Power Supply (8+) Group Test, LPS and SMPS

[DM]

10 hours ago, tipunch said:

 

Hello, thank you very much!

But where it places the AQVOX in the classification? What are the measures?

 

Best regards

 

Aqvox USB Low Noise Power Supply is just a very generic old style linear wall wart power supply, both filtering and regulator used are nothing to write home about, but it is better than generic SMPS power supply out there, so a score of 60 ?

 

1. 95 9V alkaline battery

 

2. 93 iFi iPower 9V SMPS

 

3. 90 ZERO-ZONE SUPER-PSU

 

4. 80 TeraDak TeraLink X2

 

5. 65 Breeze Audio DC-1 / S.M.S.L Panda dual 9V (both flawed)

 

6. 60 AQVOX  USB Low Noise Power Supply

 

6. 55 Apple charger

 

7. 50 Xiaomi 10040mAh battery bank

[the_doc735]

On 6/15/2018 at 2:08 AM, DM said:

 

Aqvox USB Low Noise Power Supply is just a very generic old style linear wall wart power supply, both filtering and regulator used are nothing to write home about, but it is better than generic SMPS power supply out there, so a score of 60 ?

 

1. 95 9V alkaline battery

 

2. 93 iFi iPower 9V SMPS

 

3. 90 ZERO-ZONE SUPER-PSU

 

4. 80 TeraDak TeraLink X2

 

5. 65 Breeze Audio DC-1 / S.M.S.L Panda dual 9V (both flawed)

 

6. 60 AQVOX  USB Low Noise Power Supply

 

6. 55 Apple charger

 

7. 50 Xiaomi 10040mAh battery bank

Many thanks for this test, most grateful.

I think these must be the quietest LPSU on the planet? Sean Jacobs. They claim figures of just 0.02mV ripple.

[franz159]

Good morning, is the Zero-zone PSU reviewed here similar to this one conceptually?

https://www.ebay.com/itm/S120-60W-60W-R-Core-Ultra-Low-Noise-Linear-Power-supply-2-way-DC-5V-12V-LPS-PSU/132021195239?hash=item1ebd1391e7:g:i18AAOSwWP9bKl2-

 

Many thanks

[lmitche]

43 minutes ago, franz159 said:

Good morning, is the Zero-zone PSU reviewed here similar to this one conceptually?

https://www.ebay.com/itm/S120-60W-60W-R-Core-Ultra-Low-Noise-Linear-Power-supply-2-way-DC-5V-12V-LPS-PSU/132021195239?hash=item1ebd1391e7:g:i18AAOSwWP9bKl2-

 

Many thanks

It is hard to say how similar the lpsu linked to above is to the Zero Zone used in the review. Most likely they both use Sigma 11 based regulator boards in the design, but there are many variations with different transformer types (rcore and toroid) caps (Nover and Nichicon and mosfets types (Infineon and Toshiba).

 

I have a zero zone dual lpsu in a silver aluminum box as shown in the link.  It has two Nichicon caps per regulator board and two Infineon mosfets.  It sounds great and measures well.

[lmitche]

1 hour ago, franz159 said:

Good morning, is the Zero-zone PSU reviewed here similar to this one conceptually?

https://www.ebay.com/itm/S120-60W-60W-R-Core-Ultra-Low-Noise-Linear-Power-supply-2-way-DC-5V-12V-LPS-PSU/132021195239?hash=item1ebd1391e7:g:i18AAOSwWP9bKl2-

 

Many thanks

Here is the one purchased here with Nichicon caps:

 

https://www.ebay.com/itm/ZEROZONE-HIFI-130W-Ultra-Low-Noise-DC-Linear-Power-Supply-2-Way-LPS-65W-65W-/263031412557?hash=item3d3de4d74d

 

If you buy one be sure to message the seller with your voltage requiements for both lpsus.

[Superdad]

3 hours ago, lmitche said:

 

https://www.ebay.com/itm/ZEROZONE-HIFI-130W-Ultra-Low-Noise-DC-Linear-Power-Supply-2-Way-LPS-65W-65W-/263031412557?hash=item3d3de4d74d

 

I have never understood how the Chinese get away with rating the wattage the way they do on their supplies.  Correct ratings would be based on max current at a given output voltage.  Multiply current times voltage. But do so for example with the supply you linked to, and you’ll come up with wattage in the 36-42W range.

 

Just sayin’...

