Hi Richard,
It seems that I have lost the left channel of my DAC. Turned it on the other day and noticed the left channel was out, the right channel plays perfectly. I ran through the voltages on the test points, this is what I got. All but TP6 are spot on.
PSU @ 24 vdc
TP1 : 19.33
TP2 : 10.09
TP3 : 20.58
TP4 : 2.5
TP5 : 4.94
TP6 : 1.764 seems a bit low from your target
TP7 : 10.1
TP8 : 10.1
TP9 : 10.1
TP10 : 10.1
Can you pinpoint the probable culprit from this data ?
Thanks,
Paul
It seems that I have lost the left channel of my DAC. Turned it on the other day and noticed the left channel was out, the right channel plays perfectly. I ran through the voltages on the test points, this is what I got. All but TP6 are spot on.
PSU @ 24 vdc
TP1 : 19.33
TP2 : 10.09
TP3 : 20.58
TP4 : 2.5
TP5 : 4.94
TP6 : 1.764 seems a bit low from your target
TP7 : 10.1
TP8 : 10.1
TP9 : 10.1
TP10 : 10.1
Can you pinpoint the probable culprit from this data ?
Thanks,
Paul
Paul,
I seem to recall I made a running change to the MOSFETs to lower threshold ones which meant a reduction in TP6 voltage. So that value you're showing is totally fine. If it were significantly wrong then you'd get nothing (or distortion) from both channels not just one.
I think worth checking continuity of your filter from input to output. One other customer found a potential dry joint on an inductor so I'd look into this as a first step. All your TPs aren't showing anything wrong so far.
To check if its the filter - unplug it and rotate it 180degrees before re-plugging it. If the working channel changes from right to left, its definitely the filter.
To check continuity of the filter measure with an ohmmeter between an outermost pin of the inner row of pins to an outermost pin of the outside row of pins and check both channels. Both outermost pins are wired together. If you get an open circuit rather than ~12ohm then that'll likely be an open circuit on an inductor tag for that channel.
I seem to recall I made a running change to the MOSFETs to lower threshold ones which meant a reduction in TP6 voltage. So that value you're showing is totally fine. If it were significantly wrong then you'd get nothing (or distortion) from both channels not just one.
I think worth checking continuity of your filter from input to output. One other customer found a potential dry joint on an inductor so I'd look into this as a first step. All your TPs aren't showing anything wrong so far.
To check if its the filter - unplug it and rotate it 180degrees before re-plugging it. If the working channel changes from right to left, its definitely the filter.
To check continuity of the filter measure with an ohmmeter between an outermost pin of the inner row of pins to an outermost pin of the outside row of pins and check both channels. Both outermost pins are wired together. If you get an open circuit rather than ~12ohm then that'll likely be an open circuit on an inductor tag for that channel.
Hi Richard,
I checked continuity as you suggested between the inner and outer pins on each side of the filter board, got continuity on each side, ~ 11.4 R. I'll try reversing the filter board on the DAC to see if anything changes, but I'll need to reassemble things in the case before I can run the test. I'll report back when I've done it.
Paul
Hi Richard,
I finally had the chance to reassemble the dac to test the filter board. I tested it twice with the filter board in its original orientation and again with the orientation reversed. In both cases only the right channel worked with the occasional pop coming from the left channel. So it appears that the filter board is good and that some element (opamp?) in the left channel has failed. Any ideas on a potential candidate, I do have a spare board to harvest parts from if needed.
Paul
I finally had the chance to reassemble the dac to test the filter board. I tested it twice with the filter board in its original orientation and again with the orientation reversed. In both cases only the right channel worked with the occasional pop coming from the left channel. So it appears that the filter board is good and that some element (opamp?) in the left channel has failed. Any ideas on a potential candidate, I do have a spare board to harvest parts from if needed.
Paul
Hi Paul,
So its definitely not the filter. The reason I don't think its an opamp failure is your TP7 voltage is fine at 10.1V - that seems only possible by opamp action. But there could be some mysterious opamp failure mode that I haven't come across before. So let's start at the top....
The entry point of the analog chain is Q4 MOSFET whose source receives current from the DAC chips pin6s. So check the pin6 voltage (any DAC chip will do as they're all paralleled). This should be about 1V. If that's OK then Q4 drain is the next point of interest - this is probably best checked on the filter input pin - outermost pin on the outside row on left side. The voltage there should be 10.1V. We know the filter's fine so this means TP11 should also have 10.1V on it. R40 (either side) should be 1.2V lower than that, so say 8.9V.
Assuming all those check out fine then perhaps we'll need to do AC tests, if you have the ability to generate a test waveform? I use Audacity for this on my PC.
