Keep working on the power, as you have described, it should pay off and the parts are reasonable compared to, uh, other parts. A basic snubber circuit will offer more improvement than fancy diodes.
Then the benefits from a better op amp, filter improvements, and local power decoupling will made be more apparent.
I also had gone the diode route early on, then had learned that the cheaper snubbed arrangement was better, had tried to save you that same learning experience.
Then the benefits from a better op amp, filter improvements, and local power decoupling will made be more apparent.
I also had gone the diode route early on, then had learned that the cheaper snubbed arrangement was better, had tried to save you that same learning experience.
Other people have 'learned' the opposite. Means that one or a few limited experiences may not generalize as much as we may tend to expect. Humans naturally tend to generalize from very limited experience, which probably conferred some survival advantage to our ancestors. Doesn't mean the tendency is accurate in all cases, it only needs to be about 70% accurate to have a good chance to get incorporated into progeny.
Also, in some cases diodes and snubbers might be needed, although that is more like to occur with higher current, lower loss transformers. Depending on the energy to be dissipated and sensitivity of affected circuitry, it will vary from case to case.
Last edited:
My reference is my iPad ! It sounds good enough for me.
So, If this DAC can surpass it, awesome.
I plan to use these tracks (FLAC)
1. Bruce Springsteen - Magic
2. Engima - Eyes of Truth
3. Goo Goo Dolls - Better Days
4. Mika - Emily
I might move out sooner or later. So can't invest in gears like that. But after everything is settled, I'll surely be doing it.
Regarding the snubber, I just can't do it.
I just don't know ! I tried to learn but dropped the idea for now.
I have the op-amps covered ! I currently have AD811 & LME49720, a LT1361 is in the mail but the NE5534AP not shipped...yet.
Regarding the other things like filter improvements and power decoupling, can you please tell me more ?
Like I said, its not that I haven't listened to you. Its just beyond my ability right now. So, I went the easy way out, the soft recovery diodes !
I'll deal with snubbers after I learn more (electronics).
Would this work ?
http://projectpoint.in/index.php?route=product/product&product_id=501276980154
It takes some trial and error, and/or a lot of relevant searching, gathering of information from this site and some others to gain better understanding of how the different circuits will react to different changes relating to audio. I had suggested the most simple circuit that would add the most improvement for the effort, but it will certainly be better if you can figure out the why before blindly following others.
For the analog filter capacitors that are around the op amp, my first choices are the Wima FKP2, and the Siemens KS/EMZ polystyrene mixed such that their effects are combined.
As for the decoupling, there are better(lower noise)electrolytic capacitors available to use around the wm8740 chip, and the op amp. Some of the popular ones are nichicon kz, and elna Silmic.
The choke you linked to should work to lower the noise.
Before you commit to a case, it might be nice to have a preamp in there also, just something to consider, but another project definitely.
For the analog filter capacitors that are around the op amp, my first choices are the Wima FKP2, and the Siemens KS/EMZ polystyrene mixed such that their effects are combined.
As for the decoupling, there are better(lower noise)electrolytic capacitors available to use around the wm8740 chip, and the op amp. Some of the popular ones are nichicon kz, and elna Silmic.
The choke you linked to should work to lower the noise.
Before you commit to a case, it might be nice to have a preamp in there also, just something to consider, but another project definitely.
Last edited:
@Phase
Wima & Siemens caps mixed means connected in parallel ? Like if I want 47uF, can I use 24uF+23uF ? I can get the Wima but I haven't seen Siemens around here.
I think I can get Nichicon Muse as well as Elna Silmic easily.
Regarding pre-amp, my amp is an integrated amp (Yamaha AS-500). So, is it really needed ?
Just read the PCM2704 data sheet and it was underwhelming ! 98SNR ? Seriously ?
By the way how is this DAC working ?
USB>PCM2704>SPDIF>DIR9001>WM DAC Any idea ?
Wima & Siemens caps mixed means connected in parallel ? Like if I want 47uF, can I use 24uF+23uF ? I can get the Wima but I haven't seen Siemens around here.
I think I can get Nichicon Muse as well as Elna Silmic easily.
Regarding pre-amp, my amp is an integrated amp (Yamaha AS-500). So, is it really needed ?
Just read the PCM2704 data sheet and it was underwhelming ! 98SNR ? Seriously ?
By the way how is this DAC working ?
USB>PCM2704>SPDIF>DIR9001>WM DAC Any idea ?
120dB would suit a 20+ bit DAC for sure, but this one's just 16.
On various Taobao DACs I've come across, because they want to accommodate more than one kind of input (USB, S/PDIF) they usually convert the USB to S/PDIF then use the source switching of the S/PDIF receiver to select the input.
On various Taobao DACs I've come across, because they want to accommodate more than one kind of input (USB, S/PDIF) they usually convert the USB to S/PDIF then use the source switching of the S/PDIF receiver to select the input.
@abraxalito
I think so !
By the way, what voltage range should I go for the caps for the DIR9001 and WM8470 DAC ?
Is 24V ok ?
Or should I go for the 50V ?
Also, for electrolytic, is low ESR "audio" grade is better, right ?
