i just assembled one board completely and wow ... first time i really hear the soundstage open up, i "kinda" thought its a myth lol because i never noticed the soundstage change with tweaks
the question now is wether it was the resistors or the supercap, i somehow believe its the supercap even tho my aune x8 shouldnt need it, maybe its one of the dacs that still uses usb power ....
if the resistors are the reason than this needs serious investigation... the upgrade from partial board to fully assembled board wasnt minor imo it feels like partial board did 50-60% of what the final board did
there is also the allo revolution dac, which is a interesting concept, it uses some supercaps to buffer power inside... i kinda think this is the way to go... each supercapacitor tweak i tried was a major improvement, even over linear power supplys
the ifi power X 5V had no trouble charging the supercap, i also calculated charge time it should be around 0-4 seconds, i think pretty much all power supplys can handle that, but larger caps might be really a concern here
tho now i have to try resistor and supercap upgrade seperately to be sure...
Also what i noticed now with 2 devices... the sound quality definitely gets better/smoother/more precise over time... around the first 1-2 hours, i thought it was quite noticable
the question now is wether it was the resistors or the supercap, i somehow believe its the supercap even tho my aune x8 shouldnt need it, maybe its one of the dacs that still uses usb power ....
if the resistors are the reason than this needs serious investigation... the upgrade from partial board to fully assembled board wasnt minor imo it feels like partial board did 50-60% of what the final board did
there is also the allo revolution dac, which is a interesting concept, it uses some supercaps to buffer power inside... i kinda think this is the way to go... each supercapacitor tweak i tried was a major improvement, even over linear power supplys
the ifi power X 5V had no trouble charging the supercap, i also calculated charge time it should be around 0-4 seconds, i think pretty much all power supplys can handle that, but larger caps might be really a concern here
tho now i have to try resistor and supercap upgrade seperately to be sure...
Also what i noticed now with 2 devices... the sound quality definitely gets better/smoother/more precise over time... around the first 1-2 hours, i thought it was quite noticable
more things i noticed with the fully assembled board:
IfI isilencer behind my ground filter actually sounds worse than putting it before my usb filter, i think because it negates the supercap, it probably has to sit closest to the dac..
Bass actually sounds less bloated with the usb filter, this could look as a negative or positive depending on your setup but the other things the ground filters does outweight the slight loss in sub bass i think... maybe the sub bass i was hearing was even culprit of something that is fixed now, like the shielding termination resistors (tho it couldnt be a ground loop since i have topping hs01 in the chain)
sounds seems overall "smoothed out" but dont think this means reduced dynamics, i guess people could still interpret it as such but imo it feels more like the background gets "so dark" in comparison that EFI/RFI loaded straight usb cables just make it seem like its harsher/"more dynamics" etc
high frequencys come actually cleaner, more precise trough, also because of the background thing i think
in comparison if we speak straight cable vs only topping HS01 vs IFI isilencer vs my filter
- straight wire is basicly unlistenable (IF) you heared once what better cable or tweak devices do... and i had multiple PC's, i think different PC's can sound slightly different because of a different EMI profile but overall one isnt "much" better than another, its better to filter afterwards in all cases i think
the only pc usb connection that was bearable unfiltered was the one from a raspberry PI, underclocked even better
- topping hs01... does a bit but its definitely in a region where i could say i just imagined the minor improvement, but yea i think there was one
- ifi isilencer, it was my first usb filter and i still think it does something nice on its own, definitely more than the topping HS01, but it also falls short after hearing my device and thats also why i think IFI recommends to use multiple isilencer for greater benefit...
(tho i also need to test this with my device, then i could also add 4 common mode chokes to data lines in the next revision if stacking two boards works fine...)
i can definitely say that with my device in the chain the background over usb was never darker before, but i also didnt try reclockers yet because they are rather expensive and i still wanna avoid spending 1-2k only on a usb connection...
if you know what EMI/RFI does, usually your usb connection is loaded with it and IME you need to remove majority of EMI first to even hear minor difference that seems to get otherwise masked by EMI it feels like....
IfI isilencer behind my ground filter actually sounds worse than putting it before my usb filter, i think because it negates the supercap, it probably has to sit closest to the dac..