So its definitely not the filter. The reason I don't think its an opamp failure is your TP7 voltage is fine at 10.1V - that seems only possible by opamp action. But there could be some mysterious opamp failure mode that I haven't come across before. So let's start at the top....
The entry point of the analog chain is Q4 MOSFET whose source receives current from the DAC chips pin6s. So check the pin6 voltage (any DAC chip will do as they're all paralleled). This should be about 1V. If that's OK then Q4 drain is the next point of interest - this is probably best checked on the filter input pin - outermost pin on the outside row on left side. The voltage there should be 10.1V. We know the filter's fine so this means TP11 should also have 10.1V on it. R40 (either side) should be 1.2V lower than that, so say 8.9V.
Assuming all those check out fine then perhaps we'll need to do AC tests, if you have the ability to generate a test waveform? I use Audacity for this on my PC.
Hi Richard,
All I have is a multimeter so I'll try doing as you suggest and see what find, I'll also take a hard look at my wiring from the board to the RCA out to make sure there isn't some failure there.
Paul
Hi Richard,
Finally got around to the next step, summertime chores keeping me busy. Anyway, I double checked all of the output wiring from the board to the RCA out jacks, no issues there. Checked Dac pin voltages on both sides for comparison
Pin 5 : R = 4.92v L = 4.92v
Pin 6 : R =1.09v L = 1.09v
Pin 7 : R = 0.82v L = 0.82v
Pin 8 : R = 1.1v L = 1.1v
Voltages on the outermost pin on outer row of filter installed : R = 10.17v L = 10.04v
R40 = 8.92v
TP11 = 10.1v
So basically everything seems good. Could a fault with my I2S card or I2S wiring cause only one channel to drop out ? I'm using a JLsounds usb to I2S that's pretty new, but the jumper wires seem pretty flimsy. Also do you want to take this issue offline instead of clogging up this thread.
Paul
Finally got around to the next step, summertime chores keeping me busy. Anyway, I double checked all of the output wiring from the board to the RCA out jacks, no issues there. Checked Dac pin voltages on both sides for comparison
Pin 5 : R = 4.92v L = 4.92v
Pin 6 : R =1.09v L = 1.09v
Pin 7 : R = 0.82v L = 0.82v
Pin 8 : R = 1.1v L = 1.1v
Voltages on the outermost pin on outer row of filter installed : R = 10.17v L = 10.04v
R40 = 8.92v
TP11 = 10.1v
So basically everything seems good. Could a fault with my I2S card or I2S wiring cause only one channel to drop out ? I'm using a JLsounds usb to I2S that's pretty new, but the jumper wires seem pretty flimsy. Also do you want to take this issue offline instead of clogging up this thread.
Paul
Paul - good work.
Those measurements on the DAC pins are textbook, pretty much exactly what I'd hope to see on a working DAC. The other voltages look fine too, R40 right on the nail. So nothing so far is showing up outside expected parameters. Which is very odd.
As to an I2S wiring/card fault being responsible for loss of only one channel, I can't think of a mechanism. Seeing as I2S is a stereo conduit, there's no known way a fault could turn off one channel completely without at least affecting the other.
By all means let's take to email or PM.
Richard
Those measurements on the DAC pins are textbook, pretty much exactly what I'd hope to see on a working DAC. The other voltages look fine too, R40 right on the nail. So nothing so far is showing up outside expected parameters. Which is very odd.
As to an I2S wiring/card fault being responsible for loss of only one channel, I can't think of a mechanism. Seeing as I2S is a stereo conduit, there's no known way a fault could turn off one channel completely without at least affecting the other.
By all means let's take to email or PM.
Richard
Received my Celibidache package 2 weeks ago. Thank you @abraxalito and wifey, the packaging is really neat!
Hoping to finish putting up my Celi soon!
Hi Abraxalito,
are there still Celibidache NOS DACs for sale?
I would like to buy one; but I am not able to write you a PM...
Best Regards
Fabian
are there still Celibidache NOS DACs for sale?
I would like to buy one; but I am not able to write you a PM...
Best Regards
Fabian
Hi Everyone,
some time back I began to enquire about Abraxalito's DACs as I wanted to build a music streamer, and settled on his Celibidache design. I finally have everything together and begun my listening and it certainly is living up to my expectations and the reviews of others on this forum. I spent a lot time and angst trying to get everything to work, with a lot of help from Abraxalito and guidance from the Volumio forum. So for anyone interested in building a low cost high quality music streamer here is what I used:
Rock64 4G revision 3.0 SBC from Pine Systems - the revision 3 board will not boot from a SD card with the latest and final build of Volumio by gkkpch: https://updates.volumio.org/rock64/volumio/3.212/Volumio-3.212-2022-01-28-rock64.zip so you need to use Pine Systems USB to eMMC adapter to flash the image. I used Rufus to flash the eMMC module (in my case a 8GB version).
