best PS vs battery
May 29, 2015 at 5:55 PM Thread Starter Post #1 of 13

goobicii

100+ Head-Fier
Joined
Feb 21, 2013
Posts
347
Likes
29
can you educate me swiftly on whats better? I done little research and this Hynes powersupply is considered best PS for high quality dacs & amps....  how does it compare to lets say top battery system,I never tried myself but reed each have different sound......
 
isnt it best to put the best model of Hynes powersupply to every component in DAC and amp? know it will be super expensive but I realized PS have very big impact on sound quality
 
May 29, 2015 at 9:29 PM Post #2 of 13

tomb

Member of the Trade: Beezar.com
Joined
Mar 1, 2006
Posts
10,513
Likes
743
Looks like Paul Hynes quotes his power supplies at 110dB power line noise rejection.  If my math is correct, and he says his power supplies are good to 24VDC, then the noise in his power supply is 0.075mV.  That's respectable, but far from the best out there, especially in DIY-land. 
wink.gif

 
Tangent's STEPS power supply rated 0.058mV, his TREAD was 0.060mV.  The Millett Hybrid MiniMAX has a power supply that averages about 0.045mV - and that's at 27VDC.
 
AMB's Sigma power supplies measure at 0.010mV at 30VDC.
 
I'm not sure where you got that those power supplies are "best PS for high quality dacs & amps," but as you can see, the DIY section of Head-Fi has been doing better for the last 9 years or so, at least Tangent's stuff has been here that long, AMB's Sigma stuff came a few years later. 
wink.gif

 
May 29, 2015 at 9:48 PM Post #3 of 13

tomb

Member of the Trade: Beezar.com
Joined
Mar 1, 2006
Posts
10,513
Likes
743
Just an FYI, but batteries are not magic bullets.  Yes, they're quiet compared to the typical AC-DC power supply, but they are extremely limited in voltage and some (alkaline) have huge impedance issues.  Alkaline batteries are very limited in current because they have a high output impedance.  It's one of the reasons they last so much longer.  NiCds, NiMH, and LiPo offer huge increases in current, but then the voltage is very limited.
 
Bottom line, you don't get something for nothing ...
 
May 29, 2015 at 11:08 PM Post #4 of 13

goobicii

100+ Head-Fier
Joined
Feb 21, 2013
Posts
347
Likes
29
  Just an FYI, but batteries are not magic bullets.  Yes, they're quiet compared to the typical AC-DC power supply, but they are extremely limited in voltage and some (alkaline) have huge impedance issues.  Alkaline batteries are very limited in current because they have a high output impedance.  It's one of the reasons they last so much longer.  NiCds, NiMH, and LiPo offer huge increases in current, but then the voltage is very limited.
 
Bottom line, you don't get something for nothing ...

 
so the powersupply of these popular Beta22 amps are best powersupply you can have for headphone amp?
 
May 29, 2015 at 11:38 PM Post #5 of 13

tomb

Member of the Trade: Beezar.com
Joined
Mar 1, 2006
Posts
10,513
Likes
743
 
Just an FYI, but batteries are not magic bullets.  Yes, they're quiet compared to the typical AC-DC power supply, but they are extremely limited in voltage and some (alkaline) have huge impedance issues.  Alkaline batteries are very limited in current because they have a high output impedance.  It's one of the reasons they last so much longer.  NiCds, NiMH, and LiPo offer huge increases in current, but then the voltage is very limited.

Bottom line, you don't get something for nothing ...


so the powersupply of these popular Beta22 amps are best powersupply you can have for headphone amp?


Generally? The question is would you would really hear that difference. With a DAC? Yes, probably.
 
May 31, 2015 at 1:29 PM Post #6 of 13

kevin gilmore

Señor Stax. Señor MAXX.
Can Jam '10 Organizer
Joined
Jun 21, 2001
Posts
2,863
Likes
516
I have a klone board of the golden reference power supply.
 
tests at 60 nanovolts peak to peak noise at +/-20v at up to 500ma
 
more quiet than any battery, or the electrochemical capacitors.
 
measuring these levels of noise is very hard.
 
