Hi,
Being fascinated with upcycling gear of yesteryear, I recently bought myself a Behringer DDM4000 mixer that has quite a nr of interesting features, especially considering its original as well as its current second hand price point.
The thing that has kept me busy since I bought it is that I keep reading that it is actually almost entirely a digital mixer and fx circuit, and with my inexperience eye I seem to be able to tell from its functionality (confirmed by its design schematic?) that it should have quite a nr of ADCs (4 stereo channels; Additional 2 mic channels) and at least 2 DACs (master/booth and headphones) on board.
The mixer has a single digital (coaxial spdif ) output that does not respond to any volume knob or panning, telling me it likely has a nearly direct digital output from its BlackFin fx processors.
Question 1: How is it that even after having disabled any signal processing on my HT amp, the spdif signal noticeably lags behind the analog outputs of this mixer? Is there something inherently slow about the spdif signal protocol? Lots of buffering going on?
This has me a little stumped.
Lead in to the next question(s):
My assumption is that purely design wise it would require fairly little effort to add switchable coax spdif inputs for each of the input channels, thus bypassing the ADCs.
Would this make sense, not just asking from an efficiency perspective but also sq wise? And is anyone whether perhaps some conversion from spdif to i2c or another digital signal is required to bypass the ADCs on some channels? Or is spdif signal itself already (typically) compatible with digital audio circuits?
All of this comes up mostly out of general curiosity (or rather, specific curiosity), no need for advice on buying dedicated audio interfaces or other mixers
Will update this post soon with some links and an image or two.
Cheers!
Being fascinated with upcycling gear of yesteryear, I recently bought myself a Behringer DDM4000 mixer that has quite a nr of interesting features, especially considering its original as well as its current second hand price point.
The thing that has kept me busy since I bought it is that I keep reading that it is actually almost entirely a digital mixer and fx circuit, and with my inexperience eye I seem to be able to tell from its functionality (confirmed by its design schematic?) that it should have quite a nr of ADCs (4 stereo channels; Additional 2 mic channels) and at least 2 DACs (master/booth and headphones) on board.
The mixer has a single digital (coaxial spdif ) output that does not respond to any volume knob or panning, telling me it likely has a nearly direct digital output from its BlackFin fx processors.
Question 1: How is it that even after having disabled any signal processing on my HT amp, the spdif signal noticeably lags behind the analog outputs of this mixer? Is there something inherently slow about the spdif signal protocol? Lots of buffering going on?
This has me a little stumped.
Lead in to the next question(s):
My assumption is that purely design wise it would require fairly little effort to add switchable coax spdif inputs for each of the input channels, thus bypassing the ADCs.
Would this make sense, not just asking from an efficiency perspective but also sq wise? And is anyone whether perhaps some conversion from spdif to i2c or another digital signal is required to bypass the ADCs on some channels? Or is spdif signal itself already (typically) compatible with digital audio circuits?
All of this comes up mostly out of general curiosity (or rather, specific curiosity), no need for advice on buying dedicated audio interfaces or other mixers
Will update this post soon with some links and an image or two.
Cheers!
Last edited:
This is the schematic, and on the Behringer DDM4000 product page you can get a nice impression of the product
I have plenty of hours spent on toying with soldering primitive synthdiy circuits, but will admit it is with limited understanding of the electrical behaviors of said circuits.
What I consider doable is finding a few solder points on the circuit bord and attaching some additional circuitry if there is a possibility bypassing the ADCs.
My guess is that feeding the mixer spdif over coax requires some conversion of said signal to the mixers internal digital signal, which i guess may be done with an additional ic (per channel) that performs just this task.
Am I close? Regardless of whether this is advisable?
What I consider doable is finding a few solder points on the circuit bord and attaching some additional circuitry if there is a possibility bypassing the ADCs.
My guess is that feeding the mixer spdif over coax requires some conversion of said signal to the mixers internal digital signal, which i guess may be done with an additional ic (per channel) that performs just this task.
Am I close? Regardless of whether this is advisable?
It somewhat depends on whether you want to replace the analogue I/O with digital I/O or add digital I/O to the analogue I/O though in the general scheme of things the difference in effort isn't that great. To replace or add to all the inputs would require at least 16 wires out of the mixer and 15 wires into the mixer. It would also require 5 SPDIF receivers and 5 ASRCs or 5 combined ASRC/SPDIF chips and, if adding to the I/O, a switch would also be required. And someone to layout the PCB. The attached image is for one stereo pair.
