@shoom - Hi...changing to continuous clocks didn't make any difference, but changing to Volumio did - I can play everything now, including DSD. But without being able to A/B test Volumio and Ropieee, I feel as though Volumio doesn't quite have the soundstage that Ropieee did - can anyone offer any info regarding whether there is likely to be any audible difference between them? Is there a 'preferred' OS for audio reproduction?
@Steviewunda Some of us hear the difference between various software running on Rpi, so you are not alone 🙂
https://www.diyaudio.com/community/...fight-the-jitter.192465/page-546#post-7542506
https://www.diyaudio.com/community/...fight-the-jitter.192465/page-565#post-7571014
https://www.diyaudio.com/community/...fight-the-jitter.192465/page-546#post-7542506
https://www.diyaudio.com/community/...fight-the-jitter.192465/page-565#post-7571014
Would like to have opinions / recommendations on supplying power to my streamer.
Currently I supply all 5V via StationPi SMT J3, J9 jumpers 1-2 and 3-4 shorted from LinearPi II and UcConditioner fed from a 5-24 VDC variable output powerbrick at 7 VDC.
3,3V is supplied to J5 of FifoPi Q7II from UcPure III with Eaton 2,7V supercaps fed from a 19V notebook powerbrick.
People have shared a lot of their builds here and general tendency for StationPi builds seems to feed left and right side 5V from separate power supplies supported by separate transformers. Left and right side connecting jumpers open.
Would like to go for AC, Toroidy transformers I guess. According to Ian's manuals UcPure requires 12 VAC, LinearPi with UcConditioner 6 VAC. Toroidy's standard secondary voltages start from 10V though they also do custom transformers per order.
Should I go for separate 5V supplies?
If yes, should these be supported by individual single secondary winding transformers or could I equally have one dual secondary windings one?
Thanks in advance!
- #1D FifoPi Q7II
- #20C TransportPi AES
- #19D ReceiverPi DDC
- #35C StationPi SMT
- #49B MonitorPi Pro
- #17B HdmiPi Pro II
- Amanero Combo384
- RPi 3b+
Currently I supply all 5V via StationPi SMT J3, J9 jumpers 1-2 and 3-4 shorted from LinearPi II and UcConditioner fed from a 5-24 VDC variable output powerbrick at 7 VDC.
3,3V is supplied to J5 of FifoPi Q7II from UcPure III with Eaton 2,7V supercaps fed from a 19V notebook powerbrick.
People have shared a lot of their builds here and general tendency for StationPi builds seems to feed left and right side 5V from separate power supplies supported by separate transformers. Left and right side connecting jumpers open.
Would like to go for AC, Toroidy transformers I guess. According to Ian's manuals UcPure requires 12 VAC, LinearPi with UcConditioner 6 VAC. Toroidy's standard secondary voltages start from 10V though they also do custom transformers per order.
Should I go for separate 5V supplies?
If yes, should these be supported by individual single secondary winding transformers or could I equally have one dual secondary windings one?
Thanks in advance!
Hi Steviewunda
The problem with DNLA/UPNP protocols is that they are so loose which allows software manufacturers to produce slightly different variants which means that they don't always play nice together.
I played with a few OS for UPNP with Jplay IOS App and tried Volumio, Ropieee and Moode.
With Jplay IOS Volumio playback was hard to control and disconnected often and sounded a little warm to my ears. strangely enough Vo;umio Tidal connect sounded a little bright and thin (go figure).
Ropieee sounded better IMHO and was rock solid control wise.
Moode to my ears was the best of the bunch with Jplay falling somewhere between Ropieee and Volumio tonally and again rock solid.
https://www.bing.com/ck/a?!&&p=3476...dio&u=a1aHR0cHM6Ly9tb29kZWF1ZGlvLm9yZy8&ntb=1
Do read the getting started page as you will need to set up a user name and password before first boot.
For burning images I find raspberry pi OS great for Volumio and Moode
It may be a good idea to have a few cards handy and try for yourself as suggested in one of the links provided by space jack.
Yes to my ears software can sound different.
