Nice DDDAC build! Regarding the 1:2 transformers, are you sure you need the step-up? Otherwise I'd suggest you also try balanced output with cap coupling instead of the trannies. I am running my DDDAC like this for quite a while now, and I am not looking back.The Sowters
For a really noob builder like me , can you please expand a little on this?Nice DDDAC build! Regarding the 1:2 transformers, are you sure you need the step-up? Otherwise I'd suggest you also try balanced output with cap coupling instead of the trannies. I am running my DDDAC like this for quite a while now, and I am not looking back.
I'm guessing you are talking out the DDAC board output here and use balanced instead of SE?I'd suggest you also try balanced output
How would one do this? since we are talking about balanced output, we will have 3 pin xlrs on left and right channel. where would a capacitor come into picture?output with cap coupling
Does this mean if we are using balanced output with 3pin xlr, then we don't need the transformers? Can we use balanced with transformers?instead of the trannies
See here:can you please expand a little on this?
An old question, with a new twist (I believe)...
I will use an active line stage (preamp) with balanced inputs. I will therefore use the balanced outputs from the DDDAC 1794 chips.
What is the better coupling option: capacitors or transformers? Why?
Caveats:
- So far all comparisons between caps and transformers were for SE output using caps and BAL output using transformers. It's not surprising that BAL sounded better due to the stronger drive and the partial cancellation of non-linearities.
- I am not (yet) asking for specific parts (caps or trannies). I am interested on the...
@mbrennwa what i understood from the post you linked is that to remove the DC offset for balanced output add 4 capacitors (like Wima MKP 4.7 μF) to the positive and negative outputs of the Dac. 2 caps per channel.
But this response for @dddac
also includes a resistor. Is the for SE case only?
But this response for @dddac
maybe this helps?
also includes a resistor. Is the for SE case only?
@NeoTheOne I did not use any resistors as shown in Doedes circuit for single-ended outputs. I don't think the resistors are needed with balanced signals, because they are not referenced to GND. The DC level of the balanced lines will be defined by the input circuit of the amplifier downstream of the DDDAC.
Last edited:
The resistor in the SE version drains off any DC charge from the capacitor. Without it and no cable plugged in, there will be couple of volts stored (from the DC offset from the output of the DAC) in the cap that will discharge when you plug a cable in causing a pop.
I use WIMA caps. Any no-nonsense quality part should work fine. I don't know the value from the top of my head (I am travelling). It needs to be large enough to not cut the low frequencies.
I sent you an email on this Brad....@mbrennwa What brand and value capacitors are you using? (instead of the transformers) Turns out one of my transformers is audibly distorting (I am guessing a cold solder joint inside) and I need to try this since I don't have any other immediate options.
would these output transformers work well with a 4 deck?
thanks in advance.
https://www.sowter.co.uk/specs/3232.htm
thanks in advance.
https://www.sowter.co.uk/specs/3232.htm
I am pretty sure they will work, but hard to say how well they sound and what the bandwidth would be. 10k:10k normally is used for input, but pure technically they will work as the input impedance is much lower than 10k and your load most likely is above 10k..... they seem to be laid out for higher voltage, so that is fine
Hi Doede, a few questions, I hope I don't bother you too much.... 🙂
I'm building a DDDAC streamer for my 2 kids with separate DAC boards with my own 100mA shuntregs for the analog 8,5 volt side. And made my own PCB's.
1. But now I got an idea to separate the GND's for the left side and for the right side. Even use a separate powersupply for left and right. So the DAC will be 'Dual Mono' for the analog side. Is this a good idea? Did you test or play with it? Any comments?
I think if you use 2 DDAC MK3 boards you can do the same....
2. Also the digital GND can be totally separated from the analog GND. And just on 1 point connect digital and analog ground with a 1 or 10 Ohm resistor. Because the datasheet says they have to be connected: 'TI recommends using the same ground between AGND and DGND to avoid any potential voltage difference between them.' Did you play with that? Any comments?
3. The datasheet says:
Power Supply Recommendations
The PCM1794A device requires 5-V (nominal) supplies. A 5-V supply is required for the analog circuitry poweredby the VCC1, VCC2L, and VCC2R pins. A second 5-V supply is for the digital circuitry powered by the VDD pin.These pins can be powered by the same 5-V rail but separating the supplies can assist with getting the targetSNR and THD in some cases. Place the decoupling capacitors for the power supplies close to the deviceterminals.
