Blinking LED, again

[ptonsparky]

If unused fan switch leg was on, maybe there is a staple hammered in too tight that damaged the conductor insulation and there is an intermittent arcing fault between the black and red wire.

I would expect that to cause both lamps of the fixture to come on bright. This was only one of the two and very dim. Almost as if it were on a cycle timer.

 

[synchro]

How sensitive to induction could these LEDs be given the very low demand under normal power, could a slight glow be acheived from induced current from another conductor running along side the switch leg? Another possible source of induced or leaked current could be like some installations that had the hot feed land in the fixture box first then have a switch leg down to the switch.

My thoughts as well.

The room was set up for a wall switched ceiling fan. We have only a lighting fixture installed. The switch leg to the future ceiling fan was On.

It is a 3 wire w/g. Two switch legs and neutral from 2 gang wall box to ceiling. No more than 20' in length. Seems short and with no active load on the fan SW leg, puts my theory in doubt.

The voltage across the capacitance between the fan switch leg when it's at 120V and the LED switch leg will produce some amount of current through the LED. There does not have to be any load current on the fan switch leg for that to occur. And so the theory that both of you are proposing might still be true. It could be interesting to experiment with a length of Romex and an LED bulb to see if such blinking can be induced, and if so, what length of cable would be required. I can't do it because Romex is not commonly available in the Chicago area. A convenient excuse.

 

[Sea Nile]

The voltage across the capacitance between the fan switch leg when it's at 120V and the LED switch leg will produce some amount of current through the LED. There does not have to be any load current on the fan switch leg for that to occur. And so the theory that both of you are proposing might still be true. It could be interesting to experiment with a length of Romex and an LED bulb to see if such blinking can be induced, and if so, what length of cable would be required. I can't do it because Romex is not commonly available in the Chicago area. A convenient excuse.

To add on to your theory, the induced voltage could slowly charge a capacitor in the circuitry of one or both of the LEDs until it reaches the threshold to discharge?

Wouldn't you be able to measure the voltage with the right tester?

 

[ptonsparky]

To add on to your theory, the induced voltage could slowly charge a capacitor in the circuitry of one or both of the LEDs until it reaches the threshold to discharge?

Wouldn't you be able to measure the voltage with the right tester?

My 87 might be up to the task.

MDSW might not be.

 

[synchro]

To add on to your theory, the induced voltage could slowly charge a capacitor in the circuitry of one or both of the LEDs until it reaches the threshold to discharge?

Yes. What you are describing is known as a relaxation oscillator. In the LED bulb, the capacitively coupled current could be being rectified and then charging a capacitor. The voltage on the capacitor would build up until it is large enough to make the LED circuitry draw some current and light it dimly. This would discharge the capacitor, and then the cycle of the blinking would start over again.

Wouldn't you be able to measure the voltage with the right tester?

My 87 might be up to the task.

Yes. I think the horizontal bar on the bottom of the 87's display could be used to view the build up and discharge of AC voltage on the rectifier input that is connected to the bulb terminals.