There is a misunderstanding here on power consumption and the "LF50" supply
Have a look at the schematics / circuits and you will find the answer why it is no problem… The LF50 is not supplying any way near 1A - It "only" supplies the logic chips on the mainboard itself and it supplies the 5 volt for the Waveio isolation chip.
The DAC boards itself, with their own regulation, are indeed drawing somewhat over 250mA each. But that comes straight from the 12 Volt supply
Ah I overlooked the VB knot, thanks for clarification 🙂
Now I can change the regulator with good conscious ^^
Now I can change the regulator with good conscious ^^
Do you introduce the Ian Fifo board with the modded clocks pairing with the raspberry pi or after the waveio and straight to the DDDAC board via i2s.
DAC outputs combined
Sorry, don't mean to hijack this thread but I'm in the process of building the DDDAC but maybe changing the i/v stage. One thing I read on a thread was a discussion with the designers of the PCM1794A and they were specifically asked about using the chip in mono mode, and the way the +/- outputs are conected. They did not recommend the +/- outputs of each side of the mono dac being connected directly together due to "intrernal architecture reasons". I assume that real world tests have shown no degradation in sound quality by directly combing these outputs at the i/v resistors in this design??
Sorry, don't mean to hijack this thread but I'm in the process of building the DDDAC but maybe changing the i/v stage. One thing I read on a thread was a discussion with the designers of the PCM1794A and they were specifically asked about using the chip in mono mode, and the way the +/- outputs are conected. They did not recommend the +/- outputs of each side of the mono dac being connected directly together due to "intrernal architecture reasons". I assume that real world tests have shown no degradation in sound quality by directly combing these outputs at the i/v resistors in this design??
I will! I'll also report back on the effects of swapping the PSU of the dddac, as I think this is my dynamic and low frequency "problem" culprit (it's made for 1,1A peak and about 900mA continues power, so its pretty much working on its edge right now)
the + and - from the chip are not connected together… may be I do not unterstand the real question?
I have been thinking a lot about the reason why my dddac is better when it comes down to the lower frequencies and the slam when big drums is in action. A good example is Nick Cave and the song «Red right hand» (titletrack from «peaky blinders»). With this song, it is easy to hear that dddac stands out compared to the others.
I have come to the conclusion that it is a combination of the Sowter transformers, which Doede have ordered especially for his dac, and a big, overkill psu with massiv filtering. That’s what make the magic for me.
Yup I think you're right: my sowters are already in place but PSU is not strong enough (already sent you a pm regarding your PSU design 😉 )
No pi for me too!! After my experience with Kali, I am cautious before making further investments into separate boards. I wish I could try before I buy. Let me see if someone near me has the FIFO board.
Has anyone ever experienced any jumps or improvements in sound quality when shifting from the Red Motherboard to the Blue DDDAC board, keeping all other parameters the same.
kartick, did you try the Kali with the red board? As I wrote before: search for the relevant threads and you'll understand that this is not going to work very well...
For the last time, Kali is no match for Ian’s fifo WITH Accusilicon clocks. Without theese clocks, the difference is not as big.
Hey, I did not mean trying Kali with Red board vs the blue one. I just meant shifting from the red board to the blue one on the dac. I was using the red board for the longest time and have just received the blue one. Before I put that in with the wave io (no kali), just wanted to gain from other's experiences if they saw a jump in the SQ with only these boards changed.
Hey, even I have posted elaborately in various posts about how I tried Kali with the Red motherboard. Anyway, I was not intending to ask Kali with Red board vs Kali with the blue one. Clarified that to Smooth Dancer as well.
Sorry, I did not explain myself well. In DDDac you have A Pos and B Neg directly connected together feeding an I/V resistor and A Neg and B Pos connected together feeding the second I/V resistor. The designers of PCM1794A did not recommend this...they recommend all outputs are fed into individual low impedance nodes and then summed at a later stage. They warn against direct connection. That is why I ask if there is a sound penalty to doing this.
Here are (some of) the questions to, and answers from, the designers:
1. In mono mode, how are the outputs paired? Is it L+ = R- and L- = R+, or the other way around? It's not quite clear from the datasheet, though the order of connections in Fig.26 would suggest so.
-> Yes, L+ =R- and L- = R+. But, we do not recommend to connect toghter directly each current output due to internal current source cofiguration.
2. What is the output impedance of the analog outputs, and can the corresponding ones in mono mode be added in the current domain by simply paralleling together, before the I/V? (I'm using an I/V with input impedance very close to zero.) -> Output impedance could not be specified so that output is current out, but, as noise gain analysis, output impedance would be equivalemt to around 100K-oohm.
Due to internal operation theory, as mentioned on section-1, directo connection for each current out is not recommended.
Here are (some of) the questions to, and answers from, the designers:
1. In mono mode, how are the outputs paired? Is it L+ = R- and L- = R+, or the other way around? It's not quite clear from the datasheet, though the order of connections in Fig.26 would suggest so.
-> Yes, L+ =R- and L- = R+. But, we do not recommend to connect toghter directly each current output due to internal current source cofiguration.
2. What is the output impedance of the analog outputs, and can the corresponding ones in mono mode be added in the current domain by simply paralleling together, before the I/V? (I'm using an I/V with input impedance very close to zero.) -> Output impedance could not be specified so that output is current out, but, as noise gain analysis, output impedance would be equivalemt to around 100K-oohm.
Due to internal operation theory, as mentioned on section-1, directo connection for each current out is not recommended.
Two Ian reclocker boards and the Accusilicon clocks are on the way to me. They will have to queue up behind some other projects, but I am looking forward to this very much!
Hi Theo,
If you are i2s only, there is no point using them. I am using a CD transport so they are a big benefit for me.