Bass actually sounds less bloated with the usb filter, this could look as a negative or positive depending on your setup but the other things the ground filters does outweight the slight loss in sub bass i think... maybe the sub bass i was hearing was even culprit of something that is fixed now, like the shielding termination resistors (tho it couldnt be a ground loop since i have topping hs01 in the chain)
sounds seems overall "smoothed out" but dont think this means reduced dynamics, i guess people could still interpret it as such but imo it feels more like the background gets "so dark" in comparison that EFI/RFI loaded straight usb cables just make it seem like its harsher/"more dynamics" etc
high frequencys come actually cleaner, more precise trough, also because of the background thing i think
in comparison if we speak straight cable vs only topping HS01 vs IFI isilencer vs my filter
- straight wire is basicly unlistenable (IF) you heared once what better cable or tweak devices do... and i had multiple PC's, i think different PC's can sound slightly different because of a different EMI profile but overall one isnt "much" better than another, its better to filter afterwards in all cases i think
the only pc usb connection that was bearable unfiltered was the one from a raspberry PI, underclocked even better
- topping hs01... does a bit but its definitely in a region where i could say i just imagined the minor improvement, but yea i think there was one
- ifi isilencer, it was my first usb filter and i still think it does something nice on its own, definitely more than the topping HS01, but it also falls short after hearing my device and thats also why i think IFI recommends to use multiple isilencer for greater benefit...
(tho i also need to test this with my device, then i could also add 4 common mode chokes to data lines in the next revision if stacking two boards works fine...)
i can definitely say that with my device in the chain the background over usb was never darker before, but i also didnt try reclockers yet because they are rather expensive and i still wanna avoid spending 1-2k only on a usb connection...
if you know what EMI/RFI does, usually your usb connection is loaded with it and IME you need to remove majority of EMI first to even hear minor difference that seems to get otherwise masked by EMI it feels like....
of the dac or of the ground filter device? pretty sure there are theoretical improvements after the device if you look at specific things (noise on datalines after common chokes for example), also the supercap should act kind of as a mild noise filter, not sure about the dac output tho...
ESR probably also improves with the supercap
and there is a noticable audible change if i switch the ground passtrough switch back and forth (making a straight ground connection over the two planes or going over 2x 1k ohm resistors in my case), also this should affect noise on the GND line after the resistor pretty greatly... a resistor in such usecase is basicly the best noisefilter there is... it works from DC to ultra high frequencys...
the ifi idefender designer told me even 100 ohm are like 1000x reduction of noise/ground loops... so with 500 ohm its 5000x
A isolator on the other hand also transmits noise to a pretty high degree without additional filtering... ground loops get broken but the noise is still transmitted from what i read
Last edited:
since i dont have the equipment if someone wanna measure this im definitely willing to send one "for free" if you pay the shipping cost from germany
tho i would like someone to test it that has real life expierence with such tweaks and possible audible change of that... not one of the no-sayer objectivists if you catch my drift
so we can get a second thought on the audible change too...
tho i would like someone to test it that has real life expierence with such tweaks and possible audible change of that... not one of the no-sayer objectivists if you catch my drift
so we can get a second thought on the audible change too...some more tests:
straight cable vs only the ground filter: the effects of emi filtering become really apparent in this comparison .. the ground filter clearly wins here if i only use one device
topping hs-1 + ground filter : somehow it sounds worse again, which is kinda strange...
ifi isilencer + ground filter: emi reducing effects become again more apparant with the ifi isilencer, alone the ifi isilencer does around 50-70% of what the ground filter seems todo, they dont sound exactly the same but i think they go in the same direction audible and from a design pov which is reducing EMI/RFI on the lines and improving power delivery and my device has the "ifi grounddefender" included so its kinda unfair to compare
2x ground filter after each other doesnt seem to work... i got a connection(sound) once but it broke after half a second, tho i had adapter cables and a 2m extension before all that, so i might need to reduce the test setup to see if i can get 2 running
straight cable vs only the ground filter: the effects of emi filtering become really apparent in this comparison .. the ground filter clearly wins here if i only use one device
topping hs-1 + ground filter : somehow it sounds worse again, which is kinda strange...