[lmitche]

10 minutes ago, Superdad said:

 

I have never understood how the Chinese get away with rating the wattage the way they do on their supplies.  Correct ratings would be based on max current at a given output voltage.  Multiply current times voltage. But do so for example with the supply you linked to, and you’ll come up with wattage in the 36-42W range.

 

Just sayin’...

Yes, you are right! There is no clear output current rating specified with these LPSes. Strangely they are rated by the wattage of their AC transformers 2 x 65 watts in this case.

[look&listen]

35 minutes ago, Superdad said:

4 hours ago, lmitche said:

 

https://www.ebay.com/itm/ZEROZONE-HIFI-130W-Ultra-Low-Noise-DC-Linear-Power-Supply-2-Way-LPS-65W-65W-/263031412557?hash=item3d3de4d74d

 

I have never understood how the Chinese get away with rating the wattage the way they do on their supplies.

If old eyes still work Ok, in eBay inside product picture, printing on label of R-core say "18V/50W"!

Blatant lying now prevailing fashion, endorsed by big governments all over world 

 

Also note very fine wire dressing in unit- well, except input AC wires wrapped around regulator caps (all Hi-V AC wire routing suspect IMO) 

 

Still, very nice LPSU for price. I would buy if needed that configuration & use specs conservativly (would also clean up a few things inside post purchase). 

[Superdad]

4 minutes ago, lmitche said:

Yes, you are right! There is no clear output current rating specified with these LPSes. Strangely they are rated by the wattage of their AC transformers 2 x 65 watts in this case.

 

The AC wattage ratings of Chinese R-core transformers are all over the map and not very useful.  And even there, if you look closely at the label of the one you linked to, it says 50W. 

[Superdad]

2 minutes ago, look&listen said:

Blatant lying now all the fashion, endorsed by big governments all over world.

 

Well I won’t touch that political statement! 

 

2 minutes ago, look&listen said:

Also note very fine wire dressing in unit- well, except input AC wires wrapped around regulator caps. 

 

Nah, what bothers me more is the really crappy bridge rectifiers that most all the Chinese units use.  I mean, would it kill them to spend a few Yuan more on some decent Schottky diodes?

Worse is the fact that every Chinese LPS I have had hooked up to my electronic load box failed to reach its current ratings. (Somewhere here on CA I posted photo proof of one such test with a smaller ZeroZone unit.)

Maybe I should buy this one and put it on the work bench to test and report.

[look&listen]

27 minutes ago, Superdad said:

Nah, what bothers me more is the really crappy bridge rectifiers that most all the Chinese units use.  I mean, would it kill them to spend a few Yuan more on some decent Schottky diodes?

I not see (always crappy) bridge rectifier in this advertised unit pictures. See big pass MosFet & 2 To-220 devices suspect are probably Fast Recovery diodes. Maybe 2 more hiding under PCB, since looks like S-11 variant board?

Also wonder about lack of higher current Schottky diodes in such products. Maybe non-linear pricing above 3 amps capacity?

 

27 minutes ago, Superdad said:

Worse is the fact that every Chinese LPS I have had hooked up to my electronic load box failed to reach its current ratings. (Somewhere here on CA I posted photo proof of one such test with a smaller ZeroZone unit.)

Maybe I should buy this one and put it on the work bench to test and report.

Thx for good data. 

Upvote for testing! If not destroyed, sure some audiophile happy to buy used

 

 

Just noticed V/A ratings way down in product text, seem more realistic then watt ratings in headline-

 

you can choose following any two kinds voltage as the two output:
(1: 5V 4.3A    2: 9V 4A    3: 12V 3.5A    4: 15V 2.6A    5: 18V 2.1A     7:19V  2A    8:24V 1.6A)

 

[the_doc735]

the chinese never post ripple figures, but when I wrote to them myself to ask, the replies were always 8mV +. No where near teddy pardo at just 12nV, claiming superiority over Naim's own stuff.

[Superdad]

2 hours ago, the_doc735 said:

the chinese never post ripple figures, but when I wrote to them myself to ask, the replies were always 8mV +. No where near teddy pardo at just 12nV, claiming superiority over Naim's own stuff.

 

[Teddy Pardo builds some fine low-noise, discrete regulator LPS units.  What I wrote below has nothing to do with him.]

 

Be careful when reading of quoting PS noise figures in the lower ranges.

First off, it is extremely difficult to make proper measurements, even just below the single digit milliVolt (mV) range, let alone the microVolt (µV) and nanoVolt (nV) ranges.  We are talking shielded boxes, differential probes, and measures taken right at the board level.  

Specs that are bandied about are usually just for the regulator itself.

