Jump to content
IGNORED

UpTone JS-2 Power Supply Listening Impressions


Recommended Posts

38 minutes ago, Dev said:

I have a JS-2 (borrowed from a friend) powering a i7 NUC8 (rated 28watt tdp) at 12v. When using HQP to upsample to dsd256, I see that the amber light on JS-2 flickers. Does anyone know what does it signify or is this a normal behavior ?

 

Hi Dev:

I am surprised that the front panel LED is flickering.  Have never seen that.  Perhaps the LED is failing.  Feel free to contact me directly about it.

 

Fed 120/240V AC input the JS-2 can produce a total continuous 7.2 amps at 12V (max current at 12V is slightly less when the wall voltage is 115/230V or even 110/220V).  Peak capability is higher still.

That 86.4 watts is plenty for any NUC, so you are not running short of current.  

 

Do you also have a load on the second output rail of the JS-2?  What is it?

 

A better indication of load are the red LEDs at the back of the unit.  A bit of red flickering is normal, but steady red would mean that the unit is falling out of regulation and the output voltage will be reduced.

 

Thanks,

--Alex C.

Link to comment

@Superdad, thanks for the prompt response.

 

The NUC is only connected to the JS-2, so no load on the 2nd output. I have 2x8Gb ddr4 and a 2tb SSD on this NUC but given the output capacity of 86watts, I guess its a non issue as far as power is concerned.

 

I just checked the back while playing and the red led is blinking rapidly. Is that an indication of something failing ?


 

Link to comment
5 minutes ago, Superdad said:

 

Not at all.  Just an indication of lots of big current draw peaks--consistent with what it takes to convolute DSD512.  :)

How's it sound?

 

Its only dsd256 and -2s or ext2 variant. I hooked the JS-2 with this NUC today and saw the flickering. So thought of asking. The NUC is used on the server side, so not sure if the power supply would make a difference but you never know. I will do some meaningful comparison with the stock smps. 

Link to comment

@Dev do you think perhaps the limitation is thermal throttling?  Running a NUC8i7BEH hard (with turbo boost enabled) for an extended period will likely trigger thermal throttling.  See the section on Temperature (near the end) of this review.

https://www.notebookcheck.net/Intel-NUC-Kit-NUC8i7BEH-i7-8559U-Mini-PC-Review.360356.0.html

Pareto Audio AMD 7700 Server --> Berkeley Alpha USB --> Jeff Rowland Aeris --> Jeff Rowland 625 S2 --> Focal Utopia 3 Diablos with 2 x Focal Electra SW 1000 BE subs

 

i7-6700K/Windows 10  --> EVGA Nu Audio Card --> Focal CMS50's 

Link to comment
7 minutes ago, rickca said:

@Dev do you think perhaps the limitation is thermal throttling?  Running a NUC8i7BEH hard (with turbo boost enabled) for an extended period will likely trigger thermal throttling.  See the section on Temperature (near the end) of this review.

https://www.notebookcheck.net/Intel-NUC-Kit-NUC8i7BEH-i7-8559U-Mini-PC-Review.360356.0.html

 

yes, I think this model is severely limited to thermal throttling. Unless we get a fanless case, the stock cooling doesn't seem sufficient. I reported the same earlier:

 

 

Link to comment
2 minutes ago, Dev said:

 

yes, I think this model is severely limited to thermal throttling. Unless we get a fanless case, the stock cooling doesn't seem sufficient. I reported the same earlier:

 

 

 

 

If a LPS doesn't make a difference on the server side and if you upsampling, I think a much better machine is NUC8I7HVK.

Link to comment
3 minutes ago, Dev said:

I reported the same earlier

I'm so sorry!  You're the one who pointed me to that review in the first place!

Pareto Audio AMD 7700 Server --> Berkeley Alpha USB --> Jeff Rowland Aeris --> Jeff Rowland 625 S2 --> Focal Utopia 3 Diablos with 2 x Focal Electra SW 1000 BE subs

 

i7-6700K/Windows 10  --> EVGA Nu Audio Card --> Focal CMS50's 

Link to comment
4 minutes ago, Dev said:

I think a much better machine is NUC8I7HVK

I'm really surprised to learn that this NUC manages temps so well.  Apparently, this NUC utilizes vapor chamber cooling alongside dual system fans.  

