SMPS and grounding

[One and a half]

Just to reinforce grounding of 0V connections to PE with a delivery of an Acopian Linear supply to replace and iPower unit 9V, 1.5A driving ifi Micro USB 3.0.  

 

The Acopian is physically 12 times larger than the iPower, and grounding is easy to connect via screw terminals via a short jumper. The iPower is your typical SMPS despite what ifi claim, it's full of noise, don't know what sort, but heaving it and replacing with the Acopian removed grunge. Immediately, there's micro details around instruments, there's placement of individual instruments now possible, still not on par with SACD player, but getting closer. Music is enjoyable again, the 'what can I play next' itching to find and explore.

 

It's great see a drawer full of discarded wall warts, just one more SMPS powered device on the audio power chain remaining. Another 6 weeks to wait for the final chapter.

[austinpop]

@JohnSwenson

 

I just had a strange episode with my home brew JSGT.

 

I had need to power cycle my LPS-1. The output of the Meanwell goes into the input of the JSGT, and the other end plugs into the energizing input port of the LPS-1. As it should. Well, I noticed the Meanwell green LED was cycling on and off at about 2Hz, with audible clicks to match, and there was no red light on the LPS-1. Disconnecting the Meanwell from the JSGT immediately restored the LED to steady green, and connecting it directly immediately energized the LPS-1.

 

I immediately disconnected the JSGT, and I'm trying to figure out what is going on. This was all working immediately prior to the power cycle. Luckily nothing appears to be fried or anything.

 

Could I be creating a ground loop of some kind that the Meanwell is detecting and reacting to?

[wakka992]

On 25/1/2018 at 9:19 AM, JohnSwenson said:

... ... ...

My tests on this with an LPS-1 do in fact show this. An internal shunted SMPS feeding an LPS-1 does in fact show leakage output right near the noise floor, but the same SMPS with the external adapter is not quite as good, much better than without the LPS-1, but not as good as the internal shunted supply.

... ... ...

 

John S.

Dear @JohnSwenson

I'm about to receive my LPS-1 and I was looking for a SMPS already internally shunted instead of going with the external adapter route as per your quoted message.

 

I think I've found a Meanwell unit that should do the trick, but I can't figure it out if what I'm reading from the datasheet is correct or not.

 

The Meanwell I'm speaking about are the AC-DC medical power supply from GSM40A and GSM60A series.

From the figures I think they are grounded/earthed at the "outside" of the Standard DC plug P1J.

From the datasheet PDF of this power supply I founded the following added under the Standard plug P1J pinout scheme: "-V connected to AC FG".

Please find attached the .pdf datasheet with the quote I'm mentioning at the bottom of page 3.

 

Thanks for your assistance!

GSM40A-SPEC.PDF

[Superdad]

2 hours ago, wakka992 said:

I'm about to receive my LPS-1 and I was looking for a SMPS already internally shunted instead of going with the external adapter route as per your quoted message.

 

I am assuming that you are getting a second-hand LPS-1 since the last production of them was sold out by us months ago and the enhanced UltraCap LPS-1.2 begins shipping on the 15th.  It comes with a "energizer"/charger that is already internally "ground-shunted."

 

2 hours ago, wakka992 said:

I think I've found a Meanwell unit that should do the trick, but I can't figure it out if what I'm reading from the datasheet is correct or not.

 

The Meanwell I'm speaking about are the AC-DC medical power supply from GSM40A and GSM60A series.

From the figures I think they are grounded/earthed at the "outside" of the Standard DC plug P1J.

 

Having checked a bunch of Mean Well units, including a couple of the their "medical" ones, I can tell you that the odds of the one you are looking at being designed with its DC zero-volt output "ground" common to the AC inlet ground are very slim.  For whatever reason, Mean Well does not seem to do this with any of their units.  I tried for months to get an answer from the factory about this--and to get them to make us 1,000 units with the change--and got nowhere.  That's why we moved to a different supplier for the LPS-1.2 charger.  Even then, the shunting is not something that is found in the spec sheets--we had to buy lots of samples to find one.

 

After the first 250 LPS-1.2 orders are shipped, we will begin offering our 7.5V/4.8A/36W adapter separately for $15 to those who want one.  Until then, there is a CUI brand 12V/2.5A/30W unit available in the US from DigiKey and other distributors (https://www.digikey.com/products/en?keywords=102-4104-ND.  This is one that I personally confirmed connects AC ground to DC output "ground." 

 

Hope this helps.

--Alex C.

 

[JohnSwenson]

17 hours ago, austinpop said:

@JohnSwenson

 

I just had a strange episode with my home brew JSGT.