ifi isilencer + ground filter: emi reducing effects become again more apparant with the ifi isilencer, alone the ifi isilencer does around 50-70% of what the ground filter seems todo, they dont sound exactly the same but i think they go in the same direction audible and from a design pov which is reducing EMI/RFI on the lines and improving power delivery and my device has the "ifi grounddefender" included so its kinda unfair to compare
2x ground filter after each other doesnt seem to work... i got a connection(sound) once but it broke after half a second, tho i had adapter cables and a 2m extension before all that, so i might need to reduce the test setup to see if i can get 2 running
from all the tests my believing goes even more into the direction that EMI/RFI needs to be kept out as much as possible and even a isolator doesnt seem todo a great job here tho the topping isolator is generally not well regarded i think, the intona isolators might have additional filtering inside
i also believe as long as the dac is able to play sound with no interruptions than i dont think even a impedance missmatch could alter the sound in such way... its the crappy EMI/RFI..
imo such emi filtering tweaks are also more effective than changing cables for example... which could alter EMI/RFI pickup by the cable because of different shielding and geometry...
IMO this EMI theory explains better why people hear such differences with USB than any other theory... it just makes more sense even if i cant proof it
conclusion is... i might add 3 common mode chokes in a later revision and see if it still works... apparently 4 dont (till i done more testing here)
also i might make this a "product" with a real aluminium case (so i have to adjust the PCB layout..) for shielding purposes, something like this: https://de.aliexpress.com/item/33038312237.html
i also believe as long as the dac is able to play sound with no interruptions than i dont think even a impedance missmatch could alter the sound in such way... its the crappy EMI/RFI..
imo such emi filtering tweaks are also more effective than changing cables for example... which could alter EMI/RFI pickup by the cable because of different shielding and geometry...
IMO this EMI theory explains better why people hear such differences with USB than any other theory... it just makes more sense even if i cant proof it
conclusion is... i might add 3 common mode chokes in a later revision and see if it still works... apparently 4 dont (till i done more testing here)
also i might make this a "product" with a real aluminium case (so i have to adjust the PCB layout..) for shielding purposes, something like this: https://de.aliexpress.com/item/33038312237.html
https://www.digikey.com/en/resource...sion-calculator-low-pass-and-high-pass-filter
ohh i was playing around with this...
a 1,5F supercap basicly becomes a 1Hz first order low pass filter in combination with 0,2 ohm of series resistance trough connectors and cables...
from my understand even a first order low pass is pretty effective here since it starts at 1Hz ... as we get to 100 Hz or so the attenuation should be rather big already
this kinda lets me think about super cap power buffers completely different, it basicly offers one of the lowest ESR AND is a excellent DC filter.. (without putting tens of components into the path..)
a second revision with the case also has enough room to include a power switch IC as charge current limiter and possibly bigger caps
ohh i was playing around with this...
a 1,5F supercap basicly becomes a 1Hz first order low pass filter in combination with 0,2 ohm of series resistance trough connectors and cables...
from my understand even a first order low pass is pretty effective here since it starts at 1Hz ... as we get to 100 Hz or so the attenuation should be rather big already
this kinda lets me think about super cap power buffers completely different, it basicly offers one of the lowest ESR AND is a excellent DC filter.. (without putting tens of components into the path..)
a second revision with the case also has enough room to include a power switch IC as charge current limiter and possibly bigger caps
Last edited:
i finally wanna get USB "right" at an affordable price, tho i still see myself end up one of these audiophile pcie cards
maybe in the end just to see if there is more headroom..since you use USB in any kind of (digital) Hifi setup, either its "USB to DDC" or "USB to DAC" i think the usb connection needs to be tweaked either way even if some DDC's already sound better than straight USB
maybe in the higher (inaudible) frequencys but most reviewers dont show that...
possibly jitter test could be affected i guess
My offer stands if someone wanna measure this
but i think the current completely passive layout is fairly error-proof, i would be more worried if i have switching ic's and such on it but i have to think about how i might do a next revision
yea i have some spare, since its a prototype and you are located in germany i can offer to send one for 20Euro including shipping if you leave a reviewHi ghoostknight. I just found out about idefender and googled for a diy version since it seemed a little expensive. Then k stumbled upon your posts here.