 

Secondly, single numbers are meaningless unless:

a) it is stated what is being measured--RMS noise, peak noise, spot noise--with units;

b) particular frequency or bandwidth of the measurement (the wider the better, especially for use with digital electronics).

 

For example, the Linear Technology LT3042 and LT3045, currently about the lowest noise integrated linear regulators (200mA and 500mA respectively) on the planet, have these specs as the opening to a comprehensive data sheet with graphs:

• Ultralow RMS Noise: 0.8μVRMS (10Hz to 100kHz)
• Ultralow Spot Noise: 2nV/√Hz at 10kHz
• Ultrahigh PSRR: 76dB at 1MHz

Beyond just noise specs (and the PSRR at high freq. is an important one)--which ideally would be provided for the entire supply--there is parameter even more crucial to how the supply audibly performs in an audios system, again most especially over wide-band for digital components.  I am of course speaking of output impedance.  The lower the better.  And if you see an LPS with big value capacitors--or banks of caps--AFTER the output regulator, run the other way!  It is shocking how often we see this.  Even more than a few hundred µF of capacitance after the regs is a sign that they did not know what they were doing and it is good bet that such a supply will be "slow."

[sandyk]

6 hours ago, Superdad said:

 

I have never understood how the Chinese get away with rating the wattage the way they do on their supplies.  Correct ratings would be based on max current at a given output voltage.  Multiply current times voltage. But do so for example with the supply you linked to, and you’ll come up with wattage in the 36-42W range.

 

Just sayin’...

Alex C

 That annoys me too.

 In fact, MANY of their PSUs don't have anywhere near adequate heatsinking for any more than VERY short term loads at their rated currents.

Where they use the VA ratings iof the transformer it's even more misleading, as when using a Bridge Rectifier with a capacitor input.   
IDC = 0.62 X Sec I AC

 

Regards

Alex K.

[Superdad]

39 minutes ago, sandyk said:

That annoys me too.

 In fact, MANY of their PSUs don't have anywhere near adequate heatsinking for any more than VERY short term loads at their rated currents.

 

Yeah, I routinely run test our JS-2 units at 12V/7.2A for an hour+.  Though the real torture test--and the one we use for production thermal-pad phase change, retightening, etc. on every unit--is 5V/4.8A.  The drop to 5V from the raw voltage after the rectifiers results in about 30W of heat to dissipate! Of course we designed for this:

 

In reality nobody ever runs 5V at such high current; at 12V/6A it runs cool, and folks often remark about how cool their JS-2 runs.  Note above the small. low ESR, 220uF caps post the regulators.  Our choke filter design feeds nice half-sine-wave humps into the regs as opposed to the sawtooth off just diodes. This makes the regs job easy as there are no high-frequcy components to filter out, in turn allowing us to keep the post-reg caps small.  Its a "fast," very low-impedance supply. 

[[email protected]]

14 hours ago, Superdad said:

 

I have never understood how the Chinese get away with rating the wattage the way they do on their supplies.  Correct ratings would be based on max current at a given output voltage.  Multiply current times voltage. But do so for example with the supply you linked to, and you’ll come up with wattage in the 36-42W range.

 

Just sayin’...

it is marketing

 

Dual LT3045 PSU rated for 1A (not 1.1A)

6x10F capacitors in series = 1.7F (not 60F)

 

i am right???? ?

[Superdad]

19 hours ago, [email protected] said:

it is marketing

 

Dual LT3045 PSU rated for 1A (not 1.1A)

6x10F capacitors in series = 1.7F (not 60F)

 

i am right???? ?

 

Hi Alexey:

 

Not sure what you are getting at here.  Although I was speaking of higher current supplies not meeting their stated specifications and the specific high current tests I put our 5-7A JS-2 through, you seem to be changing the subject to our popular isolated, ultracapacitor-based UltraCap LPS-1.2 .  

That is fine, but are you trying to say that our 1.1A rating for the LPS-1.2 is just marketing hyperbole and not truthful?   If so, I take umbrage with that as I personally run every unit up to at least 1.1A, and most make it to about 1.15A before the unit's monitoring software trips it into over-current protection mode.  I have run units for days on end at their 1.15A limit with no issue whatsoever.  Here s a photo of one going through normal production test:

12V output setting, 1.144A load.

 

I am aware that you produce a popular line of LT3045-based DIY DC-DC boards (though per CA rules you really ought to disclose your industry affiliation in your signature).  So I would have expected that you would be more than familiar with what this wonderful ultra-low-noise regulator is capable of when properly used.

