Behringer DCX2496 digital X-over

Re: DCX2496. Tranformer out with Sowter 9545. A good idea??

Liliya said:

2.Take the signal directly and balanced from the DAC into a Sowter 9545 transformer (ratio 1+1:5+5). (http://www.sowter.co.uk/acatalog/SOWTER_OUTPUT_TRANSFORMERS_12.html#a282)

Why? I have very good experience with this transformer in my Copland 288 CD player. I take the signal directly from PCM 63. On the other side of 9545 the signal go to Lampizator, a SSRP stage with the exellent 6N6P. Sound is superband ver y much into my taste, (http://www.lampizator.eu/LAMPIZATOR/whytubes/whytubes.html)

Eivind Stillingen, NORWAY

I am using Lundahls after DAC and I am very happy with it. I am not familiar with your transformers, but I wouldn't go more than 1:2+2 if you can. In ideal case scenario 1+1:1+1
Other than that your idea to go for tubes sounds very good.

Keep us informed.
 
AX tech editor
Joined 2002
Paid Member
novec said:
Jan: Doesn't the ADC take half the bits from the positive input and half from the negative? Or does it just take the difference between them, even if the negative is grounded, so I'll get a full resolution digital even with a single ended input?[snip]


It does take the difference, but you cannot ground one side of the ADC as both ADC inputs need to ride on a 2.5VDC reference. I use the attached circuit. The pin 'OCM' is connected to 2.5VDC derived from the supply with an LM285-2.5 .

Jan Didden
 

Attachments

  • se2bal.jpg
    se2bal.jpg
    65.7 KB · Views: 975
Hi Jan,

This specialized OPA1632 opamp is a very good solution which simplifies the analog inputs.

Hi Novec,

I'm not an electronic professor. So far I prefer learning by knowledge and not by believing otherwise we all end up most probably in blackgate caps and have to use solid silver cables and spikes for our speakers!?
Perhaps you have another look at Jans website. I have a vage idea you are reinventing the wheel?

Frank
 
AX tech editor
Joined 2002
Paid Member
I should also mention that if you want to drive that C-channel ADC input directly, be aware that it is low impedance. The schematic for the DCX2496 shows series resistors of 332 ohms and one parallel across the ADC inputs of 332 ohms as well, and then again both (stereo) ADC's are in parallel (for channel C). So you need to be able to drive a minimum of around 500 ohms differential, 250 single ended. Plus you need gain to overcome that attenuation of the series and parallel resistors. You need to know what you're doing.

Frank, do you have any idea why they use those 3rd ADC channels inparallel instead of just one? Channel C is mono anyway. Is it to keep the digital part a bit simpler?

Jan Didden
 
Hi Jan,

I have seen this 4th analog input channel already in the past but hadn't spent too much time thinking about it because I'm not using channel C at all. I only have a vage idea. Perhaps the reason is it's easier to calculate the A+C and B+C sums if there is a right and a left channel for C as well. But why are there two HF filters? They could have simply connected the Ains directly at the ADC.
I realized the first time that the C channel uses a lower grade AK5383 compared to the AK5393 for channel A+B. I fear that's mentioned in the DCX feature list.

Frank
 
To find out it would be necessary to cut one wire of the C right channel so that both Ains remain connected via R9. Then you can find out via the PC remote software whether this channel is used at all. Perhaps there is a 4th unknown analog in?
I can't do this because I have disassembled the AK8353s on my DSP boards.
 
AX tech editor
Joined 2002
Paid Member
oettle said:
Hi Jan, [snip] Perhaps the reason is it's easier to calculate the A+C and B+C sums if there is a right and a left channel for C as well. [snip]Frank

Yes, that woud be a good reason. Hadn't thought about that. So, whenever the software selected either chan A or chan B, in both cases they had chan C available.

Jan Didden
 
The OPA1632 is a very good solution, but I'm looking to spend as little as possible, and use the LM4562s I've already got. And with my economy nowadays, Jan's mod is way out of my budget...

I'm no professor either, Frank, far from it :) But I'm more of a theoretics guy than a hands-on guy, I prefer designing it right the first time rather than experimenting with a lot of different things. That might very well mean that I end up paying for a lot of fancy stuff I won't hear, but at least I'll feel good about it :) And to me, designing is half the fun. I do, however, have very good personal experiences with spikes and high quality cables, so I guess I'm a little bit of both.

I've decided to finally get my rear in gear, so I'm going for your original input circuit with a pot instead of R8, and the full three-way power bypass caps, and will be sending off the board for production tomorrow. It's time to try this baby!
 
Compatibility with CD/DVD coaxial digital out

This a simple question many here can answer.

I got my DCX2496 working with analog inputs derive from an ext DAC. Everything works well.

I read some where that channel A can accept digital input like the AEU. Is this digital input compatible with the coaxial digital output from CD or DVD player? I tried it it did not work well.

I suspect I need another box between the DVD/CD and DCX 2496 for eg. SRC or DEQ 2496 for the arrangement to work. Please confirm.

Thanks...
 
Hi ttan98,

Standard CD/DVD players provide an S/PDIF output which should work with the DCX2496 digital input. Best would be to use a 120 ohm cable and to connect GND of both devices via shield. If there is some DC voltage coming from the player a serial ~47nF cap on one input line would be helpful.
If there are clicks and cracks DCX connector might be the reason.
Also there is sometimes a 'dull sound' problem caused by the CS8420 SRC.

Frank
 
oettle said:
Hi ttan98,

Standard CD/DVD players provide an S/PDIF output which should work with the DCX2496 digital input. Best would be to use a 120 ohm cable and to connect GND of both devices via shield. If there is some DC voltage coming from the player a serial ~47nF cap on one input line would be helpful.
If there are clicks and cracks DCX connector might be the reason.
Also there is sometimes a 'dull sound' problem caused by the CS8420 SRC.

Frank


thanks for response, I seem to be experiencing 2 problems:

1. some sort of high freq. distortion, intermittent
2. the input seems to be overloading, input to A & B input display shows clipping fairly often, how to correct this? is this because of this that causes item 1 above

any way to corrrect?
 
Hi ttan98,

The red LED triggers already at -3dB. So don't worry about level it's OK!

Reason for your distortion might be caused by switching supplies of both devices. Connect GND (or housing) of both devices.
Is your player capable of driving a 120 ohm load? If not you could replace R15 on DSP board.
Use short cable.

Frank
 
oettle said:
Is your player capable of driving a 120 ohm load? If not you could replace R15 on DSP board.
Does this mean that the XLR input automatically changes it's routing and it's input impedance to suit the signal at the XLR plug?

Is the Spdif (digital) output set to drive 75ohm cable and terminating impedance?
If so, then why should it have a problem driving 110ohm?
It might have a problem with seeing the wrong terminating impedance after the 75ohm coaxial cable connection.
What tolerance can Spdif accept on it's 75ohm output?

What effect will a standard XLR plug and socket have on the terminating impedance seen by the coaxial cable?
 
Hi Andrew,

If it's a standard S/PDIF output (75ohm) it should be able to drive the 110ohm (R15) input impedance of the DCX but S/PDIF is specified for 0.5V@75ohm and AES3 for 5.0V@110ohm. So not impedance but level might be the problem.

Using short cables (<1m) these mismatching impedances aren't important but level. So increasing value of R15 might help. That's different using a 300m cable which AES/EBU is specified for.

Frank