It was never discussed Greg I did mention it a couple of times on the other thread but no one seemed interested
I might have some notes but very very little as its not my thing I usually post findings as reference instead
The only ideas that grew legs was the 1/2 clock mod for the old red mainbiards and ccs
Anyway you have a soekris DAC to mod yet
If its best sound your after I'd prefer my version of the soekris to my version of the dddac 1794 its down to naturality and realness/neutrality or maybe just the sigma delta thing
If its just compulsive soldering disorder this is very rewarding project that will match good 5k+ dacs
LOL, Nige!
I think compulsive soldering disorder might be the correct diagnosis for me!
I actually have 2 Soekris to mod... eventually I expect them to end up almost unrecognizably modified and in a dual-mono balanced setup. But at this time, I'm also working on some 'experiments' for also-ran (as the Soekris best everything else in my systems so far).
I'm planning to apply the PCM1794 single-board mods as described herein to a PCM1795-based Sony HAP Z1-ES player I've already abused mightily. I intend to keep it around as a baseline source and as something to put into the system when everything else is on the bench (and I've already modified it beyond any hope of resale anyway). But without a sacrificial PCM1795 setup to test the 8v rail for the analog side of the DAC, I'm a little cautious about going over the 6.5v max specified in the datasheet, even tho after comparing the datasheets, the two chips do appear to be VERY similar in their inner-workings. So any hints on how to juggle the voltages against the current so I don't kill the beast would be so very useful... while I COULD replace the DAC chips if I killed them, it would be a serious pain and I'd like to avoid that.
TIA!
Greg in Mississippi
P.S. If you want to take this to email, if you don't still have mine, check the SDTrans thread on TirNa HiFi, I included my email in one of my responses to Rick McI there.
Congratulations!
Enrico,
Congratulations on your latest DDDAC 8 board design which looks very professional and should compete with the SQ of the very best DACs in the world at any price. I can't image a better design for an 8 board DDDAC.
I have been contemplating a DDDAC design of my own with compactness being one of my main criteria. Of course all my proven tweaks would be built into the boards. My dream system is direct digital stream to a Najda DSP crossover/equalizer for a triamp speaker system. The Najda would output to three single board DDDACs followed by a Tortuga Audio 6-channel LDR volume control and then 3 stereo amps. This allows the Najda DSP to operate at full resolution regardless of volume.
By the way, my DDDAC has been repaired from the power supply test mishap and is sounding great. I have been designing my own PCBs for various projects such as a 7193-6BX7 tube Aikido for my Moskido amp, an AD815 gain stage for a Tuner, and buffer output boards for the Tortuga Audio LDR volume controls.
Enrico,
Congratulations on your latest DDDAC 8 board design which looks very professional and should compete with the SQ of the very best DACs in the world at any price. I can't image a better design for an 8 board DDDAC.
I have been contemplating a DDDAC design of my own with compactness being one of my main criteria. Of course all my proven tweaks would be built into the boards. My dream system is direct digital stream to a Najda DSP crossover/equalizer for a triamp speaker system. The Najda would output to three single board DDDACs followed by a Tortuga Audio 6-channel LDR volume control and then 3 stereo amps. This allows the Najda DSP to operate at full resolution regardless of volume.
By the way, my DDDAC has been repaired from the power supply test mishap and is sounding great. I have been designing my own PCBs for various projects such as a 7193-6BX7 tube Aikido for my Moskido amp, an AD815 gain stage for a Tuner, and buffer output boards for the Tortuga Audio LDR volume controls.
Ciao Pierre,
Thanks for your kind words... I am really glad to know that our work make a great improvement in your DAC and I hope the same results also for the others that participate to our adventure 🙂
Any pictures about your LDOs solution?
All the Best,
Enrico
Ciao Ross, long time since I heard you 🙂
The congratulation are also for you and Frankie, at least for the first design of the DAC board with buffer that was also the starter point to go further and design the "BUS" DAC. Maybe without your support I would never have started it all
Glad to know that you are now able to design your boards... this in my opinion is another example of the improvement that our Team work give to all of us
Take care,
Enrico
TPS7A4700
Nothing fancy, really. I Just put a board from TekDevice to replace the LF33.