Can the PCM1794a chip run on 5V also on the digital side? I thought it always should be 3.3 Volt? Did you try that before? Any comments?
4. At last, did you ever measure the current going to the PCM1794a chip on the 3.3V digital side? The datasheet says:
When I measure it is around 2.5mA? Almost 10 times lower? Any idea?
I have a LF33 on the DAC board and feed it through a 10 Ohm resistor (R2) in serie of the LF33 and I only measure around 25mV.
As far as I know a LF33 is a series regulator so I in = I out.
And the DAC is playing lovely music, everything is OK.
Thanks in advance, Walter
I'm building a DDDAC streamer for my 2 kids with separate DAC boards with my own 100mA shuntregs for the analog 8,5 volt side. And made my own PCB's.
1. But now I got an idea to separate the GND's for the left side and for the right side. Even use a separate powersupply for left and right. So the DAC will be 'Dual Mono' for the analog side. Is this a good idea? Did you test or play with it? Any comments?
I think if you use 2 DDAC MK3 boards you can do the same....
2. Also the digital GND can be totally separated from the analog GND. And just on 1 point connect digital and analog ground with a 1 or 10 Ohm resistor. Because the datasheet says they have to be connected: 'TI recommends using the same ground between AGND and DGND to avoid any potential voltage difference between them.' Did you play with that? Any comments?
3. The datasheet says:
Power Supply Recommendations
The PCM1794A device requires 5-V (nominal) supplies. A 5-V supply is required for the analog circuitry poweredby the VCC1, VCC2L, and VCC2R pins. A second 5-V supply is for the digital circuitry powered by the VDD pin.These pins can be powered by the same 5-V rail but separating the supplies can assist with getting the targetSNR and THD in some cases. Place the decoupling capacitors for the power supplies close to the deviceterminals.
Can the PCM1794a chip run on 5V also on the digital side? I thought it always should be 3.3 Volt? Did you try that before? Any comments?
4. At last, did you ever measure the current going to the PCM1794a chip on the 3.3V digital side? The datasheet says:
When I measure it is around 2.5mA? Almost 10 times lower? Any idea?
I have a LF33 on the DAC board and feed it through a 10 Ohm resistor (R2) in serie of the LF33 and I only measure around 25mV.
As far as I know a LF33 is a series regulator so I in = I out.
And the DAC is playing lovely music, everything is OK.
Thanks in advance, Walter
@WalterW - that is a lot of questions, so let's look into my personal opinions and experiences
enjoy the DAC 🙂
Doede
- Double mono is always a good idea. I describe the use of two DDDACmk3 boards to be used in double mono on my blog site. The PCBs are already designed and set up to be used like this.
- Personally, I believe this splitting of grounds is an old and very bad idea. Good PCB design can avoid the issues TI talks about. Just make sure there is a rock-solid ground plane and look if all return current tracks are above the ground plane and are not crossed by other tracks. Difficult in 2 layer, easier in 4 layer PCB
- It seems you have the 2015 datasheet. It actually says that, but it is totally wrong. May be a typo. The rest of the datasheet is very clear: VDD is 3,3V
- Yes of course I measured the VDD supply current. it depends on the FS and the datasheet is correct. Are you sure your 10 Ohm is not 1 Ohm? that would explain it.
enjoy the DAC 🙂
Doede
Of course!
There was great discussion and success, you can read it here https://www.diyaudio.com/community/threads/upgraded-single-board-pcm1794-nos-dddac.265228/
There was great discussion and success, you can read it here https://www.diyaudio.com/community/threads/upgraded-single-board-pcm1794-nos-dddac.265228/
in the meantime a sowter 3232 has been built into it. works fine.I am pretty sure they will work, but hard to say how well they sound and what the bandwidth would be. 10k:10k normally is used for input, but pure technically they will work as the input impedance is much lower than 10k and your load most likely is above 10k..... they seem to be laid out for higher voltage, so that is fine
- Home
- Source & Line
- Digital Line Level
- A NOS 192/24 DAC with the PCM1794 (and WaveIO USB input)