Give them a test drive and let us have your thoughts.
Last edited:
@wlowes +1 on making sure the contacts of clocks are making a firm contact in the socket, I affirm the procedure exactly as you described
@shoom Sure, I'll try to explain some fundamental principles as I understand it at this moment.
Essentially, every material has some intrinsic resonance. We can hear the different signature (or frequency) of any material by knocking on them and clearly each material "sounds" different. Now comes the funny part - by some (to me unknown) principle (perhaps in the realm of quantum physics), the resonance of each of our component is directly translating into the final sound. I had no problem accepting it in an amp, then by some stretch of imagination in the DAC, but then I got baffled when I discovered, how much it also affect the "sound" of the digital devices (streamer, network switches/routers, modems, etc...). Some of the change could be attributed to the widely accepted principle that clocks inside of these devices induce voltage when moving, so stabilizing them should help them in this regard, BUT this theory (at least subjectively) can't explain the extent, how much the treating of external power supplies affects the sound, where there are no clocks obviously.
I have my own hypothesis, what could be going on, but to go to more detail requires more time/space (perhaps I'll finally establish my own channel dedicated to all of my discoveries, in all other fields as well).
Now the important - practical part for us is how we can affect the resonance of our components. First and obvious possibility is the selection of the case (only applicable here for DIY community 🙂), where I would recommend to stick to use wooden bases as many of here already do. Perhaps try different types of wood. The dogma that it needs to be metal in each "case" is not holding much ground, when looking at DACs from Lessloss, which are clearly working quite well and selling for ridiculous $ (but as you can check, from Reference/Supreme versions, not many were ever sold on second-hand market). I'm not saying shielding isn't important, but I'll always prioritize resonance control (shielding, OR imo even better - noise cancelling could be done by many other means).
Second possibility is tied to another principle I found to be universally applicable: resonance of the whole device is determined by the material on which its own weight stand. And then the resonance of that material is also influenced by the underlying material. And so it continues to the centre of the gravity (let that sink in).
So this is how I'm building the foundation for every digital, non-vibrating component (meaning without own transformer coil, which I consider actually as a vibrating component and for that using a different approach)
1. wooden desk/shelve (just whatever you're now using)
2. decoupling material - some visco-elactic or gelly material with minimum intrinsic resonance and effectively absorbing whatever is happening below it
3. component base providing stable/hard support - again with minimum resonance/maximum stability - I'm using ordinary polished granite blocks, but also some thick hard wood should work just fine
4. pads or cones - the most important part is the material(s) of the smallest contact point (where the contact pressure is the largest) - I call that "primary resonance point". For cones it's the top, for pads it's the ball vs housing. Best is to try in your system and choose according to your own preference.
The thing to remember is that you can never really get rid of the resonances (at least until the gravity works), the trick is to find & match the resonances (thorough our chain) that fit our preferences and components. So we're always going to induce some resonances. For example rubbery feet get us soft resonances, which are translated into "safe" tonality with weighty bass, but smeared focus/imaging and somehow rolled-off top (there are many types of rubber, so now just generalizing). Then if you put your component onto balls/cones made of metal/glass/ceramic/wood/crystal, each will have a different sound tuning.
@miklau Hey Miklau, yeah it took a ridiculous amount of testing and still is 😆 But fortunately I'm stubborn and dedicated enough to endure the pain of constantly repeating the same songs over & over again (sometimes 20-30x in one day), to be able to draw some conclusions and with large enough set of data to establish some hypotheses, which are then again being tested multiple times in a span of multiple days/weeks/months. A lengthy process indeed in all areas (power supply/filtering, cabling, networking, software, grounding, crystals, etc...), but I'm getting there.
Hard to tell for me now, where does it matter the most, generally I recommend starting treating any domain upstream from the transducers, so AMP -> DAC -> streamer -> last switch, etc... And with speaker setup, obviously start with speakers (I'm a head-fier).
@shoom Sure, I'll try to explain some fundamental principles as I understand it at this moment.