 

Pareto Audio AMD 7700 Server --> Berkeley Alpha USB --> Jeff Rowland Aeris --> Jeff Rowland 625 S2 --> Focal Utopia 3 Diablos with 2 x Focal Electra SW 1000 BE subs

 

i7-6700K/Windows 10  --> EVGA Nu Audio Card --> Focal CMS50's 

Link to comment
1 minute ago, rickca said:

I'm really surprised to learn that this NUC manages temps so well.  Apparently, this NUC utilizes vapor chamber cooling alongside dual system fans.  

 

 

yes, other reports from friends, it capable of dsd512 ext2 filters as well. The only reason I avoided in the first place is the noisy fan and I will have to keep the server in the audio room. Looks like there isn't any news yet for a fanless model on NUC8. So while I still have a chance, I might as well return the BEH and get the HVK.

Link to comment
15 minutes ago, Dev said:

I might as well return the BEH and get the HVK

Of course, now we're talking about a NUC that's a premium product approaching $1K list price.  I guess the discrete AMD graphics card drives up the cost.

Pareto Audio AMD 7700 Server --> Berkeley Alpha USB --> Jeff Rowland Aeris --> Jeff Rowland 625 S2 --> Focal Utopia 3 Diablos with 2 x Focal Electra SW 1000 BE subs

 

i7-6700K/Windows 10  --> EVGA Nu Audio Card --> Focal CMS50's 

Link to comment

@SuperdadAlex,

 

Couple of observations:

 

1. With JS-2 and monitor connected to the NUC, I hear a weird whooshing sound (ones similar to when you have bad tubes in the system) from both my speakers when my monitor is powered on and connected via the hdmi. If I disconnect the monitor power cord, they are gone. My speakers are dead silent with the stock smps. Any idea what might be going on ? Something to do with ground loops ?

 

2. When Roon is scanning a "new" music library over the network (which means lots of processing), I don't see the yellow led flickering but I do notice that the red led is constantly lit most of the time (blips every 2/3 secs).

Link to comment
20 minutes ago, rickca said:

Of course, now we're talking about a NUC that's a premium product approaching $1K list price.  I guess the discrete AMD graphics card drives up the cost.

 

yeah, they are meant to be compact gaming rig but no different in the upsampling requirement. Wish there was a way to take advantage of the discrete graphics.

Link to comment
2 hours ago, Dev said:

With JS-2 and monitor connected to the NUC

 

Hi Dev:

Clearly you are forming a ground loop external to the JS-2.  The JS-2’s -ve zero-volt output “grounds” are “floated” (though the chassis and transformer are grounded for safety).  So what is happening is that both your HDMI monitor and whatever other peripherals are attached to the NUC (USB, etc.) are establishing the earth-grounding for your NUC and interacting.

You could defeat our float/isolation by connecting to ground the barrel of either of the output jacks and your unique issue should go away.

 

As for the rear panel red LED blips, those seem mostly normal, though I do need to ask you two questions:

1) What is the actual AC wall voltage to the JS-2? (are you running off some conditioner that is dropping a few volts?). Would like to confirm that you are working with at least 120V.

2) You said this is a loaned JS-2.  How old is it? (you can tell me s/n or buyer name). I ask because about 3 years ago we moved the transformer secondary up to 21V—to deliver full 5A at 12V in 220V countries, with the byproduct that 120/240V performance jumped to 7.2A at 12V.  Earlier units with 18 or 19V trans secondary could deliver 5A/12V with 240/120V mains.

 

That’s all I can think of tonight.  Please contact me offline for further asssitance.

(And may I ask that you and Rick curb the off-topic NUC banter in this, the JS-2 thread?  Plenty of happy NUCing around elsewhere on CA. x-D)

 

Cheers,

—Alex C.

