LED strip lighting even without a positive terminal connection

[Adhith]

hello everyone..
I have been making a project with some 12V RGB LED strips. while trying to check the LED strips I found something interesting. I just connected the ground terminal from a 12V wall adapter to the terminal "B" and left the positive terminal disconnected to the strips but, while touching the positive terminal on the strips  with my finger I saw that the led strips are lighting up with a very low brightness. This also happens even when i tough anywhere in the strip with my finger. I wonder how this is happening. anyone could explain to me about it. I'm attaching a pic of the test that I have done.

[oPossum]

Y capacitor leakage current

[Psi]

yep, pretty normal, you are acting as an antenna and coupling in some 50/60hz AC mains.  Doesnt take much to light up an led

[Zero999]

Hopefully the LEDs have some kind of static protection built-in, otherwise reverse over-voltage can damage them. Red LEDs are normally fine, but the modern blue/white variety tend to be damaged by current flowing in the reverse direction.

[Adhith]

Thank you everyone for your replies. To be honest i didn't completely understood the scene. Since i'm a beginner in electronics im pretty much still confused and if you guys don't mind could you explain a bit ?? like how is it still lighting and the path of current flow and all.

[Zero999]

Thank you everyone for your replies. To be honest i didn't completely understood the scene. Since i'm a beginner in electronics im pretty much still confused and if you guys don't mind could you explain a bit ?? like how is it still lighting and the path of current flow and all.

Most switched mode power supplies have a small capacitor, connecting the output, to the input rectifier, to divert the high frequency switching noise back to the DC bus. Because the capacitor is very small, only a tiny current flows, at mains frequencies, but it's enough to light some highly efficient LEDs. The current can also cause a tingling sensation, but it's far too low, to do any harm.

Because failure of the capacitor will cause the secondary being shorted to the mains, resulting in fire, death or serious injury, the capacitor is specially designed not to fail short circuit and must comply with very stringent safety regulations.

Here's a simplified block diagram, which you need to be logged in the forum to see.

It's also been discussed in the following thread:
https://www.eevblog.com/forum/beginners/connecting-system-ground-to-mains-earth/msg27677/#msg27677

[Adhith]

Thank you everyone for your replies. To be honest i didn't completely understood the scene. Since i'm a beginner in electronics im pretty much still confused and if you guys don't mind could you explain a bit ?? like how is it still lighting and the path of current flow and all.

Most switched mode power supplies have a small capacitor, connecting the output, to the input rectifier, to divert the high frequency switching noise back to the DC bus. Because the capacitor is very small, only a tiny current flows, at mains frequencies, but it's enough to light some highly efficient LEDs. The current can also cause a tingling sensation, but it's far too low, to do any harm.

Because failure of the capacitor will cause the secondary being shorted to the mains, resulting in fire, death or serious injury, the capacitor is specially designed not to fail short circuit and must comply with very stringent safety regulations.

Here's a simplified block diagram, which you need to be logged in the forum to see.

It's also been discussed in the following thread:
https://www.eevblog.com/forum/beginners/connecting-system-ground-to-mains-earth/msg27677/#msg27677

Thank you again Sir for your response. So what I have understood is that the small capacitor where one side is the positive line from the power supply at a 50Hz frequency creates the actual ground output of the SMPS a bit more positive and in that case, my finger at the +Ve trace of the LED strip act as the ground and thus completes the circuit resulting in a small current flow in reverse direction through the LEDs.

[Adhith]

The reason why I'm so concerned with this is because I'm actually working on making led matrix which consist of 10 rows of RGB led strips.
 The led strips are lighted by a VU meter circuit which helps to light the 10 rows of LED strips according to music played just like an equalizer. Since the leds are having this problem of lighting up without a proper positive terminal, each time when i just approach the led matrix or its associated components all the strips are lighting up in this fashion and its getting a bit annoying. Is there any way to prevent this from happening so that the leds would light up only when both the +ve and ground terminals are connected?? I'm attaching the light up pic of the matrix along with this post.

[Zero999]

The reason why I'm so concerned with this is because I'm actually working on making led matrix which consist of 10 rows of RGB led strips.
 The led strips are lighted by a VU meter circuit which helps to light the 10 rows of LED strips according to music played just like an equalizer. Since the leds are having this problem of lighting up without a proper positive terminal, each time when i just approach the led matrix or its associated components all the strips are lighting up in this fashion and its getting a bit annoying. Is there any way to prevent this from happening so that the leds would light up only when both the +ve and ground terminals are connected?? I'm attaching the light up pic of the matrix along with this post.

Yes.
You could try:

Putting a resistor in parallel with the LED strip. It will reduce efficiency slightly but cause the voltage to drop to too lower level to light the LEDs, when the tiny current through the Y capacitor flows. Try 1k to 100k.

Connecting the power supply's DC side to earth (it makes no difference whether it's the positive or negative), via a resistor, a capacitor or even directly. This will shunt the leakage current to earth, so the voltage is too low to light the LEDs.

Put a capacitor in parallel with the LED strip, to shunt the AC current around them. Try 100nF to 1μF.