For the power supply I used a
LT1083 output 3.3-9V 10-30v Regulated dc power supply board:linear/adjustable/HIF from aliexpress. You may want to use a better one.
Transformer was a
115V 230V 30W R-Core Shielded Transformer 12V+12V 9V+9V for Audio Amplifier Preamps AMP DAC R Core
Celibidache DAC. For the power supply I used Abraxalito's 24V module.
The SBC is connected to the DAC via the SBC 22 pin GPIO connector:
SBC Pin3 > DAC LRCK
SBC Pin4 > DAC BCK
SBC Pin5 > DAC DATA
SBC pin7 > DAC GND
The SBC is connected to my local network by Ethernet and I can control Volumio either via a browser or the App.
Hope this helps anyone wanting to build one.
Regards
some time back I began to enquire about Abraxalito's DACs as I wanted to build a music streamer, and settled on his Celibidache design. I finally have everything together and begun my listening and it certainly is living up to my expectations and the reviews of others on this forum. I spent a lot time and angst trying to get everything to work, with a lot of help from Abraxalito and guidance from the Volumio forum. So for anyone interested in building a low cost high quality music streamer here is what I used:
Rock64 4G revision 3.0 SBC from Pine Systems - the revision 3 board will not boot from a SD card with the latest and final build of Volumio by gkkpch: https://updates.volumio.org/rock64/volumio/3.212/Volumio-3.212-2022-01-28-rock64.zip so you need to use Pine Systems USB to eMMC adapter to flash the image. I used Rufus to flash the eMMC module (in my case a 8GB version).
For the power supply I used a
LT1083 output 3.3-9V 10-30v Regulated dc power supply board:linear/adjustable/HIF from aliexpress. You may want to use a better one.
Transformer was a
115V 230V 30W R-Core Shielded Transformer 12V+12V 9V+9V for Audio Amplifier Preamps AMP DAC R Core
Celibidache DAC. For the power supply I used Abraxalito's 24V module.
The SBC is connected to the DAC via the SBC 22 pin GPIO connector:
SBC Pin3 > DAC LRCK
SBC Pin4 > DAC BCK
SBC Pin5 > DAC DATA
SBC pin7 > DAC GND
The SBC is connected to my local network by Ethernet and I can control Volumio either via a browser or the App.
Hope this helps anyone wanting to build one.
Regards
@abraxalito I'm also interested in buying a kit but have new member restrictions and can't pm. Can you pm me?
Hey, any news on the balanced version? Could buying one more set ending up with balanced setup?
I have some PCBs for the balanced version if you'd like to order a pair built up. To go balanced you also need the I2S splitter.
Yes you can use the I2S splitter with a pair of normal Celibidache to run them balanced. The advantage of the balanced Celibidache is its using 'overclocked' DAC chips and passive I/V so will have a bit lower noisefloor.
Yes you can use the I2S splitter with a pair of normal Celibidache to run them balanced. The advantage of the balanced Celibidache is its using 'overclocked' DAC chips and passive I/V so will have a bit lower noisefloor.
I have my own designed I2S splitter board. PM me if you are interested. I posted about it here : https://www.diyaudio.com/community/...-rbcd-multibit-dac-design.324933/post-7032740
I bought the very first Celibdache a few years ago, but due to a very difficult couple of years, I only fully assembled the DAC now. I promised some, impressions so here they are.
System for evaluation:
Allo Digione->Bluejeans Coax->DAC->Bottlehead Crack w/ Speedball->HD600
BHC Tubes Tung Sol 7236/Mullard CV4003
VS The Schiit Modi 3
Not even a competition, the Celibdache delivers better performance across the board. The Schitt Modi 3 sounds like an incoherent fatiguing mess in comparison. I even tried to upgrade PSU for the modi 3 to elevate it’s performance, several different Apple ipad/iphone chargers, but the delta was just far too large.
VS The Micromega MyDAC
The Micromega suffers from a little noise and fatigue in the upper midrange/lower treble whereas the Celibdache comes off as much smoother, but still detailed there. Bass is better on the Celibdache. Overall win for the Celibdache, but it’s closer.
VS The Dragonfly Red - Bonus not really apples to apples
The Dragonfly Red suffers from USB noise and cannot really work with a really revealing system on PC, a Mac mini in this case. The Dragonly Red does manage better micro dynamics with a very noisy background. Would probably be a better matchup out of an iPhone, but I feel it would be a bit apples vs oranges. Glaring weakness of the Celibdache does apper to be microdynamics and conveying “emotion”.
Now, how does it stack up to my Denafrips Ares 2 in my main headphone rig?
