Hello, everyone.

I have a power supply that outputs 24V from 230V (with earth on the primary) and powers a user.
The problem is that the output contains a leakage current of 60-80uA (measured with a Fluke esa620 on the negative output) that I would like to reduce/eliminate.

How can it be eliminated?
When I think of a filter, I always think of interference on some frequency... is this also the case? Should I guess what frequency those uA are at?
Or something else?

Let's suppose I can't work on insulation and various shielding... but only add electronics (if any) to attenuate it...
And what about ferrite bead? (which already is present in the design, but I cannot recall its values at the moment)

 

What problem is this small leakage causing?
It's certainly not an electrocution hazard.

 

Hello, everyone.

I have a power supply that outputs 24V from 230V (with earth on the primary) and powers a user.
The problem is that the output contains a leakage current of 60-80uA (measured with a Fluke esa620 on the negative output) that I would like to reduce/eliminate.

How can it be eliminated?
When I think of a filter, I always think of interference on some frequency... is this also the case? Should I guess what frequency those uA are at?
Or something else?

Let's suppose I can't work on insulation and various shielding... but only add electronics (if any) to attenuate it...
And what about ferrite bead? (which already is present in the design, but I cannot recall its values at the moment)

Your leakage is likely normal SMPS parasitic leakage via Y-capacitors. Ferrites help high-frequency noise, not mains leakage.
You can shunt it to earth with a high-value resistor if you need it “gone” from the output. Otherwise, it’s safe and can often be ignored.

 

Your leakage is likely normal SMPS parasitic leakage via Y-capacitors. Ferrites help high-frequency noise, not mains leakage.
You can shunt it to earth with a high-value resistor if you need it “gone” from the output. Otherwise, it’s safe and can often be ignored.

What problem is this small leakage causing?
It's certainly not an electrocution hazard.

Thanks for the reply.

I use a power supply with a grounded primary and the "-" output of the secondary shorted to the primary ground (you can see it with a tester).
I also don't understand why the leakage is DC and not AC (the opposite is true if I use an identical/similar power supply without a ground).

It's not so much a question of leakage values.. but more of a theoretical and design question that I would like to understand.

 

How are you measuring this? Where are the probes? Did you break some connection ti connect an ammeter?

 

How are you measuring this? Where are the probes? Did you break some connection ti connect an ammeter?

I'm measuring the Enclosure leakage current using a prove connected to the 2-WIRE V/Ω/A (as stated in the technical datasheet)

 

It will be bulky and heavy, but could you try an isolation transformer in front of the supply?
Medical grade devices may help you reduce it to 10uA or less.

 

It will be bulky and heavy, but could you try an isolation transformer in front of the supply?
Medical grade devices may help you reduce it to 10uA or less.

The strange thing is that the power supply produces less than 80uADC... (limit imposed in the datasheet) but in some parts of the device I read more than 150uADC.
Is this "amplification" somewhere plausible?

 

Stranger still is that the leakage is DC, because it is usually AC, caused by the Y caps as mentioned previously by other poster, or transformer interwinding capacitance.

 

Stranger still is that the leakage is DC, because it is usually AC, caused by the Y caps as mentioned previously by other poster, or transformer interwinding capacitance.

Honestly, I don't know if there's a Y capacitor... I don't have the schematic and I can't open it.
But the negative output is definitely grounded because I can see it with the tester.
Some time ago I had an identical power supply (of the same brand) but without the ground earth on the primary side and the leakage was in AC
In fact, one thing I don't understand is how, in theory, a power supply with earth produces ADC leakage and one without produces AAC (same voltage, current, and power specifications).

 

Honestly, I don't know if there's a Y capacitor... I don't have the schematic and I can't open it.
But the negative output is definitely grounded because I can see it with the tester.
Some time ago I had an identical power supply (of the same brand) but without the ground and the leakage was in AC

are you convinced on the measurement?
you say it changes when you change things
all very strange.
try wearing an anti static strap, turning off the lights ,
can you share picture of you doing the test, showing the environment please.

 

are you convinced on the measurement?
you say it changes when you change things
all very strange.
try wearing an anti static strap, turning off the lights ,
can you share picture of you doing the test, showing the environment please.

Sorry! I made a mistake in my last comment... I'll update it.

 

This is confusing indeed!!
My interpretation is that the leakage current is DC that would flow from the negative output terminal of the 24 volt supply to the mains "safety green wire" ground connection. That makes no sense to me, if this supply is actually isolated. an ISOLATED power supply DOES NOT include any current path between any portion of the output and "Ground", whatever that is cosen to be.
Please describe the whole path of this leakage current

 

if this supply is actually isolated. an ISOLATED power supply DOES NOT include any current path between

But the negative output is definitely grounded because I can see it with the tester.