 

By the way, the UltraCap LPS-1.2 actually has 14 pieces of 10Farad ultracaps (7 per alternating bank), and yes we know how capacitors in series divide down.  The ultracaps are used as a battery-like, isolated from AC and charging side, low-impedance power source, so while they have some filtering function, the 1.43F per bank is more than ample.  

 

Wishing you the best of success,

--Alex C.

[the_doc735]

@DM 

 

So, are you going to test (if you can get hold of samples?):

uptone

paul hynes

paul pang

ted pardo

sean jacobs

Baldwin

thelinearsolution

SOtM

 

....and other hi-end appliances like these/ (not the low quality chinese)? Interested to know how these would 'fair' against SMPS? Better or worse? (for audio of course!)?

[the_doc735]

17 hours ago, the_doc735 said:

@DM 

 

So, are you going to test (if you can get hold of samples?):

uptone

paul hynes

paul pang

ted pardo

sean jacobs

Baldwin

thelinearsolution

SOtM

 

....and other hi-end appliances like these/ (not the low quality chinese)? Interested to know how these would 'fair' against SMPS? Better or worse? (for audio of course!)?

Mojo

 

etc......

[Kelly]

Some very established but forward looking companies are using SMPS in their units. All the latest NAD master series is SMPS and I believe Linn also uses SMPS in all of their products. I know my Kii Theee speakers use SMPS power supplies. I believe the main problems with SMPS implementations is  that they emit a lot of noise back into the power mains if care is not exercised to prevent it. So that is why an accessory linear supply may sound better than an SMPS even if it does not provide cleaner output in a lot of subjective tests: the linear supply does not crud up the mains for the rest of the system.

[the_doc735]

1 hour ago, Kelly said:

Some very established but forward looking companies are using SMPS in their units. All the latest NAD master series is SMPS and I believe Linn also uses SMPS in all of their products. I know my Kii Theee speakers use SMPS power supplies. I believe the main problems with SMPS implementations is  that they emit a lot of noise back into the power mains if care is not exercised to prevent it. So that is why an accessory linear supply may sound better than an SMPS even if it does not provide cleaner output in a lot of subjective tests: the linear supply does not crud up the mains for the rest of the system.

So where does that leave this?

How about this?

"Up to ten times quieter than than iFi's already excellent ultra-low-noise stock AC/DC adapter, and 20 times quieter than conventional audiophile linear power supplies, iPower has a virtually nonexistent noise floor of 0.000001 Volts (1μV). " 

 

Still back polluting the mains? (to rest of gear)...

[Kelly]

Just to be clear, I don't personally claim that an LPS sounds better than an SMPS. I don't subscribe to any particular dogma other than if you design it well then it will sound good. But that is indeed one reason I have heard to keep cheap SMPS away from your audio system. My understanding is that extra care can be taken to minimize the high frequency noise the SMPS emits back into the mains, but that is expensive and larger. I would bet that something like the Mola Mola Kaluga (one of the absolute finest amplifiers at any price) does not pollute back into the mains, but a small wall-wart is more likely to.

 

To address the iFi, if someone were to compare it to an inferior measuring LPS and still feel the LPS sounded better. Then the iFi maybe generating more noise back in the mains would be a possible rationale, and perhaps the next thing to measure, if you could figure out how to measure such a thing accurately.

[the_doc735]

50 minutes ago, Kelly said:

My understanding is that extra care can be taken to minimize the high frequency noise the SMPS emits back into the mains, but that is expensive and larger.

Yes, it is expensive!

So, instead of spending all that money on preventing back pollution (from SMPS) electronics, why not just spend the money on a supremo LPSU in the first place, that does not suffer from those problems in the first place? Cheers!

[Cornan]

17 hours ago, the_doc735 said:

Yes, it is expensive!

So, instead of spending all that money on preventing back pollution (from SMPS) electronics, why not just spend the money on a supremo LPSU in the first place, that does not suffer from those problems in the first place? Cheers!

Or you could buy yourself a cheap floating SMPS (like Gophert csp-3205II) followed by LT3045 in series (making it a LPS with a SMPS feeder supply). That would give you some extra cash to improve other things like isolation transformer, DC blocker and better ICs. ?

[the_doc735]

13 minutes ago, Cornan said:

Or you could buy yourself a cheap floating SMPS (like Gophert csp-3205II) followed by LT3045 in series (making it a LPS with a SMPS feeder supply). That would give you some extra cash to improve other things like isolation transformer, DC blocker and better ICs. ?

hybrid?