Link to comment
22 hours ago, Dev said:

 

With JS-2 and monitor connected to the NUC, I hear a weird whooshing sound (ones similar to when you have bad tubes in the system) from both my speakers when my monitor is powered on and connected via the hdmi. If I disconnect the monitor power cord, they are gone. My speakers are dead silent with the stock smps. Any idea what might be going on ? Something to do with ground loops ?

 

Just to close the loop (pardon the pun) on Dev's issue--since we chatted via PM about it--here is what was found and resolved:

Dev said:

I connect everything to a Synergistic Research Powercell 10UEF, except the Amp which connects directly but in the same phase where the Powercell is also connected. When I reported the problem earlier, the monitor was connected directly to a spare the wall outlet (same phase). When I re-connected the monitor to the Powercell outlet directly, I don't hear the buzz anymore. 

 

My reply to him included the following comments:

Clearly it seems you found the source of the buzz--some sort of loop with the monitor.  

As for why it does not occur with the ungrounded SMPS, I can only guess that the significant AC leakage coming though (from the wall, through the SMPS, and into the NUC) somehow conducts enough of your line neutral to be a diversion of the loop.

As mentioned you could test (and thereby test my theory) by grounding one of the DC outputs of the JS-2 and seeing what happens.  

I looked at photos of the inside of the Powercell 10UEF and there clearly is a lot of stuff increasing impedance between grounds, etc. so it is hard to know for sure what is happening.

Link to comment
  • 2 weeks later...

I have been using a JS-2 to power the dirty side of an Allo DigiOne Signature and a LPS-1.2 (that in turn powers the clean side of the same Signature) since about one month. I am very pleased with the results. However, there is an issue that is puzzling me.

 

Last weekend I did some re-cabling and realized that the phase of the JS-2 is just the opposite than the phase of all my other components: the Naim DAC, the Supernait 2, the Oppo 203 all have the phase at the lower socket pin. On Naim cables, this is marked with L. The JS-2 phase is just on the opposite pin, the one marked with N on Naim cables.

 

If one has a power strip this is not a problem: one can turn the Schuko plug around and have all components connected consistently. But with a power distribution block or hydra, there is no way to adjust for one load with inverted phase.

 

Is there a special reason why the JS-2 has a different phase than all my other components?  I have measured the phase of nDAC, SN2, JS-2, etc. using an Oehlbach phaser: http://www.oehlbach.com/en/power/phaser.

 

Thanks, nbpf

Link to comment
49 minutes ago, wanta911 said:

I'll let Alex answer but as far as I know the standard is to have the bottom pin (red arrow) live in the case of the JS-2 plug orientation. I would assume that's what it is?

 

 

 

 

JS-2_rear_master_1024x1024.jpg

Yes, that is what I would have expected but a phase detector clearly indicates the phase to be at the other (upper) pin, hence my surprise. For the SN2 and the Naim DAC, the same detector indicates the phase to be at the lower pin. The plug orientation is the same in these devices. 

Link to comment

What do you mean by phase in this respect? There is a switch, a fuse then the two sides connect to the primary winding of the transformer. What does "phase" mean in this arrangement? Neither side is connected to anything else other than the primary of the transformer.

 

How does it matter which pin gets connected to which side of the primary winding? Neither side of the jack gets connected to safety ground.

 

Can you provide a schematic of what you mean by phase? That link provided gives no indication what it is actually measuring.

 

John S.

Link to comment
4 minutes ago, JohnSwenson said:

What do you mean by phase in this respect? There is a switch, a fuse then the two sides connect to the primary winding of the transformer. What does "phase" mean in this arrangement? Neither side is connected to anything else other than the primary of the transformer.

 

How does it matter which pin gets connected to which side of the primary winding? Neither side of the jack gets connected to safety ground.

 

Can you provide a schematic of what you mean by phase? That link provided gives no indication what it is actually measuring.

 

John S.

Thanks for replying! Unfortunately I cannot provide you a schematic of the Oehlbach device or a precise specification of what it measures.

 

The purpose of the device is to indicate which pin of an audio device should ideally be connected to the live pin of a Schuko female socket of a power strip. The idea is that, for optimal sound quality, all devices connected to a power strip should be connected consistently.

 

As reported, the device indicates the lower pin (the one marked with L on Naim cables) for SN2, nDAC and Teddy Pardo power supplies and the upper pin for the JS-2. Hence my question.

 

No matter what the Oehlbach device does measure or indicate: what is your advice? Shall one connect the lower pin of the JS-2 to the phase (live), the upper pin or it just doesn't matter?

 

Thanks, nbpf

 

Link to comment
12 hours ago, nbpf said:

Yes, that is what I would have expected but a phase detector clearly indicates the phase to be at the other (upper) pin, hence my surprise. For the SN2 and the Naim DAC, the same detector indicates the phase to be at the lower pin. The plug orientation is the same in these devices. 

 

We did go over all this with you back in June (this and the 8 posts that follow it: https://www.computeraudiophile.com/forums/topic/20528-js-2-power-supply-installed/?do=findComment&comment=829340).

 

The JS-2 is wired to the IEC60320-C14 standard.

 

Not that any of this matters since the R-core transformer in the JS-2 is wired balanced on both the primary and secondary sides, and its DC outputs are “floated.”

(The chassis and the transformer are grounded for safety of course.)

 

So other than the fact that the fuse and switch are in-line only on one side of the AC mains line, the JS-2 is entirely symmetrical.

 

Link to comment
1 hour ago, Superdad said:

 

We did go over all this with you back in June (this and the 8 posts that follow it: https://www.computeraudiophile.com/forums/topic/20528-js-2-power-supply-installed/?do=findComment&comment=829340).

 

The JS-2 is wired to the IEC60320-C14 standard.

 

Not that any of this matters since the R-core transformer in the JS-2 is wired balanced on both the primary and secondary sides, and its DC outputs are “floated.”

(The chassis and the transformer are grounded for safety of course.)

 

So other than the fact that the fuse and switch are in-line only on one side of the AC mains line, the JS-2 is entirely symmetrical.

 

Thanks Alex! 

 

In June I had no JS-2 and thus my interest was pretty much theoretical. Now I have a JS-2 and I am planning to buy a distribution block. For this reason I have checked the polarity of all my components, included the JS-2.

 

With a distribution block, I will obviously not be able to ensure that each device is connected to the mains live through the pin indicated by the phase detector unless the detector indicates the same pin for all devices.

 

The JS-2 is currently the only device in my system for which the phase detector indicates that the upper pin should be connected to the mains live. For all other devices, it indicates the lower pin. Hence my question. 

 

I'm taking from your answer that it does not actually matter which pin of the JS-2 is connected to the mains live.  

Link to comment
18 minutes ago, nbpf said:

The JS-2 is currently the only device in my system for which the phase detector indicates that the upper pin should be connected to the mains live. For all other devices, it indicates the lower pin. Hence my question. 

No information on what magic tester try to detect.

No differences in JS-2 internal circuit, but switch & fuse.

So what to worry about ?

Maybe retry phase detector test with JS-2 power switch in On position, see if change  :)

Link to comment
Just now, look&listen said:

No information on what magic tester try to detect.

No differences in JS-2 internal circuit, but switch & fuse.

So what to worry about ?

Maybe retry phase detector test with JS-2 power switch in On position, see if change  :)

I do not think there is anything magic with the Oehlbach device. It is just a device that attempts at detecting which pin of a device should be connected to the mains live, the idea being that in an audio system it could be beneficial if all devices were connected consistently.

   

I have tried the phase detector with the JS-2 power switch off and on and, in the latter case with and without connected loads. In all these three configurations the phase detector indicates the upper pin of the JS-2 as the one that should be connected to the live. As reported, the same detector indicates the lower pin for all my other devices. Hence my surprise.

 

I am not worried but considering the cost of a distribution block, I want to be sure that it does not actually matter which pin of the JS-2 is connected to the mains live.  

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...