Interesting !
What clock are you going to use and how are you going to reclock the signal coming from your USB chip ?
Use a low jitter clock with a proper supply.
I'd say, put the master clock next to the DAC chip, and :
- use the clock to reclock incoming data with a simple 74HC latch.
- buffer the clock and send it to the USB chip,
As for resistor I/V, it sucks, folded cascode is a lot better.
What clock are you going to use and how are you going to reclock the signal coming from your USB chip ?
Use a low jitter clock with a proper supply.
I'd say, put the master clock next to the DAC chip, and :
- use the clock to reclock incoming data with a simple 74HC latch.
- buffer the clock and send it to the USB chip,
As for resistor I/V, it sucks, folded cascode is a lot better.
goon-heaven said:
I tried TDA1545A, but others (notably, Rudolf Broertjes) report similar improvements with other chips.
Basically using a resistor IV is a compromise because you need a high resistor value to get useful signal voltage, but you also need a small resistor value because any voltage on the DAC output pin will distort. Remember these DACs are designed to drive a zero-ohm impedance, and can be traumatized by driving a high IV resistor. Most notably PCM63, with its output protection diodes, which clip at 0.6V and distort at much less...
The folded cascode is not concerned with this compromise so it can present a low impedance to the DAC and use a high value resistor to get a decent output voltage.
When listening, it just sounds vastly better : improved detail, dynamics, etc. There is no comparison...
Thanks Pierre-Frederic for the useful information. Sounds good sense for some DACs - I wonder whether there would be any mileage using folded cascode in my DDDAC1543? (using passive I/V showed superior dynamics (though less detail and refinement) in a comparision with my sons TDA1541 S2 DAC).
Steve
(RME soundcard on the shopping list)
Steve
(RME soundcard on the shopping list)
A word clock is a square signal at 44100 Hz (or whatever your sample rate is, 96k, 192k).
To generate it, use a dumb binary divider from the oscillator which you use for reclocking in your DAC.
If you don't have a stable clock for reclocking in your DAC, your sound probably sucks...
Or, you can also generate a bit-clock and a word-clock (using the same divider, very easy), feed them to an SPDIF-Encoder (very easy too) and voila ! you have a synchronized SPDIF signal.
Look at the datasheets, I'm not going to spoon-feed
To generate it, use a dumb binary divider from the oscillator which you use for reclocking in your DAC.
If you don't have a stable clock for reclocking in your DAC, your sound probably sucks...
Or, you can also generate a bit-clock and a word-clock (using the same divider, very easy), feed them to an SPDIF-Encoder (very easy too) and voila ! you have a synchronized SPDIF signal.
Look at the datasheets, I'm not going to spoon-feed
AL1402G to 4 stereo dacs
Same question :
Has anyone made or used a multichannel dac using the OPTOREC wavefront (alesis) AL1402G chip ?
I am very interested in buying a RME HDSP9652 for my linux/jack/brutefir convolving crossover PC so as to output 2 ADAT (8channel each) lines to a DIY mutiDAC device with OPTOREC inputs and many PCM1794A stereo DAC chips.
All dacs working at 44100hz 24 bits and master clock in DAC enclosure slaving the RME card.
The idea being that HDSP9652 has ALSA drivers and 3 ADAT outputs which offer galvanic isolation (from the PC) being optical.
Any comments, thanks, Eric
Same question :
Has anyone made or used a multichannel dac using the OPTOREC wavefront (alesis) AL1402G chip ?
I am very interested in buying a RME HDSP9652 for my linux/jack/brutefir convolving crossover PC so as to output 2 ADAT (8channel each) lines to a DIY mutiDAC device with OPTOREC inputs and many PCM1794A stereo DAC chips.
All dacs working at 44100hz 24 bits and master clock in DAC enclosure slaving the RME card.
The idea being that HDSP9652 has ALSA drivers and 3 ADAT outputs which offer galvanic isolation (from the PC) being optical.
Any comments, thanks, Eric
Thanks peufeu
Do you know if the pcm1794A is perfectly compatible with the AL1402G chip in terms of PCM format and timing ? The AL1402G not being I2S compatible, does its FMT mode00 or mode01 or mode11 suit the PCM1794 ?
http://focus.ti.com/lit/ds/symlink/pcm1794a.pdf
www.wavefrontsemi.com/DataSheetsFolder/WavefrontAL1402.pdf
Also does the clocking solution from the DAC side to the RMEhdsp9652 clock input break the ground isolation between RME soundcard and outside DAC ?
Does anybody see possible problems going this way ?
Do you know if the pcm1794A is perfectly compatible with the AL1402G chip in terms of PCM format and timing ? The AL1402G not being I2S compatible, does its FMT mode00 or mode01 or mode11 suit the PCM1794 ?
http://focus.ti.com/lit/ds/symlink/pcm1794a.pdf
www.wavefrontsemi.com/DataSheetsFolder/WavefrontAL1402.pdf
Also does the clocking solution from the DAC side to the RMEhdsp9652 clock input break the ground isolation between RME soundcard and outside DAC ?
Does anybody see possible problems going this way ?
> Do you know if the pcm1794A is perfectly compatible with the AL1402G chip in terms of PCM format and timing ?
No idea ! épluchage de datasheet...
> Also does the clocking solution from the DAC side to the RMEhdsp9652 clock
> input break the ground isolation between RME soundcard and outside DAC ?
- isolation transformer (ethernet, pulse, etc)
- fast opto
- or feed it an ADAT signal with no data, needs an ADAT encoder, add an ADC, then you can feed vinyl through brutefir
No idea ! épluchage de datasheet...
> Also does the clocking solution from the DAC side to the RMEhdsp9652 clock
> input break the ground isolation between RME soundcard and outside DAC ?
- isolation transformer (ethernet, pulse, etc)
- fast opto
- or feed it an ADAT signal with no data, needs an ADAT encoder, add an ADC, then you can feed vinyl through brutefir
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.