May 31, 2015 at 11:03 PM Post #7 of 13

AudioCats

Headphoneus Supremus
Joined
Feb 25, 2007
Posts
3,418
Likes
80
Location
CO, USA
I have compared the Sigma-22 vs. NiMh bank (powering the low voltage tube section in a hybrid), and I prefer the battery bank. Calmer sound, darker background.
Either one is a bit better than the common 78/79xx regulator though.
 
 
Quote:
  I have a klone board of the golden reference power supply.
 
tests at 60 nanovolts peak to peak noise at +/-20v at up to 500ma
 
more quiet than any battery, or the electrochemical capacitors.
 
measuring these levels of noise is very hard.

 
How does it compare to batteries (sound wise), when driving a small amp?
 
Jun 1, 2015 at 5:33 AM Post #8 of 13

kevin gilmore

Señor Stax. Señor MAXX.
Can Jam '10 Organizer
Joined
Jun 21, 2001
Posts
2,863
Likes
516
I'm going to be doing some testing this week.
 
The problem with batteries is that they come in only 6V and 12V
versions (sealed gel cells) So I have to get some more 6V batteries
to test my current amplifier on +/-18  (which should get close to +/-20)
 
Jun 6, 2015 at 11:01 AM Post #9 of 13

kevin gilmore

Señor Stax. Señor MAXX.
Can Jam '10 Organizer
Joined
Jun 21, 2001
Posts
2,863
Likes
516
1)  So a total of 6 x 6V/7.2AH batteries wired as +/-18V  (actually closer to +/-19.6 fully charged)
 
2) newest 3 stage goldenreference wired at +/- 19.8V  driven from AC transformer
 
3) newest golden reference driven from 8 x 6V/7.2AH batteries
 
1 is the most noisy, but you have to be in a very dead room and listen with sensitive iem's to notice.
 at 500ma, the batteries start to drift voltage wise after as little as 1 hour and then the servo has to start working.
Balancing batteries over time is impossible as far as I know.
 
Can't tell any difference between 2 and 3
 
all on the latest ssdynalobal board
 
virtually all of the noise from the batteries is in the range of 1 to 10hz
 
conclusion, won't be doing battery power supplies any time in the future except for
portable applications.
 
Jun 6, 2015 at 5:21 PM Post #10 of 13

AudioCats

Headphoneus Supremus
Joined
Feb 25, 2007
Posts
3,418
Likes
80
Location
CO, USA
Thanks Kevin, great info.
 
My take on your test result is that, for low current application (say, <10mA for the input stage), battery might be a rather viable option? noise mostly in 1-10Hz range probably won't matter to my equipment much.
 
how much are those GoldenReference power supplies? Just curious.
 
Jun 6, 2015 at 8:11 PM Post #11 of 13

kevin gilmore

Señor Stax. Señor MAXX.
Can Jam '10 Organizer
Joined
Jun 21, 2001
Posts
2,863
Likes
516
as long as you can get a battery (or batteries) that is the voltage you want, gel cells at 10ma would last quite a long time.
you still have to find a way to charge them.
 
way back when, the stax phono cart amplifier box had a battery that was half the size of the box. It would play
about 30 hours at one time, when you turned it off, it automatically charged.
 
The goldenreference board is about $18
 
lt1021 is $7.04  2 needed
opa134 is $2.94  2 needed
power transistors $5.87 2 needed
big electrolytic $2.48 2 needed
wima caps $2.39  2 needed (not sure if this is the right part)
heatsinks $11.08 2 needed
transformer $30
 
the rest of the stuff is cheap
 
so say about $110 in parts
 
you can buy the original version from china for $48 plus shipping, cash only (no transformer)
which is why I did my version
 
I'm using much more expensive and accurate reference chips and
an extra pre-reg stage, so its more money.
 
board layout in my boards directory
http://gilmore.chem.northwestern.edu/goldenreference.pdf
 
with resistor changes, will do any voltages from +/-12 to +/-36
 
Jun 7, 2015 at 9:06 AM Post #13 of 13

kevin gilmore

Señor Stax. Señor MAXX.
Can Jam '10 Organizer
Joined
Jun 21, 2001
Posts
2,863
Likes
516
slightly off, and the transformer listed is way overkill, I used it for the ksa5 clone.
 
http://gilmore.chem.northwestern.edu/ssdynalobal.xlsx
 
all mouser parts except 1 digikey part
 

Users who are viewing this thread

Top