 

I had need to power cycle my LPS-1. The output of the Meanwell goes into the input of the JSGT, and the other end plugs into the energizing input port of the LPS-1. As it should. Well, I noticed the Meanwell green LED was cycling on and off at about 2Hz, with audible clicks to match, and there was no red light on the LPS-1. Disconnecting the Meanwell from the JSGT immediately restored the LED to steady green, and connecting it directly immediately energized the LPS-1.

 

I immediately disconnected the JSGT, and I'm trying to figure out what is going on. This was all working immediately prior to the power cycle. Luckily nothing appears to be fried or anything.

 

Could I be creating a ground loop of some kind that the Meanwell is detecting and reacting to?

Just to clarify, the Mean Well was connected to the adapter, then the adapter was connected to the LPS-1? Did the LED on the LPS-1 ever come on? Did it come on red then go off? Did it go read, then amber then sometime during amber go to red? Did it flashing red? Something else?

 

The sequence of events is very critical to trying to figuring what is going on.

 

Thanks,

 

John S.

[wakka992]

1 hour ago, Superdad said:

 

I am assuming that you are getting a second-hand LPS-1 since the last production of them was sold out by us months ago and the enhanced UltraCap LPS-1.2 begins shipping on the 15th.  It comes with a "energizer"/charger that is already internally "ground-shunted."

 

 

Having checked a bunch of Mean Well units, including a couple of the their "medical" ones, I can tell you that the odds of the one you are looking at being designed with its DC zero-volt output "ground" common to the AC inlet ground are very slim.  For whatever reason, Mean Well does not seem to do this with any of their units.  I tried for months to get an answer from the factory about this--and to get them to make us 1,000 units with the change--and got nowhere.  That's why we moved to a different supplier for the LPS-1.2 charger.  Even then, the shunting is not something that is found in the spec sheets--we had to buy lots of samples to find one.

 

After the first 250 LPS-1.2 orders are shipped, we will begin offering our 7.5V/4.8A/36W adapter separately for $15 to those who want one.  Until then, there is a CUI brand 12V/2.5A/30W unit available in the US from DigiKey and other distributors (https://www.digikey.com/products/en?keywords=102-4104-ND.  This is one that I personally confirmed connects AC ground to DC output "ground." 

 

Hope this helps.

--Alex C.

 

Hi Alex,

exactly, I'm buying a used ISO Regen LPS-1 combo.

 

Thanks for the tip on the CUI SMPS.

I'll wait for your shunted power supply to be avayable (I've found it from shop but the minimum order is for 500units XD); in the meantime I'll externally shunt the stock Meanwell one.

 

Regarding LPS-1 and SMPS grounding, for what I know in general you should have only one earth in the entire system, to avoid ground loop.

 

If that is the case, since my DAC is already powered by a shunted SMPS, adding a second shunted SMPS for the LPS-1 could create ground loop OR the LPS-1 isolates the ground at the input from the output (in addition to isolate the output from the AC)?

 

Thanks for your assistance!

[wakka992]

1 hour ago, wakka992 said:

Hi Alex,

exactly, I'm buying a used ISO Regen LPS-1 combo.

 

Thanks for the tip on the CUI SMPS.

I'll wait for your shunted power supply to be avayable (I've found it from shop but the minimum order is for 500units XD); in the meantime I'll externally shunt the stock Meanwell one.

 

Regarding LPS-1 and SMPS grounding, for what I know in general you should have only one earth in the entire system, to avoid ground loop.

 

If that is the case, since my DAC is already powered by a shunted SMPS, adding a second shunted SMPS for the LPS-1 could create ground loop OR the LPS-1 isolates the ground at the input from the output (in addition to isolate the output from the AC)?

 

Thanks for your assistance!

 

On the light of the following lines from @JohnSwenson to my question I add: should I ground both SMPS input to LPS-1 AND LPS-1 output to ISO Regen? 

 

On 17/10/2017 at 10:22 PM, JohnSwenson said:

"... ... ...

IF the LPS-1 is driving an ISO REGEN there can be a situation where the whole audio system is floating with respect to earth ground and a charge can build up which can show up as clicks and pops. ONE earth ground in such a system can alleviate this. ONE way to do this is to ground the negative of the supply powering the ISO REGEN. If this supply is an LPS-1 then you can try grounding the output of the LPS-1 to see if it fixes the clicks and pops.

 

Grounding the INPUT to the LPS-1 can help in other situations by shunting the high impedance leakage."

 

I want to both "shunt the high impedance leakage" and "prevent system floating with respect to earth ground" 

[austinpop]

3 hours ago, JohnSwenson said:

Just to clarify, the Mean Well was connected to the adapter, then the adapter was connected to the LPS-1?

 

Yes, the output barrel of the Meanwell was connected to the female end of the JSGT, and the male end of the JSGT connected to the LPS-1.