Reading your post about LDO, I meant that expensive solutions are not necessarily the way to go.
Nothing fancy, really. I Just put a board from TekDevice to replace the LF33.
Reading your post about LDO, I meant that expensive solutions are not necessarily the way to go.
Attachments
Ross
Question
Finally getting round to looking at signal buffers I see a lot of obsolete parts been mentioned varieties of 2sk's etc and the bf862 which seems widely used I see it was noted in this https://www.google.ie/url?sa=t&sour...MM_9CL_Qy_g5Rp6lA&sig2=Lty1rUXuOuQqCKu8MjX6RA
For its low input capacitance at 13pf
The 2sk208 we used in the CCS has a input capacitance of 8.2pf
Which I think is one if the reasons it worked good there
And we know its low noise
I was wondering if there was any reason it could not be used and if so was it part if your buffer circuit testing
I know I've a bunch of 208r's but is there low idss make them unusable?
I suppose I could get 208gr's
Cheers
Nige
Question
Finally getting round to looking at signal buffers I see a lot of obsolete parts been mentioned varieties of 2sk's etc and the bf862 which seems widely used I see it was noted in this https://www.google.ie/url?sa=t&sour...MM_9CL_Qy_g5Rp6lA&sig2=Lty1rUXuOuQqCKu8MjX6RA
For its low input capacitance at 13pf
The 2sk208 we used in the CCS has a input capacitance of 8.2pf
Which I think is one if the reasons it worked good there
And we know its low noise
I was wondering if there was any reason it could not be used and if so was it part if your buffer circuit testing
I know I've a bunch of 208r's but is there low idss make them unusable?
I suppose I could get 208gr's
Cheers
Nige
2sk208 as Buffer JFET
Nige2000,
I never tested the 2sk208 as a buffer JFET. There is virtually nothing about using this JFET in audio circuits. It was developed in 1981 and has a much lower Ids than the BF862.
The 2sk246 showed up in an obscure article about zero tempco and the 2sk208 was found to be a substitute component. Its our dumb luck that the 2sk208 turned out to have a near zero tempco at 0.4ma and 2.4Vds while sounding better than any other JFET tried in the pin 20 CCS.
From experience I can tell you that both the BF862 and 2SK208 vary widely in Idss from the same lot. They must be tested and matched to get the best results. I strongly recommend that you stay with matched BF862s for buffer use.
Nige2000,
I never tested the 2sk208 as a buffer JFET. There is virtually nothing about using this JFET in audio circuits. It was developed in 1981 and has a much lower Ids than the BF862.
The 2sk246 showed up in an obscure article about zero tempco and the 2sk208 was found to be a substitute component. Its our dumb luck that the 2sk208 turned out to have a near zero tempco at 0.4ma and 2.4Vds while sounding better than any other JFET tried in the pin 20 CCS.
From experience I can tell you that both the BF862 and 2SK208 vary widely in Idss from the same lot. They must be tested and matched to get the best results. I strongly recommend that you stay with matched BF862s for buffer use.
What is better, this
0.8uV Ultralow noise DAC power supply regulator +-9/12/15V 1.5A*x2 - DIYINHK
or this
DDDAC Power Supply - Audio Creative Shop
0.8uV Ultralow noise DAC power supply regulator +-9/12/15V 1.5A*x2 - DIYINHK
or this
DDDAC Power Supply - Audio Creative Shop
I use a complete different approach:
Separate PSUs for each voltage.
Audiophiler Spannungsregler, einstellbar, Voltage Regulator
Followed by many parallel capacities:
Netzteil-Siebung
NO REGULATION on board! Instead some additional capacities (3*470µF VCC and 2 * 470µF VDD for each PCM1794) close to the chips.
This kind of PSU made a very big step in sq!
Separate PSUs for each voltage.
Audiophiler Spannungsregler, einstellbar, Voltage Regulator
Followed by many parallel capacities:
Netzteil-Siebung
NO REGULATION on board! Instead some additional capacities (3*470µF VCC and 2 * 470µF VDD for each PCM1794) close to the chips.