Essentially, every material has some intrinsic resonance. We can hear the different signature (or frequency) of any material by knocking on them and clearly each material "sounds" different. Now comes the funny part - by some (to me unknown) principle (perhaps in the realm of quantum physics), the resonance of each of our component is directly translating into the final sound. I had no problem accepting it in an amp, then by some stretch of imagination in the DAC, but then I got baffled when I discovered, how much it also affect the "sound" of the digital devices (streamer, network switches/routers, modems, etc...). Some of the change could be attributed to the widely accepted principle that clocks inside of these devices induce voltage when moving, so stabilizing them should help them in this regard, BUT this theory (at least subjectively) can't explain the extent, how much the treating of external power supplies affects the sound, where there are no clocks obviously.
I have my own hypothesis, what could be going on, but to go to more detail requires more time/space (perhaps I'll finally establish my own channel dedicated to all of my discoveries, in all other fields as well).
Now the important - practical part for us is how we can affect the resonance of our components. First and obvious possibility is the selection of the case (only applicable here for DIY community 🙂), where I would recommend to stick to use wooden bases as many of here already do. Perhaps try different types of wood. The dogma that it needs to be metal in each "case" is not holding much ground, when looking at DACs from Lessloss, which are clearly working quite well and selling for ridiculous $ (but as you can check, from Reference/Supreme versions, not many were ever sold on second-hand market). I'm not saying shielding isn't important, but I'll always prioritize resonance control (shielding, OR imo even better - noise cancelling could be done by many other means).
Second possibility is tied to another principle I found to be universally applicable: resonance of the whole device is determined by the material on which its own weight stand. And then the resonance of that material is also influenced by the underlying material. And so it continues to the centre of the gravity (let that sink in).
So this is how I'm building the foundation for every digital, non-vibrating component (meaning without own transformer coil, which I consider actually as a vibrating component and for that using a different approach)
1. wooden desk/shelve (just whatever you're now using)
2. decoupling material - some visco-elactic or gelly material with minimum intrinsic resonance and effectively absorbing whatever is happening below it
3. component base providing stable/hard support - again with minimum resonance/maximum stability - I'm using ordinary polished granite blocks, but also some thick hard wood should work just fine
4. pads or cones - the most important part is the material(s) of the smallest contact point (where the contact pressure is the largest) - I call that "primary resonance point". For cones it's the top, for pads it's the ball vs housing. Best is to try in your system and choose according to your own preference.
The thing to remember is that you can never really get rid of the resonances (at least until the gravity works), the trick is to find & match the resonances (thorough our chain) that fit our preferences and components. So we're always going to induce some resonances. For example rubbery feet get us soft resonances, which are translated into "safe" tonality with weighty bass, but smeared focus/imaging and somehow rolled-off top (there are many types of rubber, so now just generalizing). Then if you put your component onto balls/cones made of metal/glass/ceramic/wood/crystal, each will have a different sound tuning.
@miklau Hey Miklau, yeah it took a ridiculous amount of testing and still is 😆 But fortunately I'm stubborn and dedicated enough to endure the pain of constantly repeating the same songs over & over again (sometimes 20-30x in one day), to be able to draw some conclusions and with large enough set of data to establish some hypotheses, which are then again being tested multiple times in a span of multiple days/weeks/months. A lengthy process indeed in all areas (power supply/filtering, cabling, networking, software, grounding, crystals, etc...), but I'm getting there.
Hard to tell for me now, where does it matter the most, generally I recommend starting treating any domain upstream from the transducers, so AMP -> DAC -> streamer -> last switch, etc... And with speaker setup, obviously start with speakers (I'm a head-fier).
Wow - interesting read and a complex subject
Some would state having the big cables attached (which I needed of course) negates all this anti vibration and isolation and the big cables can put that back in?
Some would state having the big cables attached (which I needed of course) negates all this anti vibration and isolation and the big cables can put that back in?
If you use usb audiolinux is BY FAR the best os hands down have a nice day. By FAR. If you need gpio use gentoplayer while very good not quite audiolinux level.
The rest are absolutely trash compared to these two.
