lights that light up when they shouldn't

[Fulvio Romano]

I have a question I'm ashamed of.
I bought four led ceiling lights and mounted them at home. each has a double insulation ac/dc plastic transformer with neutral input and phase.
three of these, from off, have a flebleable residual light clearly visible if the room is completely dark.

1. Why?
2. How do I solve this?

 

[New Rider]

Hi.

phase and neutral are connected correctly?
Do you have a line up nearby?

 

[PierArg]

I knew with the led lights happen

 

[Fulvio Romano]

phase and neutral are connected correctly?

I connected them. I had even wrong, if I did it means I thought it was right, so I would hardly be able to grasp the error.
Anyway, what do you mean "right"? are two threads, there are two ways to connect them, both correct. What mistakes could I have made?

Do you have a line up nearby?

No, I'm in the middle of the ceiling.

 

[Stefano_GS]

I have a question I'm ashamed of.
I bought four led ceiling lights and mounted them at home. each has a double insulation ac/dc plastic transformer with neutral input and phase.
three of these, from off, have a flebleable residual light clearly visible if the room is completely dark.

1. Why?
2. How do I solve this?

on why I would say that there can be 3 hypotheses.
1. The switch doesn't do his job well. imaginative hypotheses
2. the led element consists of a fluorescent base, which, even if switched off in the dark preserves a visible fluorescence with total darkness (which however degrades over time). try to see if disconnected completely give the same effect. imaginative hypothesis since a ceiling lighter does not present this behavior
3. the problem could be given by a phenomenon of inductance if the chandelier return cable to the switch passes in tubes with other cables in phase (the phenomenon worsens so much the length of the stroke). can be solved with bipolar switch or dissecting the line near the led with a relay

 

[exxon]

the reason is the parasite coupling of the wire left open by the switch with the grounding system. the coupling is capacitive (not inductive, being the open circuit).

if you are lucky (depending on the debalancing of neutral and phase compared to the earth), it is resolved by moving the switch on the other of the power cables (it is not enough to reverse the cables at the input of the power supply. . . )

if you are less fortunate (if phase and neutral have both a difference of potential consistent with ground) then you need to use a bipolar switch or insert in parallel to the input of the power supply a resistance of bleeding (560 kohm/1w) that solves the issue. resistance dissipates less than 0.1 w, which is irresistible.

if you have the ability to do it, the resistance is the fastest method.

 

[Fulvio Romano]

1. The switch doesn't do his job well. imaginative hypotheses

little plausibile

2. the led element consists of a fluorescent base, which, even if switched off in the dark preserves a visible fluorescence with total darkness (which however degrades over time). try to see if disconnected completely give the same effect. imaginative hypothesis since a ceiling lighter does not present this behavior

No, stay on all night.

3. the problem could be given by a phenomenon of inductance if the chandelier return cable to the switch passes in tubes with other cables in phase (the phenomenon worsens so much the length of the stroke). can be solved with bipolar switch or dissecting the line near the led with a relay

I don't understand.
Suppose I interrupt the neutral. the phase passes into the led and discharges on a non-stop neutral of another user.
Suppose I interrupt the phase. the neutral has not enough tension to do the same game.

It's enough for me.
1. use bipolar contact, or
2. interrupt the other cable, or
3. pull out the cables, twist them and re-enter them in the channel.

Right?

 

[New Rider]

the reason is the parasite coupling of the wire left open by the switch with the grounding system. the coupling is capacitive (not inductive, being the open circuit).

if you are lucky (depending on the debalancing of neutral and phase compared to the earth), it is resolved by moving the switch on the other of the power cables (it is not enough to reverse the cables at the input of the power supply. . . )

if you are less fortunate (if phase and neutral have both a difference of potential consistent with ground) then you need to use a bipolar switch or insert in parallel to the input of the power supply a resistance of bleeding (560 kohm/1w) that solves the issue. resistance dissipates less than 0.1 w, which is irresistible.

if you have the ability to do it, the resistance is the fastest method.

I connected them. I had even wrong, if I did it means I thought it was right, so I would hardly be able to grasp the error.
Anyway, what do you mean "right"? are two threads, there are two ways to connect them, both correct. What mistakes could I have made?


No, I'm in the middle of the ceiling.

Here...

 

[Fulvio Romano]

the reason is the parasite coupling of the wire left open by the switch with the grounding system. the coupling is capacitive (not inductive, being the open circuit).

if you are lucky (depending on the debalancing of neutral and phase compared to the earth), it is resolved by moving the switch on the other of the power cables (it is not enough to reverse the cables at the input of the power supply. . . )

Okay, what I wrote here:

2. interrupt the other cable, or

if you are less fortunate (if phase and neutral have both a difference of potential consistent with ground) then you need to use a bipolar switch or insert in parallel to the input of the power supply a resistance of bleeding (560 kohm/1w) that solves the issue. resistance dissipates less than 0.1 w, which is irresistible.

if you have the ability to do it, the resistance is the fastest method.

If I have to dissipate 100mw I make them dissipate directly to the lamp, so that fleble light does not bother me.
[EDIT] as not said. making two accounts probably the coupling is draining about 5 or 6 w...

 

[Fulvio Romano]

the reason is the parasite coupling of the wire left open by the switch with the grounding system. the coupling is capacitive (not inductive, being the open circuit).

but sorry, and the differential doesn't notice anything? drena coupling less than 30ma?

 

[exxon]

but sorry, and the differential doesn't notice anything? drena coupling less than 30ma?

It's all about microamps.

the small LEDs of the lamps used in the ceiling lights (type 3014), have a current of operation to regime of some ten milliampere. to that current emit a lot of light and have efficiencies in the order of 100 lm/w.

if the current is reduced, the efficiency increases. first linearly, then with increasing speed. in the range of the few microamps, efficiency is hundreds of times higher and also with difficult to measure currents, it is however noticeable the light output.

the led element consists of a fluorescent base, which, even if switched off in the dark preserves a visible fluorescence with total darkness (which however degrades over time).

the garnets of ittrio and aluminum (yag) used in white leds (they are called phosphors, but do not contain phosphorus). are strictly fluorescent crystals and have no phosphorescent behaviour. Therefore, once the excitement is removed (the blue light of the underlying led), they instantly stop (there is nanoseconds) from generating secondary emission.

the residual brightness that is known after the shutdown is due to the charge present in the output capacitors of the power supply. when the voltage arrives near the threshold voltage of the gallium nitride (semiconductor with which the led is made, about 3 v), the current drops to microscopic values and the capacitors discharge more for internal loss than for energy transfer to the load. this current is however sufficient to make the led emit a visible light. this condition can last for several seconds.