 

3 hours ago, JohnSwenson said:

 

Did the LED on the LPS-1 ever come on? Did it come on red then go off? Did it go read, then amber then sometime during amber go to red? Did it flashing red? Something else?

 

Nothing whatsoever happened on the LPS-1. In fact the behavior was the same even with the JSGT detached from the LPS-1. Something about connecting the Meanwell output to the JSGT caused the Meanwell to behave in that fashion - the green LED on the Meanwell brick was flashing. Detaching from the JSGT immediately fixed it.

[JohnSwenson]

4 hours ago, wakka992 said:

 

On the light of the following lines from @JohnSwenson to my question I add: should I ground both SMPS input to LPS-1 AND LPS-1 output to ISO Regen? 

 

 

I want to both "shunt the high impedance leakage" and "prevent system floating with respect to earth ground" 

You just want to shunt the SMPS feeding the LPS-1, the LPS-1 will isolate its output from the ground of the feeder. So in your case, shunt the feeder to the LPS-1 and DON'T ground the negative of the LPS-1 output.

 

John S.

[wakka992]

6 hours ago, JohnSwenson said:

You just want to shunt the SMPS feeding the LPS-1, the LPS-1 will isolate its output from the ground of the feeder. So in your case, shunt the feeder to the LPS-1 and DON'T ground the negative of the LPS-1 output.

 

John S.

thank you very much for your assistance John!

can't wait for my LPS-1 and ISO Regen to arrive 

 

BTW I think I think I discovered what Meanwell means with that "-V connected to AC FG" on the datasheet --> it shoul be "external of the DC barrel conected to the AC Frame Ground".

 

Since I'm not an expert on this, @JohnSwenson @Superdad does this make any sense to you?

does that mean the Meanwell GSM40A and GSM60A I posted about earlier are shunted?

 

thanks as always

[asdf1000]

On 29/10/2017 at 6:46 AM, JohnSwenson said:

Theoretically you could do the same with USB, BUT USB is not just AC, it requires DC connectivity through the data pair, so a transformer will not work. This has made series blocking very difficult to deal with. There are a few solutions, but none of them block the high impedance components, so you still need to shunt the all the high impedance source before they get to the USB cable if you want to stop ALL the leakage from getting through to a DAC. 

 

Hi @JohnSwenson

 

This thread has mostly been about shunting high impedance leakage currents for network switches and the LPS-1 (proven to be very helpful in my system, so thanks).

 

Does the ISO REGEN block both low and high impedance leakage currents?

 

Or do you recommend this high impedance shunting with the ISO REGEN too, by grounding the USB source?

 

For example if I have a powered USB hub which feeds the ISO REGEN, then grounding one of the hub's USB ports should shunt the high impedance leakage to ground, away from entering the ISO REGEN? And this would be completely separate to grounding the LPS-1 input of course, with a separate purpose.

 

Or not required with the ISO REGEN?

 

Cheers

[Superdad]

11 hours ago, wakka992 said:

BTW I think I think I discovered what Meanwell means with that "-V connected to AC FG" on the datasheet --> it shoul be "external of the DC barrel conected to the AC Frame Ground".  Does that mean the Meanwell GSM40A and GSM60A I posted about earlier are shunted?

 

Good catch.  It might mean that.  You would have to buy one and check it with a meter.

[JohnSwenson]

17 hours ago, Em2016 said:

 

Hi @JohnSwenson

 

This thread has mostly been about shunting high impedance leakage currents for network switches and the LPS-1 (proven to be very helpful in my system, so thanks).

 

Does the ISO REGEN block both low and high impedance leakage currents?

 

Or do you recommend this high impedance shunting with the ISO REGEN too, by grounding the USB source?

 

For example if I have a powered USB hub which feeds the ISO REGEN, then grounding one of the hub's USB ports should shunt the high impedance leakage to ground, away from entering the ISO REGEN? And this would be completely separate to grounding the LPS-1 input of course, with a separate purpose.

 

Or not required with the ISO REGEN?

 

Cheers

The ISO REGEN has the same issue, it only blocks low impedance leakage. That means it is effective if whatever drives it does not have any high impedance leakage going through its USB output. I have never tried shunting high impedance leakage through the USB ground connection, it MAY work, but it may not.

 

I have just followed the practice of making sure that whatever is feeding the USB going into the ISO REGEN has had its SMPS negative output grounded. That works for sure, trying to do it with the USB ground may or may not work.

 

John S.

[asdf1000]

2 minutes ago, JohnSwenson said:

I have never tried shunting high impedance leakage through the USB ground connection, it MAY work, but it may not.

 

I have just followed the practice of making sure that whatever is feeding the USB going into the ISO REGEN has had its SMPS negative output grounded. That works for sure, trying to do it with the USB ground may or may not work.