This kind of PSU made a very big step in sq!
Dear Friends
Can someone condense me all the upgrades possible to the original DDDAC 1794 (4 boards fist version) that i can do to it.
Since there is the 2016 new version CCS etc can someone explain me what to do in order for me to do the upgrades to this wonderful sounding DAC, were can i get and what the parts to, Etc.
Sorry to bug you all.
Regards
João
Can someone condense me all the upgrades possible to the original DDDAC 1794 (4 boards fist version) that i can do to it.
Since there is the 2016 new version CCS etc can someone explain me what to do in order for me to do the upgrades to this wonderful sounding DAC, were can i get and what the parts to, Etc.
Sorry to bug you all.
Regards
João
I have 3 DDDAC to upgrade.
1st- is the red , first version with 4 boards with swoter output transformerd
2nd- is a red mainboard 2 boards, cap output
3rd- is a blue board (digital delay/spdif) 4 boards output
From what i´ve read n the first page i will probably need :
-2-clock delay circuit
-CCS for 10 dac boards
-Digital Vreg for 10 boards
-For the 2 old motherboards dont know whot can i do more?!?
Are there any instructions on how to install? (So many posts that i cant filter the visual noise)
All the rest i think its not as relevant but please feel free to advise.
Regards
João
1st- is the red , first version with 4 boards with swoter output transformerd
2nd- is a red mainboard 2 boards, cap output
3rd- is a blue board (digital delay/spdif) 4 boards output
From what i´ve read n the first page i will probably need :
-2-clock delay circuit
-CCS for 10 dac boards
-Digital Vreg for 10 boards
-For the 2 old motherboards dont know whot can i do more?!?
Are there any instructions on how to install? (So many posts that i cant filter the visual noise)
All the rest i think its not as relevant but please feel free to advise.
Regards
João
Hi João,
Actually on pages 1 and 2 of this thread you'll find a great deal of information. It's a good idea to read all the posts by Carlsor and emyeuoi to get you started.
Jan
Actually on pages 1 and 2 of this thread you'll find a great deal of information. It's a good idea to read all the posts by Carlsor and emyeuoi to get you started.
Jan
only getting back to this now
the 2sk208/2sk879 is a closer/identical match to the 2sk30atm which is mentioned plenty online in audio circuits
the 20 2sk208's i got varied very little in idss i could have used any of them
ive tested 2sk880,bf862,MMBF4416 and the 2sk208's in a pass b1 type config as a basic method to test individual jfets for sound quality
to my surprise there wasn't a vast amount of variance in the sound signatures of these jfets i had expected more
my order of preference was the 208's, mmbf4416, 2sk880, then the bf862
the 208's are sweeter sounding than 4416 which has better speed over the 880's that have better clarity, decay and reverb over the bf862 which is good anyway in what normal people would call a hair splitting exercise
so all is recommended as part of my usual nitpicking
Thanks for the summary, very informative. I have couple of questions: Is Doede using 208's in his latest DAC and how would you describe the sound difference between the resistor and the CCS?
Much appreciated,
Nick
Last edited:
2sk880,bf862,MMBF4416 and the 2sk208 Tests
Can you please provide circuit specifics so I can duplicate your test?
What version of the 2SK208 did you use?
I find that the BF862 is fast, detailed, and neutral and use it in all JFET applications except the PCM1794 pin 20 CCS. Just like not using the same brand/type of resistors and capacitors throughout a system there is probably a sq advantage to using different JFET types.
Can you please provide circuit specifics so I can duplicate your test?
What version of the 2SK208 did you use?
I find that the BF862 is fast, detailed, and neutral and use it in all JFET applications except the PCM1794 pin 20 CCS. Just like not using the same brand/type of resistors and capacitors throughout a system there is probably a sq advantage to using different JFET types.
No not part of deodes new boards
Ross here has posted a set of mods for the original dddac
Including a buffer he has designed
I was interested in building a buffer to ease the impedance difference so i was asking for advice before i started experimenting since i had done the first experiments i thought id better report the results