Problem with Behringer DCX2496

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
AX tech editor
Joined 2002
Paid Member
Hi all,

I've been working on my DCX2496 and it appears I broke something: When I connect the SP/DIF from the CD transport to the DCX digital inputs, the DCX sees no input signal. The CD transport is OK (I have another DCX and that works OK with the same transport).

I checked the waveform on the input of the SRC 8420 (pins 4 and 6) and that is identical with the input waveform, so it does get to the SRC.

I think I checked all settings on the DCX menus to make sure it has AES/EBU selected at channel A and B, not muted, etc, even went so far as to reset it to factory presets. Still no luck.

Is there anything I can check with a scope etc to find out what is wrong? I'm an analog type of guy so some help would be appreciated.

Jan Didden
 
AX tech editor
Joined 2002
Paid Member
fredsonqc said:
sorry,
set up, page one, down select input on A+B AES/EBU;
input C works olso for micro
FRED


Fred, sorry, you lost me. My digital input doesn't work. This is selected by selecting inputs A & B, then selecting AES/EBU. It is not possible to use input C for digital; the hardware doesn't support it.
So, I don't see the connection between my problem and the C-input.

jan Didden
 
dcx 2496 steer clear

I purchased a DCX 2496 mid Dec ..I had to search high and low to find one and eventually found one in Adelaide.Its prime purpose was to help me sort out the correct crossover points and slopes for some speakers which have been designed and built for mastering purposes.The speakers use high-end Focal Audiom components 2 x 12wx bass, 2 x 7nv2mids , 1 x tlr tweeter, all in their own cabs ...5 cabs per side.This is a no compromise system designed to be a reference monitoring system.
The DCX was bought soley as a piece of test equipment with the idea of substituting it with something better once the various points were sorted.
Well for 2 weeks of tearing my hair out trying to achieve some sort of resemblence to a decent sound my faith in the unit gave up.I've owned Behringer gear before and been quite happy with it. so the DCX with all its bells and whistles ,i thought would have been a reasonable purchase.
I had a 3way analog pioneer car x-over (cd 635) lying around which i put into the system ,with the aid of a car battery....and what you know ,the sound starts coming together.......so whats wrong with the DCX.Well to find out I ran test tone through it with its outputs running into a RCF spectrum analyser ,(an old unit but very accurate) What i found was that the bass response is all over the shop, and certain bass frequencies also push up mid-range frequencies on the low outputs.To explain further, we had a 180hz 12 db slope x-over point between the bass and mids . A 100hz tone peaking at 0db would be accompanied by a 315hz tone about 3db down .This happened on the low outputs regardless of output configeration. (LMH LMH or LLMMHH) .This was the most obvious fault and explained my original dilemma that achieving anything near a flat resonse is totally impossible.
So basically you get what you pay for...you cannot expect a unit of this price to give decent results.
And finally the car x-over which was a blessing considering my frustration is being replaced by a x-over from Marchand Electronics.
 
AX tech editor
Joined 2002
Paid Member
rfbrw said:
Are OLRCK and OSCLK present at the CS8420?

RFBRW,


OK, finally got around to measure it. On OSCLK (pin 16) I see what appears to be a 7Mhz wave, distorted square as if gone through a low pass filter, but that may also be my scope. This is with music playing and an active signal at the AES input (pins 4,5). The signal at 16 is steady, no modulation.
On pin 17 OLRCK I see an almost perfect 1Mhz square wave, also not modulated.

I have found out that the CS8420 runs in hardware mode 1 which is the default data flow with AES input mode. So OSCLK should be the serial output bit clock, and OLRCK should be the word clock for the output sample rate.

BUT, the plot thickens. MUTE (pin 14) is asserted! Now, according to the DCX schematic, MUTE is connected to pin 11, RERR, with a pulll-up to +5 to define the start-up option. (MUTE is called SDIN on the DCX schematic, but that is for software mode; in hardware mode 1 this pin is MUTE). Now RERR is asserted whenever there is a problem with the AES receiver (parity error, coding error, loss of PLL lock). Question: is MUTE asserted or is RERR asserted??