Do you sell your version?
write me a PM if you are interested
i just need to wait for the CDA supercaps i ordered, they probably arrive in 3-5 days
just keep in mind since its a kind of a prototype to
1. only add resistance between grounds if you use a external power supply, tho most dacs grounded elsewhere can probably handle it anyway
2. and watch the power supply because of the "unprotected" supercap, it will overload small power supplys for a few seconds, most should handle it, tho i dont wanna give a gurantee here... you can reduce the risk with using a high ampere power supply like 5V 3A, a 500mA usb port gets overloaded far longer than a 3A one... once the cap is fully charge there is no need to worry anymore
Last edited:
not interested anymore @Dnhaus ?
i think If you use not the cheapest external power supply there is nothing to worry about, the issues above come non-existent, as long you use a good quality external power supply....
power supply switched to EXT and i have currently 3x 1k ohm resistors switched on ground for 333 ohm.... the 500 ohm mentioned before actually interrupted/disconnected the dac (Aune X8) after some listening (after a few hours), going to 333 ohm fixed this
i can also send you a pcb without supercap (or with a smaller electrolytic cap), that would eliminate the power supply risk, the other issue with gnd resistance but no additional low resistance ground shouldnt really possible fry something (in most cases), tho one should still avoid it...
i just really someone wanna test this to get a second (or third/fourth..) opinion on this unfortunaly no one i know personally is into hifi... but i think i done enough testing to categorize pros/cons pretty well on my own, but still, more thoughts on this are invalueable IMO
_____
i just received yesterday the Black Lion Audio 2x2 Revolution... great sounding dac, i think the BLA people know what they are doing even if measurements dont show it... its basicly on a another level if i compare it to the "audiophile" Aune X8
tho i havent played around with the usb ground filter yet, its plugged in since i received the device (the same as with the Aune X8) so i could compare, i will later remove the ground filter to see what it does on the BLA
also i have to test if the BLA can handle 500 ohm or if this is more dependend on cable length and general setup than what dac you use
1k ohm as resistance option is probably too much
also funny that BLA actually uses usb cables with ferrites on both ends... you dont see that very often but it speaks words IMO, they actually listening to the difference...
build quality alone is also atleast 100 worth IMO specially compared to others interfaces
i think If you use not the cheapest external power supply there is nothing to worry about, the issues above come non-existent, as long you use a good quality external power supply....
power supply switched to EXT and i have currently 3x 1k ohm resistors switched on ground for 333 ohm.... the 500 ohm mentioned before actually interrupted/disconnected the dac (Aune X8) after some listening (after a few hours), going to 333 ohm fixed this
i can also send you a pcb without supercap (or with a smaller electrolytic cap), that would eliminate the power supply risk, the other issue with gnd resistance but no additional low resistance ground shouldnt really possible fry something (in most cases), tho one should still avoid it...
i just really someone wanna test this to get a second (or third/fourth..) opinion on this
unfortunaly no one i know personally is into hifi... but i think i done enough testing to categorize pros/cons pretty well on my own, but still, more thoughts on this are invalueable IMO_____
i just received yesterday the Black Lion Audio 2x2 Revolution... great sounding dac, i think the BLA people know what they are doing even if measurements dont show it... its basicly on a another level if i compare it to the "audiophile" Aune X8
tho i havent played around with the usb ground filter yet, its plugged in since i received the device (the same as with the Aune X8) so i could compare, i will later remove the ground filter to see what it does on the BLA
also i have to test if the BLA can handle 500 ohm or if this is more dependend on cable length and general setup than what dac you use
1k ohm as resistance option is probably too much
also funny that BLA actually uses usb cables with ferrites on both ends... you dont see that very often but it speaks words IMO, they actually listening to the difference...
build quality alone is also atleast 100 worth IMO specially compared to others interfaces
Ok i did remove the ground filter now and wow.... it sounds "noisier" overall (typical EMI effect ime) and bass was pretty much gone in comparison
tho the bass thing i definitely account to the supercap, this is also what i noticed with other supercap tweaks before... bass becomes way cleaner and deeper (and its not just the bass, but thats the most obvious thing, overall resolution seems to increase too..)
