Well, not easy to describe sound, but I can try.
I have noticed that if there is any change after a tweak, I always hear that in the lower frequencies first, bass is more defined or tight. The same thing happens after this reg tweak, but this is not the only improvement. I can hear more into the recording with a lot of details, superb instrument separation and improved 3D. Every instrument is now easier to localize between the speakers. At last, but not least, realism, the feeling that you are there is spooky with live recordings.
Dddac is very good with stock regulators, but is improving quite a bit in my setup with the new regulators.
I own a big thank you to «ekstreme-boky» who first wrote about this tweak earlier in this tread.
I have noticed that if there is any change after a tweak, I always hear that in the lower frequencies first, bass is more defined or tight. The same thing happens after this reg tweak, but this is not the only improvement. I can hear more into the recording with a lot of details, superb instrument separation and improved 3D. Every instrument is now easier to localize between the speakers. At last, but not least, realism, the feeling that you are there is spooky with live recordings.
Dddac is very good with stock regulators, but is improving quite a bit in my setup with the new regulators.
I own a big thank you to «ekstreme-boky» who first wrote about this tweak earlier in this tread.
Hello,
There are some other contestants to use for the 5 volt wave io power supply.
The one used by by Smoothdancer is a bit tricky because you have to take care about the input voltyage which should not be to high because then it will fry when also delivering 385 mA for the wave io.
Smoothdancer is using a regulated supply ( i think part of the dddac kit) to lget the right input voltage for the LT3045 regulator which then feeds the wave io which incorporates several ldo regulators too.
I remember a Dutch member ( Stefan Supersurfer) who skipped the 7810 regulator. I dont wanna do it, afraid to fry the Tent shunts.
After getting my big choke i could try some other regulators than the Belleson.
greetings, Eduard
There are some other contestants to use for the 5 volt wave io power supply.
The one used by by Smoothdancer is a bit tricky because you have to take care about the input voltyage which should not be to high because then it will fry when also delivering 385 mA for the wave io.
Smoothdancer is using a regulated supply ( i think part of the dddac kit) to lget the right input voltage for the LT3045 regulator which then feeds the wave io which incorporates several ldo regulators too.
I remember a Dutch member ( Stefan Supersurfer) who skipped the 7810 regulator. I dont wanna do it, afraid to fry the Tent shunts.
After getting my big choke i could try some other regulators than the Belleson.
greetings, Eduard
Ok here's an odd one. I've implemented 4 Kali reclockers with Wave IO boards and DDDacs and all have run successfully with no issues. All have been implemented leaving the Kali floating with no earth connection and simply bck, lck and data connected.
I then removed the Kali board from my own DAC to use in another DAC and installed a new Kali board.
This board works perfectly with my MacBook Pro but refuses to work with my Mac Mini. The lock light doesn't come on on the Kali board and all I get is some crackling.
I've tried various earthing schemes others have used but nothing makes any difference.
Any ideas anyone?
I then removed the Kali board from my own DAC to use in another DAC and installed a new Kali board.
This board works perfectly with my MacBook Pro but refuses to work with my Mac Mini. The lock light doesn't come on on the Kali board and all I get is some crackling.
I've tried various earthing schemes others have used but nothing makes any difference.
Any ideas anyone?
I never use the 7810 regulators. SQ to my ears is best when the Tents are directly fed with a choke PSU with no pre regulation.
I fully understand tho that Doede wanted to protect the DAC boards from oversupply which would try them.
Hey Juancho, on a side note, does this mean that you first take the signal out of the WaveIO and feed it to Kali (which is separately powered and does not use any Rpi/other computer) and from Kali you take it to the DDDAC's motherboard? If this understanding is correct, can you suggest what kind of sound differences do you observe when using the Kali in the chain and when you take it out? It would be better if you could share a small hand drawn diagram of how you connect these components. I actually use an Allo isolator and Kali but use it with Rpi and not directly, this way I also remove WaveIO altogether from the chain. So am curious to learn using the Kali without an Rpi and with WaveIO. Thanks.
Hello ,
Doede simulated my present power supply with the new LL2771 choke. I am using transformer with several windings on the secundairy side which can be put in series or in parallel to get the closest to the optimum voltage.
The power supply gives a maximum dc voltage during switching on of about 25 volt before going down to the value i need to safely use the 7810. Guido Tent once told me that i could use a 5 watt zener diode to tackle that high voltage. If you choose the right zener it will only work when working as protection. If you use a 12 volt zener and the actual voltage is 10,5 volt you weill not have a negative side effect of the zener..