I have everything in my setup isolated with sorbothane, the only thing to use imo. I disagree with quite a bit of experience that dac should be isolated first, then streamer and anything digital... Then move to amp or preamp. Even under my crossovers in my DIY GR RESEARCH nx otica made a tremendous difference.
The best is constrained layer dampening with layers of grainte and sorbothane sandwiches. Incredible. Can use grainte tiles as well.
Attachments
@badd99 Agree. Dampen everything. If you have to choose only one component, I'd dampen the crystal of the clock first. Silicon, sand, blue tac. To me it is akin to a buying a clock with lower phase noise. Some things you clearly could not pick out in isolation, but it all adds up IMHO. I just get in the habit. For me, the basic structure is wood. I tac stuff down with silicon. And if there is some empty space on a pcb, I toss in a bag of sand for good measure. 👍
So in a effort to understand power supplies
I found this
https://a.co/d/dcFUvL9
It has 2x5v, 2x3.3v, 2x12v,
and 2xadjustable.
Would one simply add a single 32v toroidal transformer to this?
You could then feed up to six Uc Connditioners for 3.3v and 5v
I found this
https://a.co/d/dcFUvL9
It has 2x5v, 2x3.3v, 2x12v,
and 2xadjustable.
Would one simply add a single 32v toroidal transformer to this?
You could then feed up to six Uc Connditioners for 3.3v and 5v
Last edited:
Hello,
Sometimes optimal clock use is written about by other members .
You can have a perfect power supply but if the clock is not properly attached what will be the benefit. Closest resemblance to a clock would be an old fashioned cartridge in a tone arm? If you make it cost 500 $ it will be a very critical device so EVerything possible should be done to make it work as it should. The better you get your cartridge to follow the groove the better it will extract information. I guess once your clock is not properly connected it will function like a cartridge with a broken suspension.
The ufl connectors i mentioned it before if you read the technical information page given by the manufacturer you can read they ( i think the one connected to the cable) is not made to be connected and disconnected on a daily basis for a long time. WHY do you think it won't matter because we only use it for diy audio? Why then all the fuss about power supplies?
Greetings Eduard
Sometimes optimal clock use is written about by other members .
You can have a perfect power supply but if the clock is not properly attached what will be the benefit. Closest resemblance to a clock would be an old fashioned cartridge in a tone arm? If you make it cost 500 $ it will be a very critical device so EVerything possible should be done to make it work as it should. The better you get your cartridge to follow the groove the better it will extract information. I guess once your clock is not properly connected it will function like a cartridge with a broken suspension.
The ufl connectors i mentioned it before if you read the technical information page given by the manufacturer you can read they ( i think the one connected to the cable) is not made to be connected and disconnected on a daily basis for a long time. WHY do you think it won't matter because we only use it for diy audio? Why then all the fuss about power supplies?
Greetings Eduard
I think the best Damping is 1/2 a bottle of wine. Rather than damping the source components why just not dampen the destination components; and its cheaper
Hello,
I read there is a plan to introduce pints of wine into the UK. It will take some time to get it inside your country.
Greetings Eduard
Yes fair points.
I've used gentoo and audiolinux on PC but they where still quite young then. I must revisit and thanks for the reminder.
just waiting for the DAC to settle before starting the software merry-go-round.😉
Wladimir
A truly epic post which I'm going to read through when i have a little more free time.
Many thanks for taking the time to share.
Haaaa
I prefer to damp both self and source components just to be sure.
Hello,
Of course reading some books about music will also enhance your listening experience. It is like with literature if you acquire some knowledge about the time the book was written you will appreciate/ understand it much more. But the fables of La Fontaine are still as vivid when they were written 350 years ago. And hard to imagine the spirit of that time for some of us. Here we probably have members who don't even own more than ten books.
Greetings Eduard
Of course reading some books about music will also enhance your listening experience. It is like with literature if you acquire some knowledge about the time the book was written you will appreciate/ understand it much more. But the fables of La Fontaine are still as vivid when they were written 350 years ago. And hard to imagine the spirit of that time for some of us. Here we probably have members who don't even own more than ten books.
Greetings Eduard