 

Got it. Thanks John.

 

[pas]

On 2/5/2018 at 11:38 AM, Superdad said:

 

I am assuming that you are getting a second-hand LPS-1 since the last production of them was sold out by us months ago and the enhanced UltraCap LPS-1.2 begins shipping on the 15th.  It comes with a "energizer"/charger that is already internally "ground-shunted."

 

 

Having checked a bunch of Mean Well units, including a couple of the their "medical" ones, I can tell you that the odds of the one you are looking at being designed with its DC zero-volt output "ground" common to the AC inlet ground are very slim.  For whatever reason, Mean Well does not seem to do this with any of their units.  I tried for months to get an answer from the factory about this--and to get them to make us 1,000 units with the change--and got nowhere.  That's why we moved to a different supplier for the LPS-1.2 charger.  Even then, the shunting is not something that is found in the spec sheets--we had to buy lots of samples to find one.

 

After the first 250 LPS-1.2 orders are shipped, we will begin offering our 7.5V/4.8A/36W adapter separately for $15 to those who want one.  Until then, there is a CUI brand 12V/2.5A/30W unit available in the US from DigiKey and other distributors (https://www.digikey.com/products/en?keywords=102-4104-ND.  This is one that I personally confirmed connects AC ground to DC output "ground." 

 

Hope this helps.

--Alex C.

 

Thanks for the CUI recommendation, functions and sounds great. Quite a significant improvement.

[sockpit]

This may have been asked above.  I skimmed through the many pages and did not see it, though.

 

If one wanted to avoid using a separate outlet for the shunt, could one simply run the ground wire back to the ground pin of the stock Meanwell that comes with the LPS-1 (and is not in fact being used by that SMPS)?  

 

I have in mind soldering it to the pin at its very base, or using a connector (attached) that slides tightly over the ground pin.

[Superdad]

38 minutes ago, sockpit said:

This may have been asked above.  I skimmed through the many pages and did not see it, though.

 

If one wanted to avoid using a separate outlet for the shunt, could one simply run the ground wire back to the ground pin of the stock Meanwell that comes with the LPS-1 (and is not in fact being used by that SMPS)?  

 

I have in mind soldering it to the pin at its very base, or using a connector (attached) that slides tightly over the ground pin.

 

Yes, that would work fine! 

[sockpit]

43 minutes ago, Superdad said:

 

Yes, that would work fine! 

Ok, so assuming you can get a good connection, could one do the same with the DC barrel of the meanwell and dispense with the male-female screw terminal extras?  It’s the outer barrel of the Meanwell dc output I want to ground, right?

 

strikes me as cheaper and more elegant solution . . . 

[Superdad]

37 minutes ago, sockpit said:

Ok, so assuming you can get a good connection, could one do the same with the DC barrel of the meanwell and dispense with the male-female screw terminal extras?  It’s the outer barrel of the Meanwell dc output I want to ground, right?

 

strikes me as cheaper and more elegant solution . . . 

 

Yes, that will work as well.

[sockpit]

?   Thanks.  It took 36 pages for this idea to spring up.  Strange.  ?

[MikeyFresh]

2 hours ago, sockpit said:

?   Thanks.  It took 36 pages for this idea to spring up.  Strange.  ?

 

For the IEC side perhaps, I don't remember anyone doing it that way and I'm curious how you implement it.

 

On the DC barrel end it has definitely been done by a few members, myself included:

 

[sockpit]

Yep.  That's what I had in mind on both sides, if possible.  Thanks.

 

(Did you really hear a difference?)

[MikeyFresh]

Just now, sockpit said:

Yep.  That's what I had in mind on both sides, if possible.  Thanks.

 

(Did you really hear a difference?)

 

You mean a difference with vs. without the ground shunt I presume?

 

If so the answer is yes and it was much more than I was expecting, suggesting at least in my particular set-up there was a leakage loop or some AC making it through to the DAC affecting it's performance.

 

If you meant did I hear a difference in using a ring terminal vs. the DC barrel adapters I can't answer that question as I've never actually made one with the adapters, only the ring terminals, so nothing for me to compare there.

 

 

[sockpit]

11 minutes ago, MikeyFresh said:

 

You mean a difference with vs. without the ground shunt I presume?

 

If so the answer is yes and it was much more than I was expecting, suggesting at least in my particular set-up there was a leakage loop or some AC making it through to the DAC affecting it's performance.

 

If you meant did I hear a difference in using a ring terminal vs. the DC barrel adapters I can't answer that question as I've never actually made one with the adapters, only the ring terminals, so nothing for me to compare there.

 

 

I meant the first.  Thanks, will try it!

[MikeyFresh]

27 minutes ago, sockpit said:

Thanks, will try it!

 

Post pictures!