Attachments
Thanks for your input; I will admit though that I am not entirely clear on where there is overlap with your image and the mixers' existing internals.
Please correct me if I am wrong in my interpretation so far.
Let's say that I would sacrifice just the dedicated Line input on Channel 4 (the leftmost two RCA ports) by disconnecting that particular analog path into the CS4272 (maintaining the switchable Phono/Line input).
I would have to add an IC that takes raw SPDIF signal through the hijacked white RCA connector and converts it into whatever digital signal the CS4272 below can interpret.
Based on the image below (from the pdf attached to my earlier post) I take it the AINL/AINR+/- are the Line level analog inputs, I take it the AOUT(LR)+- are analog outputs of the processed signal. What confuses me is that there is already circuitry attached to the pins 3-7, which my guess is where I would have to insert a (preprocessed) digital audio signal.
Is it possible/probable this is used for PFL (prefader listening/cueing headphone monitoring)?
And if that were the case, would it still be possible to somehow also inject the SPDIF signal through those pins?
Thanks again
Please correct me if I am wrong in my interpretation so far.
Let's say that I would sacrifice just the dedicated Line input on Channel 4 (the leftmost two RCA ports) by disconnecting that particular analog path into the CS4272 (maintaining the switchable Phono/Line input).
I would have to add an IC that takes raw SPDIF signal through the hijacked white RCA connector and converts it into whatever digital signal the CS4272 below can interpret.
Based on the image below (from the pdf attached to my earlier post) I take it the AINL/AINR+/- are the Line level analog inputs, I take it the AOUT(LR)+- are analog outputs of the processed signal. What confuses me is that there is already circuitry attached to the pins 3-7, which my guess is where I would have to insert a (preprocessed) digital audio signal.
Is it possible/probable this is used for PFL (prefader listening/cueing headphone monitoring)?
And if that were the case, would it still be possible to somehow also inject the SPDIF signal through those pins?
Thanks again
The CS4272 is a codec, a combined ADC and DAC. Pins 3-7 form the shared digital audio interface. Pins 3-5 are the clock signals. SDOUT is the digitized output for balanced analogue inputs AINL and AINR and SDIN is the digital signal for AOUTL and AOUTR. The digitized analogue inputs route straight to the DSP chips and the outputs from the DSP chips route straight back to the dacs and SPDIF transmitter. The two DSP chips do all the audio processing.
When you turn an analogue input into a digital input without any option to switch back to analogue, you won't need the switch in rfbrw's schematic of post #6, but you still need the S/PDIF receiver and the asynchronous sample rate converter, as the incoming digital signal will not be synchronized with the internal clocks of the mixer. There are ICs that are both S/PDIF receiver and asynchronous sample rate converter, for example the Texas Instruments SRC4392. Most of them need a microcontroller to set all of the internal bit settings to whatever you need.
Reading up on the 8422, I get the impression that I would have to figure out the internally used bit depth and some other aspects of the signal (one of: i2s, tdm, left or right justified) in order to then have a certain pin through a certain value resistor connect to ground and have it send a chosen signal format to the 4272(s).
Is the 8422 agnostic about the incoming spdif signal, or does that require some additional tweaking as well? And would it make a difference in (reducing) signal latency if I happen to choose one hardware configuration over another?
Is the 8422 agnostic about the incoming spdif signal, or does that require some additional tweaking as well? And would it make a difference in (reducing) signal latency if I happen to choose one hardware configuration over another?
Nothing is being sent to the CS4272. All ADC side does is convert analogue audio into digital audio. If the SPDIF input is replacing the analogue input then the output of the ADC in the CS4272 is out of circuit replaced by the output of the ASRC. If the SPDIF input is an alternative to the analogue then the digital output of the CS4272 is switched out for that of the ASRC when the SPDIF input is selected.
The internals of the CS8422 don't matter. It for the the user to configure the input as required and configure the output as required. The CS4271, CS4272 and the CS4351 digital outputs use the left justified serial audio format so the CS8422 output will be set to that. The input is the SPDIF signal.
As to the latency you would have to compare the group delay figures, where given, for the various devices that meet your need.
The internals of the CS8422 don't matter. It for the the user to configure the input as required and configure the output as required. The CS4271, CS4272 and the CS4351 digital outputs use the left justified serial audio format so the CS8422 output will be set to that. The input is the SPDIF signal.
As to the latency you would have to compare the group delay figures, where given, for the various devices that meet your need.