But for removing the 7810 you will have to disabble the tower??
Greetings, Eduard.
P.s i am going to listen to the effect of the bigger choke before removing the 7810
I also had issues where one particular Kali would work okay between the WaveIO and the DDDAC, but another one would not. In my case the issue was that I got crackles during playback (but still get music, just in a bad way). Not as extreme as in your case. All I could figure out was that there seem to be slight differences between Kalis of different batches / revisions. I gave up on the Kalis after a while. I now don't use the WaveIO anymore, as I moved to the RPi world. The HifiBerry I2S outputs sound just as good as my "good" Kali.
There should be a few posts scattered in this thread where I described exactly this. Most of my posts were about trying to figure out why things didn't work out as well with my second build as they did in my first build.
I am sure but for me to search for all that from various posts and then join the dots, if it is easy for you maybe you just post that easily once again for my benefit. Just draw a simple diagram and help me understand. I want to really try out the kali in conjunction with the waveio. Doable?
You basically change the I2S conncetion from this:
WaveIO ---> DDDAC
to this:
WaveIO ---> Kali ---> DDDAC
This post has some details of how I did things.
Note that the Kali introduces a slight audio delay. That's irrelevant when listening to music, but not fun when you're using the DDDAC for the audio of a movie or similar.
Thanks. I've been through all the posts and tried most of the alternative schemes where people earthed the Kali. But none of these help the situation where it works on MacBook and not on Mac Mini
I've attached a dia a friend did for Kali Wave DDDac connections. In my versions I've never used /needed to use the earth connections
Hi.
I have just build a 4 deck DDDAC are very pleased with the outcome. I am in the process of choosing an output option.
A nice capacitor sounds very good, but of course I would like to try transformer output.
I have a pair of http://www.lundahl.se/wp-content/uploads/datasheets/7903.pdf i can wire 1:2 and try just to get an idea.
Since there is some (20mV) DC left on the output, do you think I could harm the transformer by magnetizing the core?
Edit: I found out. Lundahl states that these transformers should be used with DC-free signal sources or with a blocking capacitor.
I have just build a 4 deck DDDAC are very pleased with the outcome. I am in the process of choosing an output option.
A nice capacitor sounds very good, but of course I would like to try transformer output.
I have a pair of http://www.lundahl.se/wp-content/uploads/datasheets/7903.pdf i can wire 1:2 and try just to get an idea.
Since there is some (20mV) DC left on the output, do you think I could harm the transformer by magnetizing the core?
Edit: I found out. Lundahl states that these transformers should be used with DC-free signal sources or with a blocking capacitor.
Last edited:
You should be able to null out that dc offset on the dddac. If you put a resistor in parallel with the load resistor with the highest dc output you can adjust the output voltage to match the other side. If for instance the highest voltage you have is 2.720volts output and 33 ohms load then the current is 2.720/33 = 82.424 ma. Given that the dac output is a current source then if we want an output of 2.70 volts (20mv lower) then we need a resistance of 2.7v/ 082424amps =32.76 ohms. A resistor of 4.45k across 33 ohms would give you approx. 32.76 ohms and should be close to giving you a dc balance.
Easy to do- if you have a 10k 10 turn trim pot, solder or clip lead it across the output resistor and adjust until you have nulled out the offset. Then measure the pot and replace with a fixed resistor or resistors of the same value.
I have never done this to my own dac but it seems like it should work. You can see that a pretty large resistor is required to null out that offset because it is a very small percentage of the approx 2.7volt dc output.
I had the same idea before, but I never really tried it. Yes, it should work -- but is the DC offset fixed, or will it drift a bit after a few weeks/months/years?
I have measured mine a few times and it was very consistent - within a milivolt. I have one dac board and the offset is less than 5mv on the worst channel.
Ok. I've done a bit more digging and my old MacBook outputs 44.1 24 bit. A newer MacBook and my Mac Mini output 32 bit which results in crackles EXCEPT when the Audio Midi is set to output 384 32bit when the music sounds great but is upsampled.
So, anyone know what's happening here?
David
Do the crackles just get moved out of the frequency band that you can hear by the upsampling? Does the crackling at lower sampling frequencies sound different at different sampling frequencies?
There's no music at the lower sampling frequencies. So the crackling goes, music appears!