 

Now we have a problem because it is seems that the complaint is that a current is flowing between the power supply negative output terminal and the mains side "green wire earth ground" connection.
This tends to conflict with the statement from post #1: " I use a power supply with a grounded primary and the "-" output of the secondary shorted to the primary ground "!!!
IF the output negative IS ACTUALLY CONNECTED to that "green wire earth ground" by a very low resistance, then OBVIOUSLY there is an intentional connection, OR AN INTERNAL DEFECT in that particular power supply.
Since the claim is " But the negative output is definitely grounded because I can see it with the tester.", indicating some sort of internal connection, what remains is the option of disconnecting that "green wire earth ground" connection to use this supply for your application, which we have not been told what it is.
AND, twice, in post #5 and in post #13, the question is asked about the exact points of measurement of said leakage current.
Now I pose an additional question: Is any load connected to the output of the supply at the time this leakage is measured?? AND, if so, what else is the load connected to??
AND one more question: What sort of power supply is this: a transformer based supply? A switch-mode power supply? What is the output current rating of the supply??

 

dont know why the op is worried about the leakage, how they are actualy measuring it or if its real and not a measurement technique problem,
but
I have seen this in past, where there are capacitors or even 10 Mohm resistor from ac side to dc side .
https://recom-power.com/en/rec-n-very-low-noise-filter-for-isolated-dc!sdc-converters-46.html?3
Just wondering if this is the sort of thing the op has.

 

Now we have a problem because it is seems that the complaint is that a current is flowing between the power supply negative output terminal and the mains side "green wire earth ground" connection.
This tends to conflict with the statement from post #1: " I use a power supply with a grounded primary and the "-" output of the secondary shorted to the primary ground "!!!
IF the output negative IS ACTUALLY CONNECTED to that "green wire earth ground" by a very low resistance, then OBVIOUSLY there is an intentional connection, OR AN INTERNAL DEFECT in that particular power supply.
Since the claim is " But the negative output is definitely grounded because I can see it with the tester.", indicating some sort of internal connection, what remains is the option of disconnecting that "green wire earth ground" connection to use this supply for your application, which we have not been told what it is.
AND, twice, in post #5 and in post #13, the question is asked about the exact points of measurement of said leakage current.
Now I pose an additional question: Is any load connected to the output of the supply at the time this leakage is measured?? AND, if so, what else is the load connected to??
AND one more question: What sort of power supply is this: a transformer based supply? A switch-mode power supply? What is the output current rating of the supply??

dont know why the op is worried about the leakage, how they are actualy measuring it or if its real and not a measurement technique problem,
but
I have seen this in past, where there are capacitors or even 10 Mohm resistor from ac side to dc side .
https://recom-power.com/en/rec-n-very-low-noise-filter-for-isolated-dc!sdc-converters-46.html?3
Just wondering if this is the sort of thing the op has.

Hi everyone, I'll clarify a few things because I might have confused some terms while translating.

The power supply has the earth on the primary and the negative terminal of the secondary (output) shorted to the earth of the primary.
I don't have it handy, so I can't give you the part number or model, but I remember there was no schematic, so it's a sealed box.

The first question was how to eliminate or mitigate the a 70uA leakage assuming you don't want to affect the mechanical parts, i.e., without touching insulation, conductive paint, plastic washers, etc.
Is there a need to know if that leakage is a differential mode or common mode?
Theoretically, I say

The second question is how is it possible that, taking two identical power supplies, the one without a earth on the primary has an AC output leakage, while the one with a earth There is a DC leak on the primary output.
Thinking about it a bit I fear that the DC leakage is taken from the earth and brought directly to the secondary since they are shorted

EDIT: earth is not the "-" terminal of the supply, it's another pin Obviously

If you ask me how that leakage was measured I tell you using fluke esa620 and selecting enclosure touch current, then placing the probe on some points of the device such as the "-" terminal of the input pin that carries the power, on screws or other mechanical parts

 

Hi everyone, I'll clarify a few things because I might have confused some terms while translating.

The power supply has the earth on the primary and the negative terminal of the secondary (output) shorted to the earth of the primary.
I don't have it handy, so I can't give you the part number or model, but I remember there was no schematic, so it's a sealed box.

The first question was how to eliminate or mitigate the a 70uA leakage assuming you don't want to affect the mechanical parts, i.e., without touching insulation, conductive paint, plastic washers, etc.
Is there a need to know if that leakage is a differential mode or common mode?
Theoretically, I say

The second question is how is it possible that, taking two identical power supplies, the one without a earth on the primary has an AC output leakage, while the one with a earth There is a DC leak on the primary output.
Thinking about it a bit I fear that the DC leakage is taken from the earth and brought directly to the secondary since they are shorted

EDIT: earth is not the "-" terminal of the supply, it's another pin Obviously

If you ask me how that leakage was measured I tell you using fluke esa620 and selecting enclosure touch current, then placing the probe on some points of the device such as the "-" terminal of the input pin that carries the power, on screws or other mechanical parts

I asked for some pictures of you doing the tests as at these level of currents , just how you hold the cables or the desk the equipments resting on can affect the measurements
the results being strange point to this being so .

 

Now again it makes no sense atall, based on this: "
The power supply has the earth on the primary and the negative terminal of the secondary (output) shorted to the earth of the primary."
That says that there is a connection between the input and the output, so how can there be a leakage current?????
At this point the complaint of a leakage current makes no sense to me.