I think so !
By the way, what voltage range should I go for the caps for the DIR9001 and WM8470 DAC ?
Is 24V ok ?
Or should I go for the 50V ?
Also, for electrolytic, is low ESR "audio" grade is better, right ?
Regarding caps, no point in going way above the working voltage - DIR9001 runs on 3.3V so a 4V cap would suffice. For WM8740 a 6.3V cap I believe would be appropriate. Usually lower ESR is better but that is based on the assumption that the layout's not been screwed up (which on Taobao DACs it normally has). No experience of 'audio' grade caps so can't comment sensibly.
Are you proposing a chip transplant ?
Are you sure the layout for 8804 match that of the 8740?
Ok !
Do I need an LCR meter for checking capacitance as most of the caps I find are +-10% ?
I wouldn’t sweat the pcm2706 at this point, there are other factors that will contribute to a decent sounding, rewarding outcome. A shorter, decent quality usb cable with a ferrite and careful routing away from other cables has helped me when using this type of usb receiver chip.
The analog filter caps are the ones in the pf range, I think one was 1,000, another 4,700 if I recall from looking at the picture you had posted. What I meant by mixing is to try and have a combination of different types of capacitors; namely polystyrene, and polypropylene film/foil in those locations. Some of those parts tend to be larger than some of the more commonly used ones, and it may be necessary to mount some from under the board. These have provided the most improvement when changed that I have noticed.
Abroxolito is more experienced with dacs of this nature, way more in fact, having developed several from scratch. A much higher voltage rated capacitor will have more inductance and is not a good idea for such a sensitive device.
On my wm8741 dac, I have used a 47uf on the mid-rail decoupling, sounded better than the data sheet example of 10uf, but your results may vary.
The analog filter caps are the ones in the pf range, I think one was 1,000, another 4,700 if I recall from looking at the picture you had posted. What I meant by mixing is to try and have a combination of different types of capacitors; namely polystyrene, and polypropylene film/foil in those locations. Some of those parts tend to be larger than some of the more commonly used ones, and it may be necessary to mount some from under the board. These have provided the most improvement when changed that I have noticed.
Abroxolito is more experienced with dacs of this nature, way more in fact, having developed several from scratch. A much higher voltage rated capacitor will have more inductance and is not a good idea for such a sensitive device.
On my wm8741 dac, I have used a 47uf on the mid-rail decoupling, sounded better than the data sheet example of 10uf, but your results may vary.
Soldering smd parts ?noooo, I just mentioned the best way to get very low jitter, 8804(spdif receiver could replace dir9001) even not similar to 8740
The issue with me is the availability of caps (polymer caps). All of them are above 100V
I need to a lot of googling left to do !
Ok, I'll check out the data sheet and find that cap out.
I often spend way more time researching and looking around at others experiences than you might imagine for a inexpensive part like a small capacitor for a specific location.
The polymer caps may lead to resonance if not quite small (22-47uf)in value if you are intending to use those. Only good on the digital supplies, and not close to a regulator either.
I would stick to Panasonic FM for digital supply probably, especially near the regulators, and leave the smd parts near the chips alone. You really need to look up the regulator part numbers and find out what the data sheet says about the esr of the output.
All of the polymer caps I have seen are rated at lower voltages, haven’t seen any that were anywhere near 100v.
The polymer caps may lead to resonance if not quite small (22-47uf)in value if you are intending to use those. Only good on the digital supplies, and not close to a regulator either.
I would stick to Panasonic FM for digital supply probably, especially near the regulators, and leave the smd parts near the chips alone. You really need to look up the regulator part numbers and find out what the data sheet says about the esr of the output.
All of the polymer caps I have seen are rated at lower voltages, haven’t seen any that were anywhere near 100v.
The regulators, not the diodes have a part number, that part number will lead to a data sheet that has valuable information regarding the performance, limitations, and suggested application.
Doing that, I found this;
Frequency compensation for the regulator is provided by capacitor Cout and its use is mandatory to ensure output stability. A minimum capacitance value of 4.7 mF with an equivalent series resistance (ESR) that is within the limits of 33 mW (typ) to 2.2 W is required. See Figures 12 and 13. The capacitor type can be ceramic, tantalum, or aluminum electrolytic as long as it meets the minimum capacitance value and ESR limits over the circuit’s entire operating temperature range. Higher values of output capacitance can be used to enhance loop stability and transient response with the additional benefit of reducing output noise.
The various capacitors will have ratings also, and can be used to establish a higher performing power supply.
Doing that, I found this;
Frequency compensation for the regulator is provided by capacitor Cout and its use is mandatory to ensure output stability. A minimum capacitance value of 4.7 mF with an equivalent series resistance (ESR) that is within the limits of 33 mW (typ) to 2.2 W is required. See Figures 12 and 13. The capacitor type can be ceramic, tantalum, or aluminum electrolytic as long as it meets the minimum capacitance value and ESR limits over the circuit’s entire operating temperature range. Higher values of output capacitance can be used to enhance loop stability and transient response with the additional benefit of reducing output noise.
The various capacitors will have ratings also, and can be used to establish a higher performing power supply.