System is:
Pi2AES->Bluejeans coax/Bluejeans toslink->DAC->Flux FA-12->Hifiman Anandas
This is a bit of unfair competition as the Denafrips Ares 2 is running balanced to the Flux FA-12, and that doubles it’s output power. With that said, the Celibdache is a much more aggressive sounding DAC. It sounds a little dynamically compressed compared to the Ares 2, but some might like that sound. This compression was also confirmed on the BHC tube amp. The Ares 2 is a much more polite sounding DAC that is also more delicate in its presentation. They are similar in detail and the both sound like r2r dac’s without the chip DAC harshness or hardness.
If I were to choose just one DAC, it would be the Ares 2 as it can be used for hours upon hours without fatigue, but the Celibdache’s aggressiveness, it can get grating over time. The Celibdache has a very strong bass presence, but the Ares 2 has a nuanced more textured bass presentation. The Ares 2 has better treble extension and detail. Mids are similar on both, but one is aggressive and the other relaxed. Microdynamics is where the Ares 2 jumps ahead. For example, around 1:30 of Father and Son by Cat Stevens, emotion is better portrayed on the Ares 2 and you can really feel it gushing out of the singer.
Personal preference and system matching would likely be key here and for the price I spent, about $425 CAD all in, it does an admirable job. Unfortunately, it only gets second headphone rig status. They way I’m describing the Celibdache kind of reminds me of the Bifrost 2, but without one on hand to compare I don’t want to say much more.
I recently acquired a very nice Marantz 1250 vintage AIO I need to refurbish, so I might try this DAC out again with it and my LRS speakers. This DAC was a bad pairing for the Akitika Z4 since the Z4 didn’t let the bass from the Celibdache shine.
System for evaluation:
Allo Digione->Bluejeans Coax->DAC->Bottlehead Crack w/ Speedball->HD600
BHC Tubes Tung Sol 7236/Mullard CV4003
VS The Schiit Modi 3
Not even a competition, the Celibdache delivers better performance across the board. The Schitt Modi 3 sounds like an incoherent fatiguing mess in comparison. I even tried to upgrade PSU for the modi 3 to elevate it’s performance, several different Apple ipad/iphone chargers, but the delta was just far too large.
VS The Micromega MyDAC
The Micromega suffers from a little noise and fatigue in the upper midrange/lower treble whereas the Celibdache comes off as much smoother, but still detailed there. Bass is better on the Celibdache. Overall win for the Celibdache, but it’s closer.
VS The Dragonfly Red - Bonus not really apples to apples
The Dragonfly Red suffers from USB noise and cannot really work with a really revealing system on PC, a Mac mini in this case. The Dragonly Red does manage better micro dynamics with a very noisy background. Would probably be a better matchup out of an iPhone, but I feel it would be a bit apples vs oranges. Glaring weakness of the Celibdache does apper to be microdynamics and conveying “emotion”.
Now, how does it stack up to my Denafrips Ares 2 in my main headphone rig?
System is:
Pi2AES->Bluejeans coax/Bluejeans toslink->DAC->Flux FA-12->Hifiman Anandas
This is a bit of unfair competition as the Denafrips Ares 2 is running balanced to the Flux FA-12, and that doubles it’s output power. With that said, the Celibdache is a much more aggressive sounding DAC. It sounds a little dynamically compressed compared to the Ares 2, but some might like that sound. This compression was also confirmed on the BHC tube amp. The Ares 2 is a much more polite sounding DAC that is also more delicate in its presentation. They are similar in detail and the both sound like r2r dac’s without the chip DAC harshness or hardness.
If I were to choose just one DAC, it would be the Ares 2 as it can be used for hours upon hours without fatigue, but the Celibdache’s aggressiveness, it can get grating over time. The Celibdache has a very strong bass presence, but the Ares 2 has a nuanced more textured bass presentation. The Ares 2 has better treble extension and detail. Mids are similar on both, but one is aggressive and the other relaxed. Microdynamics is where the Ares 2 jumps ahead. For example, around 1:30 of Father and Son by Cat Stevens, emotion is better portrayed on the Ares 2 and you can really feel it gushing out of the singer.
Personal preference and system matching would likely be key here and for the price I spent, about $425 CAD all in, it does an admirable job. Unfortunately, it only gets second headphone rig status. They way I’m describing the Celibdache kind of reminds me of the Bifrost 2, but without one on hand to compare I don’t want to say much more.
I recently acquired a very nice Marantz 1250 vintage AIO I need to refurbish, so I might try this DAC out again with it and my LRS speakers. This DAC was a bad pairing for the Akitika Z4 since the Z4 didn’t let the bass from the Celibdache shine.
Last edited by a moderator: