SMPS and grounding

[tboooe]

12 hours ago, JohnSwenson said:

See the post right above yours, it seems there are some power conditioners that do not work well with grounding scheme, I have no idea why, I don't have any of these to try out, and personally I don't really want to get into testing such things right now. If the grounding doesn't work with the power conditioner and it does work when plugged into a regular (non-conditioned) outlet, then listen to both ways:

 

non-grounded SMPS plugged into conditioner, powering the LPS-1

grounded SMPS into regular outlet, powering LPS-1

 

Whichever sounds best, go with that.

 

John S.

Thank you @JohnSwenson...here is more strangeness...I tested the plug on the DC terminal adapter for continuity and verified with my digital multimeter that the output was seeing 7.58VDC from the Meanwell that came with the LPS-1.  I plugged the ground plug into the wall outlet and the LPS-1 still would not turn on.  I then decided to just rewire the DC terminal adapters.  Again I verified with the meter everything was working properly.  I plugged it into my P-3 AC regenerator and this time it worked.  A few hours later I notice the LPS-1 was again not powered on.  I unplugged and replugged everything which "fixed" the issue.  Checked this morning and everything seems to working ok.  

 

Not sure whats going on but for now its working.

[H-Man]

Reading about this tweak I had to test it since I had all the parts lying around so for a total cost of less than 5 USD I put this together:

 

 

 

I have have been using the mentioned Netgear switches for years so I connected to the Switch and LPS-1 powering my UltraRendu. I am very surprised at how noticeable this tweak is... and with the forks it has been very easy to A/B test. The only downside is how ugly it is... This should obviously be standard going forward. 

 

While the multimeter was was out I measured the internal PS on the Brooklyn DAC and that was already grounded. 

[oneguy]

For those looking for a more elegant solution you could your LPS-1 with something like this: 

 

https://rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.com%2Fulk%2Fitm%2F122625352090

 

I know it may be considered overkill to use a linear power supply to power the LPS-1 but it would do the job cleanly without having to add any taps or grounds. As John has stated, normal linear power supply only let through low impedance noise and the LPS-1 doesn’t let any low impedance noise so the two in series equates to no low or high impedance noise getting through. 

 

It may may not be as cheap a solution as SMPS grounding but you could run a Y cable from one of the LPS outputs to power a second LPS-1 (they can make a 7.5V 5A output section) leaving the two other outputs to run whatever you want. This at least makes the cost more manageable. 

 

I am considering it as a future upgrade because between my two LPS-1’s Mean Wells, the NetGear switch and the grounding plugs that’s 4 electrical outlets that are taken up. This combined unit would allow me to go down to 1. 

 

Anyway, my 2 cents on a unit I found and an idea I had. 

[Cornan]

32 minutes ago, H-Man said:

While the multimeter was was out I measured the internal PS on the Brooklyn DAC and that was already grounded.

 

Thanks! ? I was wondering if the Brooklyn DACs internal SMPS was grounded or not? Now I do not need to open it up and check myself.

[H-Man]

I am a bit lost for words... I have been part of the camp that has said that ethernet does not have any sound... Now I have tested this tweak on the switch I have in front of my Kef LS50W and this tweak impacts the sound... To be fair this is about the SMPS powering the Ethernet switch but still I would have considered this “snake-oil” before I tested. I need some time before passing final word and getting rid of any kind of confirmation bias.

[thyname]

What is better in terms of powering the "approved" ethernet switch?

 

1 - Apply the grounding technique from John S to the SMPS output that power the switch

 

2 - Use a linear power supply to power the Switch

[JohnSwenson]

first off if you use a linear supply, the output still has to be grounded. It is the grounded power supply that allows the switch to shunt the leakage from other devices.

 

In general a decent linear supply is always preferred over an SMPS, but they can cost a lot more. In situations like the switch the LPS is probably not going to be much of a difference sound wise so spending hundreds of dollars on an LPS VS $12 for an SMPS may not be a  good tradeoff. But you still have to look into whether the output of the LPS is grounded. If it is NOT grounded you need to use one of the grounding schemes to ground its output in order to shunt the leakage from other devices.

 

Just as an example in MY system using one of approved switches with a grounded SMPS made a significant difference in SQ VS  using the SMPS without grounding. Using an un-grounded very good LPS sounded worse than the grounded SMPS, the grounded LPS was no better sounding than the grounded SMPS. Thus there was no reason to go with the LPS FOR THE SWITCH, this does NOT mean I'm saying that LPS are useless. For DACs, streamers etc the LPS was always much better than the SMPS.

 

For your system a grounded LPS may sound slightly better, but the only way to tell is try it yourself.

 

John S.

[JohnSwenson]

1 hour ago, oneguy said:

For those looking for a more elegant solution you could your LPS-1 with something like this: 

 

https://rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.com%2Fulk%2Fitm%2F122625352090

 

I know it may be considered overkill to use a linear power supply to power the LPS-1 but it would do the job cleanly without having to add any taps or grounds. As John has stated, normal linear power supply only let through low impedance noise and the LPS-1 doesn’t let any low impedance noise so the two in series equates to no low or high impedance noise getting through. 

 

It may may not be as cheap a solution as SMPS grounding but you could run a Y cable from one of the LPS outputs to power a second LPS-1 (they can make a 7.5V 5A output section) leaving the two other outputs to run whatever you want. This at least makes the cost more manageable. 

 

I am considering it as a future upgrade because between my two LPS-1’s Mean Wells, the NetGear switch and the grounding plugs that’s 4 electrical outlets that are taken up. This combined unit would allow me to go down to 1. 

 

Anyway, my 2 cents on a unit I found and an idea I had. 

This PS looks very nice. The 9V and 12V outputs can each power an LPS-1 (IF the ratings are true). The LPS-1 is not designed to work from a 5V input so should not be used with the 5V output.

 

This looks like a very well built robust supply, it probably has enough juice to power a couple LPS-1s. Some customers have found that some inexpensive LPS have specs that are not true, the voltage sags when powering an LPS-1. But from just looking at it, it looks like this supply will not do that and should power LPS-1s just fine. But you really don't know until it is tried out.

 

John S.

[thyname]

16 minutes ago, JohnSwenson said:

first off if you use a linear supply, the output still has to be grounded.

 

Thanks! It makes sense, good advice as always.

 

Did you mean grounding the output of the SMPS that powers the linear power supply?

[oneguy]

22 minutes ago, JohnSwenson said:

This PS looks very nice. The 9V and 12V outputs can each power an LPS-1 (IF the ratings are true). The LPS-1 is not designed to work from a 5V input so should not be used with the 5V output.

 

This looks like a very well built robust supply, it probably has enough juice to power a couple LPS-1s. Some customers have found that some inexpensive LPS have specs that are not true, the voltage sags when powering an LPS-1. But from just looking at it, it looks like this supply will not do that and should power LPS-1s just fine. But you really don't know until it is tried out.

 

John S.

The outputs are customizable to customer needs. Back when this thread was developing I once asked the if it were possible to make a 48V 1.25A output to power my PoE switch (which used to directly connect to my uR). It was possible and the charge was an extra $24-25. 

 

Thanks for the the comments on an LPS still needing to be grounded for the NetGear switch to do its job. I currently have mine plugged into an LH Labs LPS4 since as a matter of convenience since I was only using 2 of the 5 outputs. I’ll try to find out if the output negative is grounded and if not then ground it. 

[charlesphoto]

You may want to check out the HDPLEX 200w new version. I have the 100w and like it as it’s so flexible, esp with the 5-19v adjustable rail. The new one looks even better. Also much more resale value and flexibility, plus you could power a server with it if need be. Not sure what shipping would be, (they have a warehouse in LA and Germany if I’m correct) but you’re only $75 more than the Chinese one with it’s $60 shipping. My two cents. 

 

https://www.hdplex.com/hdplex-fanless-200w-linear-power-supply-for-pc-audio-and-ce-device.html

 

[JohnSwenson]

1 hour ago, thyname said:

 

Thanks! It makes sense, good advice as always.

 

Did you mean grounding the output of the SMPS that powers the linear power supply?

No,

if you use an LPS with the switch, the output of the LPS has to be grounded, either inside the LPS, or with a separate grounding device as shown in this thread.

 

You use EITHER a grounded SMPS (again either internal or with separate grounding device) OR a grounded LPS.

 

The only time you use one power thingy driving another is if you use an LPS-1. It has to be supplied DC power from some other DC power supply.

 

John S.

[JohnSwenson]

1 hour ago, charlesphoto said:

You may want to check out the HDPLEX 200w new version. I have the 100w and like it as it’s so flexible, esp with the 5-19v adjustable rail. The new one looks even better. Also much more resale value and flexibility, plus you could power a server with it if need be. Not sure what shipping would be, (they have a warehouse in LA and Germany if I’m correct) but you’re only $75 more than the Chinese one with it’s $60 shipping. My two cents. 

 

https://www.hdplex.com/hdplex-fanless-200w-linear-power-supply-for-pc-audio-and-ce-device.html

 

You should be able to drive 5 LPS-1s from that supply (4 off the 12V and 1 off the adjustable set fro 9V or so).

 

It explicitly says the output grounds are not connected to AC ground or to each other, thus is you want to use one of the outputs to drive one of approved switches you must ground the output that is powering the switch, the others do not have to grounded.

 

John S.

[auricgoldfinger]

I think what they mean is each rail is grounded separately from the others.

[Superdad]

1 hour ago, charlesphoto said:

You may want to check out the HDPLEX 200w new version. I have the 100w and like it as it’s so flexible, esp with the 5-19v adjustable rail.

 

24 minutes ago, JohnSwenson said:

You should be able to drive 5 LPS-1s from that supply (4 off the 12V and 1 off the adjustable set fro 9V or so).

 

The UltraCap LPS-1 does not tolerate "energizing"/charging input above 12V.  Even a short spike above 12V can fry one of its input regulators--turning the board into scrap.  Four of the six destroyed LPS-1 boards we have in a box here came from people using an HDPlex power supply. 

Now I do not know the vintage of the offending HDPlex units owned by those who have suffered, and while HDPlex seems to offer good value (and I know they offer good support), at this time we must advise against using that brand as a charger for our LPS-1.  Perhaps one set at 9V would be safe, but that is no guarantee. 

 

It is easy for us to inspect a failed LPS-1 board to determine if an over-voltage killed it:

 

 

Such is not covered by our warranty.

 

As this thread--and the graphs that we published made clear--there is no advantage to powering an LPS-1 with a linear power supply versus the Mean Well we include, once the Mean Well's output is shunted.

 

 

I am still trying to make arrangements with a quality SMPS manufacturer to supply us with units that already have their zero-volt output ("ground") tied to the AC mains ground pin.  There are some that do so already, but we have multiple factors to consider in making our choice before I make the required 1,000 unit stock purchase.

 

Thank you all for your enthusiasm, patience, and understanding.

--Alex C.

[Superdad]

7 minutes ago, auricgoldfinger said:

I think what they mean is each rail is grounded separately from the others.

 

No, it can not be both ways.  Either all the rails share a common ground, or they are all floated from mains ground.

[auricgoldfinger]

2 minutes ago, Superdad said:

 

No, it can not be both ways.  Either all the rails share a common ground, or they are all floated from mains ground.

 

Thanks for the clarification. 

[tboooe]

50 minutes ago, Superdad said:

 

No, it can not be both ways.  Either all the rails share a common ground, or they are all floated from mains ground.

So now I am a bit confused.   I thought it didnt make sense to ground the output of an LPS because given how they work there is no high freq noise to deal with with.

[Superdad]

14 minutes ago, tboooe said:

So now I am a bit confused.   I thought it didnt make sense to ground the output of an LPS because given how they work there is no high freq noise to deal with with.

 

a) It's not high frequency noise, it is is high-impedance common-mode leakage (which has both high and low freq. components);

 

b) You are correct that grounding the output of a linear supply is not needed--except with regards to the specific Netgear switches that John tested, for which grounding whatever power supply is used allows that Ethernet switch to shunt leakage currents coming in on the CAT cable from other upstream gear.  So it is not so much that you are grounding the PS for this switch, just that it is the easiest place to ground those unique switches (and which if you are using an SMPS for it--just fine--will also take care of ITS high-impedance leakage).

[tboooe]

9 minutes ago, Superdad said:

 

a) It's not high frequency noise, it is is high-impedance common-mode leakage (which has both high and low freq. components);

 

I knew I should have paid more attention in my EE courses is college.  Thank you sir for the explanation.

[austinpop]

2 hours ago, JohnSwenson said:

one power thingy driving another

 

Ah, engineers and their technical jargon!

[austinpop]

@Superdad

 

Once you get Meanwell to start internally grounding their SMPSes for you, will you have them change their name to DoWell?

[lmitche]

3 minutes ago, austinpop said:

@Superdad

 

Once you get Meanwell to start internally grounding their SMPSes for you, will you have them change their name to DoWell?

How about GroundsWell?

[sandyk]

2 hours ago, Superdad said:

The UltraCap LPS-1 does not tolerate "energizing"/charging input above 12V.  Even a short spike above 12V can fry one of its input regulators--turning the board into scrap.  Four of the six destroyed LPS-1 boards we have in a box here came from people using an HDPlex power supply. 

 Hi Alex

 I have a friend with a 300W HDPlex PSU that also owns an LPS-1

 I have forwarded this post to him and asked if he is able to check the actual output voltage of it's 12V rail.

IIRC, he also had a module fail some time back, and that later revisions provided some protection.

Do you know from these customers if their HDPLex PSU modules failed at the same time, with perhaps a series voltage regulator failure resulting in full unregulated output voltage to the load ?

 I believe that this may have happened.

 

Kind Regards

Alex K

[austinpop]

2 minutes ago, lmitche said:

How about GroundsWell?

 

Jealous I didn't think of that.   It's brilliant!