[Adhith]

Thank you again for your suggestion. To be specific the leds are powered by a separate led driver through the 10 transistors seen in the circuit which i have attached previously.
1) Since its a RGB led strip with common anode, by "Putting a resistor in parallel with the LED strip"  and "Put a capacitor in parallel with the LED strip" did you mean connecting the positive terminal of the strip to any one of the R or G or B terminal or to the power ground??

2) since the DC adapter is having only 2 pins there is no earth pin from it to make it earthed

3)one thing which i couldn't figure out was that, why is the strip so sensitive to the things near it. the strip is even lighting up when I just approach my hand around it and not even touching it. would there be any interaction to worry about between the terminals  +, R, G, B on the LED strips since they are very close to each other and to be on the safe side should I just insulate these pins with some hot glue??

[Adhith]

just shorted each of 10 output wires positive wires to the power ground through a 10K resistor and the problem is solved. I could move forward with this idea right?? also 10K resistor wont do much of power dissipation right? or should i try with a much more higher value for the resistor??

[exe]

yep, pretty normal, you are acting as an antenna and coupling in some 50/60hz AC mains.  Doesnt take much to light up an led

I don't think so. Never seen an LED lit this way. A good 50Hz must be huge.

[Adhith]

I haven't seen this type of disturbance before but now I did. any how its only noticeable at dark and so probably I would have missed before

[exe]

I haven't seen this type of disturbance before but now I did. any how its only noticeable at dark and so probably I would have missed before

Are you sure it's not leakage to ground? That's totally different from AC coupling through the body.

[Gyro]

I haven't seen this type of disturbance before but now I did. any how its only noticeable at dark and so probably I would have missed before

Some modern LEDs can be seen glowing at insanely low currents in a dark environment...

https://www.eevblog.com/forum/projects/led-visibility-optimization-at-very-low-currents/

[Adhith]

I haven't seen this type of disturbance before but now I did. any how its only noticeable at dark and so probably I would have missed before

Are you sure it's not leakage to ground? That's totally different from AC coupling through the body.

Thank you for your kind reply.well the leds are driven by a seperate led driver. small lighting up problem is only seen when my hand goes near it and also when the VU meter starts operating all the led strips are lighted up in a very low brightness just like the picture that i have attached before. This problem is solved through the 10k resistor to the ground at each of the 10 output pins form VU meter. But I'm a bit confused of this weird behavior.

[Adhith]

I haven't seen this type of disturbance before but now I did. any how its only noticeable at dark and so probably I would have missed before

Some modern LEDs can be seen glowing at insanely low currents in a dark environment...

https://www.eevblog.com/forum/projects/led-visibility-optimization-at-very-low-currents/

Thank you for your kind reply. In my case the led light up seems more over like a single dot of light.

[Zero999]

yep, pretty normal, you are acting as an antenna and coupling in some 50/60hz AC mains.  Doesnt take much to light up an led

I don't think so. Never seen an LED lit this way. A good 50Hz must be huge.

I have seen this before. If I take a single high efficiency green LED, connect one to an earthed object, such as a tap or radiator and hold the other in my hand, I the LED lights dimly.

Thank you again for your suggestion. To be specific the leds are powered by a separate led driver through the 10 transistors seen in the circuit which i have attached previously.
1) Since its a RGB led strip with common anode, by "Putting a resistor in parallel with the LED strip"  and "Put a capacitor in parallel with the LED strip" did you mean connecting the positive terminal of the strip to any one of the R or G or B terminal or to the power ground??

You effectively have three LED strips, so of course you need a separate resistor for the red, green and blue elements.

2) since the DC adapter is having only 2 pins there is no earth pin from it to make it earthed

3)one thing which i couldn't figure out was that, why is the strip so sensitive to the things near it. the strip is even lighting up when I just approach my hand around it and not even touching it. would there be any interaction to worry about between the terminals  +, R, G, B on the LED strips since they are very close to each other and to be on the safe side should I just insulate these pins with some hot glue??

As long as you use a resistor per red, green and blue set of LEDs, then it shouldn't cause any problems.

[Adhith]

Thank you Hero999 for your valuable reply. So you were saying that since its an R,G,B led strip i want to use 3 such resistors one for each colour. But when I shorted the +ve terminal of a strip to the ground through a 10K resistor the problem was solved. Does this also work?? or its a wrong thing to do??

[Zero999]

Thank you Hero999 for your valuable reply. So you were saying that since its an R,G,B led strip i want to use 3 such resistors one for each colour. But when I shorted the +ve terminal of a strip to the ground through a 10K resistor the problem was solved. Does this also work?? or its a wrong thing to do??

Thinking about this again. I was wrong about needing a separate resistor for each colour. A single resistor on the +V rail is all that's needed and as it works, keep it as it is.

[Adhith]

Thinking about this again. I was wrong about needing a separate resistor for each colour. A single resistor on the +V rail is all that's needed and as it works, keep it as it is.

thank you for your reply. I'll keep the resistor as it is.