation Filter with or without Direct Down-Sampling option enabled) and the direction of the I2S output port (Master / Slave option) using only 2 jumpers. The momentary switch toggles between all the inputs available (USB, Optical, RCA, BNC and External I2S).

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

Here's one example where this board is used with ESS Sabre DAC eval. module:
http://www.diyaudio.com/forums/showthread.php?postid=1485660#post1485660 and
http://www.diyaudio.com/forums/showthread.php?postid=1485726#post1485726

Best Regards!
Lucifix
Click to expand...

Looks great. And the price is?
 
Moby

Moby

Member
Joined 2003
janneman said:
From a technical point, I can give at least two reasons for the change in the analog input.

- In the stock situation, the analog input signals goes through 3 not-so-hot opamps. In my active mod, that trio has been replaced by a single very high performance fully balanced opamp (OPA1632);

- In the stock situation, Behringer slightly rolls UP the mid/high frequencies (up to +6dB at 20kHz) as shown in the simulator pic. I don't know why, possibly to improve the subjective 'presence'. For me it sounded a bit unnatural, though. My new analog input is ruler flat.

Jan Didden
Click to expand...
Jan, I'm in the progress of replacing analog input stage of my DCX, But I'm really confused with this +6db shelving
:confused: How the stock unit can be flat (well almost flat) if they tried just to make "subjective presence"? Regarding SPL curve boost of 2db at 3khz is not subtle ;-). Did you measured your input modification flat? I'm talking about A/A mode. Also, if the Behringer added -6db rolloff in the DSP it can be counted like digital input dull sound. Anyway, I'm still confused, please let me know does your unit with modified analog input (with BB1632) measure flat at the output?
 
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
Moby said:

Jan, I'm in the progress of replacing analog input stage of my DCX, But I'm really confused with this +6db shelving
:confused: How the stock unit can be flat (well almost flat) if they tried just to make "subjective presence"? Regarding SPL curve boost of 2db at 3khz is not subtle ;-). Did you measured your input modification flat? I'm talking about A/A mode. Also, if the Behringer added -6db rolloff in the DSP it can be counted like digital input dull sound. Anyway, I'm still confused, please let me know does your unit with modified analog input (with BB1632) measure flat at the output?
Click to expand...

Hi Moby,

Sorry for the delay. I did a measurement today, from analog in to analog out, everything flat, no filters, no digital gain. The analog input was at +6dB. See below.

Jan Didden
 

Attachments

  • dcx analog.pdf
    2.9 KB · Views: 225
voodooless

voodooless

Member
Joined 2004
I measured the original output stage, and it does not contain any shelving... It is ruler flat.

Anyway, my new plan is to replace the complet DAC and output section with a new one :) I'm going to use the new ESS sabre DAC for it. I even have two channels left that I could use for bypassing the DSP to make a headphone output or something else... Should be a fun project. But for now, I'll have to wait until the boards are ready, that might take a while. Also I don't think this is going to fit into the original casing ;)
 
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
4real said:
I measured the original output stage, and it does not contain any shelving... It is ruler flat.
[snip]
Click to expand...


OK, thanks, I hadn't done that. My measurement from analog in through analog out was with my replacement board.

But, since the shelving is in the stock input stage (both shown clearly in the circuit diagram as well as confirmed by my sims), that would mean that the stock unit has additional 'presence'. Maybe that is what some people actually didn't like.

Jan Didden
 
AR2

AR2

Master Burner
Joined 2002
burningamp.com
4real said:
I measured the original output stage, and it does not contain any shelving... It is ruler flat.

Anyway, my new plan is to replace the complet DAC and output section with a new one :) I'm going to use the new ESS sabre DAC for it. I even have two channels left that I could use for bypassing the DSP to make a headphone output or something else... Should be a fun project. But for now, I'll have to wait until the boards are ready, that might take a while. Also I don't think this is going to fit into the original casing ;)
Click to expand...


I have been following that thread too, and I am excited about it as well. That takes me again at the point where I really haven't seen much mods where the digital signal is taken out of DCX. I am still unclear how well that could be done. I am assuming that digital transmitters needs to be inserted in the first place in order to send this signals out of the box?
 
Moby

Moby

Member
Joined 2003
Now I'm totally confused...I remember that somebody measured stock DCX2496 and claimed that it's pretty flat. Right now I can't find that article or thread:confused: If that person measured correctly I wonder what "burned" stock "presence":confused:Anyway, burned or not I have a question about input of AK AD chips. Does it have to be terminated with 332-332-332ohm (around 500ohm)?
 
Moby

Moby

Member
Joined 2003
Thanks, I checked. So I don't see any reason for using something like this. This is much more sound vice better circuit. Also, it can be arranged for about 10k input impedance. Symmetrical or fully differential. What do you think about?
5393-input.jpg
 
You must log in or register to reply here.
Top Bottom