Actually.... its strange that that bass resolution seems to increase that much and that the supercap acts as an excellent DC filter starting at 1 Hz.... is it a correalation here?
___________________
Tho im actually not sure if i wanna go the isolator route, it actually involves more complicated IC's on the clean side, where im unsure if they even can preserve the original signal "quality", pretty sure there is some degredation going on, either fixable with a usb hub afterwards that basicly reclocks the signal
OR
we just try to preserve the quality of the original signal as best as possible and filtering unwanted stuff... from tests so far this seems to work really well and i havent noticed any real degredation from removing the topping HS01, it even got slightly better imo
tho i remember from some tests 2-3 years ago with USB hubs... each hub/implemenation sounds different, so to get something seriously good sounding alot of testing is required, whats interesting still here is wether a good usb filter actually makes the differences of usb hubs or cables less if the filter sits right at the dac... (thats kinda my expectation but i havent tested with cables alot yet)
tho the bass thing i definitely account to the supercap, this is also what i noticed with other supercap tweaks before... bass becomes way cleaner and deeper (and its not just the bass, but thats the most obvious thing, overall resolution seems to increase too..)
Actually.... its strange that that bass resolution seems to increase that much and that the supercap acts as an excellent DC filter starting at 1 Hz.... is it a correalation here?
___________________
Do you have schematics for me?
Tho im actually not sure if i wanna go the isolator route, it actually involves more complicated IC's on the clean side, where im unsure if they even can preserve the original signal "quality", pretty sure there is some degredation going on, either fixable with a usb hub afterwards that basicly reclocks the signal
OR
we just try to preserve the quality of the original signal as best as possible and filtering unwanted stuff... from tests so far this seems to work really well and i havent noticed any real degredation from removing the topping HS01, it even got slightly better imo
tho i remember from some tests 2-3 years ago with USB hubs... each hub/implemenation sounds different, so to get something seriously good sounding alot of testing is required, whats interesting still here is wether a good usb filter actually makes the differences of usb hubs or cables less if the filter sits right at the dac... (thats kinda my expectation but i havent tested with cables alot yet)
yea definitely, all about USB seems to be a compromise, but you are right i agree with the general consent to keep cables as short as possible, tho for my setup it has to be 2m... i dont wanna park my hifi directly beside my PC...
I don't have schematics but the descriptions of the experiments around the DAC and an accompanying AC Filter box are in a very long thread over at audiophilestyle.com. Look for 'AC Filter box' to find the thread, at least if the search engine is better than it was. It all started with an inline USB Power filter which wasn't too good for my DAC, but I kept that circuit and re-used it while powering the DAC with a custom low-noise Linear PSU.
ah it must be this one https://audiophilestyle.com/forums/...rounding-boxes-linear-psu-and-balanced-power/
i take a read when i can thank you, any insight is helpful with this stuff, specially listening expierence...
thanks
i take a read when i can thank you, any insight is helpful with this stuff, specially listening expierence...
thanks
i actually also still wanna build a power distrubution unit ... i already have the right EMI filter for it which im already using in a non-ideal plastic box, also this made a noticable difference EMI/RFI wise... also ferrite beads help alot, you can specially place 5-10 of them on the power cable of your PC for example... it really helps
tho the power distribution unit i have in mind is a bit special...
i wanna take this one: https://www.thomann.de/de/the_tracks_racksteckdose_19_mkii.htm
rip out the insides besides the socket, route everything with 2,5 or 4mm2 OFC/OCC Copper, maybe exchange one socket with a plate that can take an IEC socket, put emi filter and possibly DC filter inside and fill the whole thing with some kind of mineral mix, as sort of grounding box
i think this is also a good test if the minerals really matter, theres no way such a design could act as "antenna" (since the ground wire is routes to the grounding box) either the minerals really do a difference or it makes no difference at all...
tho one thing still holds me back.... if i wanna go with the rack format, most racks are rather flimsy or pricey...