LED flickers when touching moving soldering iron (other pin on my hand): ESD?

[741]

I'm assembling a new PCB. On this PCB is an 0805 green LED. One pin goes nowhere (because the series resistor is yet to be fitted).

If I put either LED pin in good contact with my hand, then touch the direct-mains soldering iron tip to the other contact there is a very brief flicker of dim light.

The flicker only recurs if I remove the iron, then re-apply it. If I unplug the iron, nothing happens.

I can only think of obvious things like mains earth not same potential as my hand, and I'm looking at tiny ESD events generated by my clothing? I can see the LED would rectify any AC and hence the effect is observed 'either way'. However, I'd think ESD would be unipolar, with polarity dependent upon the actual composition of the clothing I'm wearing. I have in mind those electrostatic experiments where a glass rod is rubbed with different materials and the result is either a net deficit or excess of electrons.

[Gyro]

You don't say what iron you have, but clearly the tip isn't grounded. The reason for the flash is that there is some capacitance involved it the leakage path to the voltage appearing on the tip, either between the element and barrel if it is a basic mains heating element, or maybe interference suppression Y capacitor leakage if it is a low voltage element iron with a separate controller or power brick. You need to ensure that the tip is grounded before working on any sensitive components (the LED flicker is a warning).

[Zero999]

You don't say what iron you have, but clearly the tip isn't grounded.

Not necessarily. It's possible he is charged or is picking up mains hum. If I hold one lead of a high efficiency green LED in my hand and touch the other lead on an earthed object, such as a tap or radiator, it glows dimly.

[741]

I've just 'beep-tested' the tip and it is connected OK to the earth pin on the mains plug - the iron is an Antex CS18.

Interesting that this kind of effect happens just touching a radiator...

In this case though I'm surprised that mains hum would need me to re-apply the iron - it should give a continuous glow surely? That was what made me think it was not hum. The flicker is one-shot. If I tap the tip twice per second, there is not a flash each time, it's a bit erratic.

[magic]

Try two LEDs in anti-parallel and see if it helps

[SiliconWizard]

You don't say what iron you have, but clearly the tip isn't grounded.

Not necessarily. It's possible he is charged or is picking up mains hum. If I hold one lead of a high efficiency green LED in my hand and touch the other lead on an earthed object, such as a tap or radiator, it glows dimly.

Yeah. It's possible he is just not grounded himself. There is a potential difference between him and the soldering iron tip. That's pretty much it.
Just tested myself (not grounded, with shoes) between one finger and the earth pin of a mains socket. Several volts. Absolutely normal, that could be much worse depending on conditions.
And, those SMD LEDs can glow dimly with only a very low current (in the µA or tens of µA range), so a tiny leakage current and a potential difference of a couple volts is all that's needed here.

[Gyro]

I've just 'beep-tested' the tip and it is connected OK to the earth pin on the mains plug - the iron is an Antex CS18.

Interesting that this kind of effect happens just touching a radiator...

In this case though I'm surprised that mains hum would need me to re-apply the iron - it should give a continuous glow surely? That was what made me think it was not hum. The flicker is one-shot. If I tap the tip twice per second, there is not a flash each time, it's a bit erratic.

Ah yes, non grounded user is clearly the other alternative. The hit and miss flash is probably due to you picking up the surrounding AC mains field - near waveform peak (of the right polarity), you get a flicker; near a zero crossing, you don't.   I'm not sure why it's a one off flicker, it could be body capacitance discharging or it could be some effect of the LED not conducting (possibly going into reverse breakdown) on alternating half cycles. I suppose it's possible that the difference between the LED forward and reverse voltages is 'biassing' the average DC potential of your body negative (basically acting as a rectifier) to the point where you can no longer provide sufficient forward current to light the LED.

It would be a good idea to invest in an ESD mat and strap.

[Zero999]

It would be a good idea to invest in an ESD mat and strap.

Agreed. It doesn't take much to damage some components. Modern InGaN LEDs can also be very sensitive and can be damaged when reverse biased to too higher voltage.

[tszaboo]

I had this with a board lying on an ESD mat. Turns out (pun intended) the ESD mat wasn't connected to earth. The LED efficient enough to light up with mains hum.

[amyk]

This topic seems to come up periodically...

https://www.eevblog.com/forum/chat/led_s-light-when-soldering/

https://www.eevblog.com/forum/beginners/soldering-iron-is-giving-off-usable-power/

https://www.eevblog.com/forum/beginners/led-powered-by-soldering-iron/

And, those SMD LEDs can glow dimly with only a very low current (in the µA or tens of µA range), so a tiny leakage current and a potential difference of a couple volts is all that's needed here.

Less than 1 uA is needed, few nA according to some of the others in those posts above.