[Edit: MUTE is an input so it must be RERR that asserts MUTE]

Any ideas what I should check next?

Jan Didden
 
janneman said:


RFBRW,


OK, finally got around to measure it. On OSCLK (pin 16) I see what appears to be a 7Mhz wave, distorted square as if gone through a low pass filter, but that may also be my scope. This is with music playing and an active signal at the AES input (pins 4,5). The signal at 16 is steady, no modulation.
On pin 17 OLRCK I see an almost perfect 1Mhz square wave, also not modulated.

The information I have states the DCX has an internal sample rate of 96K. Therefore you should expect to see OLRCK at 96k and OSCLK at 64Fs (about 6MHz).


I have found out that the CS8420 runs in hardware mode 1 which is the default data flow with AES input mode. So OSCLK should be the serial output bit clock, and OLRCK should be the word clock for the output sample rate.

But is it in master or slave mode. If it is in slave mode, the wrong OLRCK would mean a fault where the clocks are generated. If it is in hardware mode, it could be related to a faulty AES3 section.


BUT, the plot thickens. MUTE (pin 14) is asserted! Now, according to the DCX schematic, MUTE is connected to pin 11, RERR, with a pulll-up to +5 to define the start-up option. (MUTE is called SDIN on the DCX schematic, but that is for software mode; in hardware mode 1 this pin is MUTE). Now RERR is asserted whenever there is a problem with the AES receiver (parity error, coding error, loss of PLL lock). Question: is MUTE asserted or is RERR asserted??

[Edit: MUTE is an input so it must be RERR that asserts MUTE]

Any ideas what I should check next?

Jan Didden

There isn't really much to check where the AES3 rx section is concerned. Either there is something wrong with the input signal or at the input pins, dry joint, grounded, pulled high or shorted or the chip is bad.
 
Ex-Moderator
Joined 2003
You're feeding S/PDIF (0.5Vpk-pk) into an AES3 input (5Vpk-pk) and you're wondering why it doesn't work? I have a DCX2496 and it only just works on S/PDIF. It's always on the verge of crashing and clearly needs to see data of a proper amplitude that it can get a grip on.
 
AX tech editor
Joined 2002
Paid Member
rfbrw said:
The information I have states the DCX has an internal sample rate of 96K. Therefore you should expect to see OLRCK at 96k and OSCLK at 64Fs (about 6MHz).
[snip]


rwbrw,

My bad. The OLRCK is indeed around 96kHz, about 10 divs on the scope at 1uS/div.

The OSCLK looks OK too, 6MHz, 7Mhz, it's all rock and roll to me ;)
Seriously, I think it is OK. Next I will take EC8010's suggestion to check input levels. I will try to look at the PLL filter pin, maybe that'll give me a clue.

I asked my dealer for a repair, he wants to take it on but warned that probably it is as expensive as a new one. Go figure.

Maybe I should first try to replace the 8420 before I write the unit off. The analog inputs are OK so the ADC, DSP, DAC's etc are OK.
Anybody wants to buy a DCX for analog-only :D ?

Jan Didden
 
AX tech editor
Joined 2002
Paid Member
rfbrw said:
You could try boosting the input with a SN75176 or something similar. There is also a TTL circuit based around the 74HCU04 and the 74HC86 you could try. Both options give a balanced output a lot higher than 0.5 volts


The input signal is around .7V pk-pk, which gets at the primary of the input xformer. The input of TXR and TXP is also .7V pk-pk, of course.
The xformer floats, and the TXN is just a few 10's of mV wrt ground (and also at around 100mV DC wrt ground).
Is this normal? I'll check the recommended level for the 8420. Are you guys saying this input circuit is a klutz, really??

Jan Didden
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.