Neon lamp indicators embedded in power strip switches indicate whether the switch is ON or OFF.

When the neon lamp is ON in a dark environment (lights of the room are off), the lamp begins flickering and it sometimes goes off for a few seconds.

When I switch the lights of the room on, it stops flickering and remains bright. Even when I unlock my phone in the dark near the lamp, it stops flickering!

Is the light-making-mechanism in neon lamps affected by the light intensity of the surrounding environment?

enter image description here

UPDATE: I captured my observations with my iPhone and uploaded it to YouTube. Here is the link of the video: https://youtu.be/1jlUmEfGHZA

All the lights of my room were turned off. There were no plugs connected to the power strip. The only source of light was the screen of my MacBook Pro, which was not plugged in. I increased and decreased the brightness of the computer screen during the video. I then produced two additional light environments using my phone.

  • \$\begingroup\$ The light of gas discharge lamp is due to ionization, mostly caused by aceleration due to high voltage and collison of electrones with gas molecule. The ionization however can happen also due to external causes. \$\endgroup\$ Apr 24, 2016 at 20:01
  • 3
    \$\begingroup\$ Maybe here is an another guy making some hypothesis arrond your thougts: technicaltom.wordpress.com/2013/09/14/… \$\endgroup\$ Apr 24, 2016 at 20:09

5 Answers 5


This is a common phenomenon: neon pilot lights have a limited lifetime, and after many years of use, they begin to flicker, then they finally go dark. They no longer can operate at line voltage, but instead require a higher voltage for stable operation.

Also, neon pilot lights can act as photosensors. Try this with a flickering neon bulb: shine a red LED on it. Then shine a blue LED on it. If both LEDs are roughly the same brightness, then the blue LED should have a much greater effect. You're seeing the same photoelectric effect that won Einstein the 1921 Nobel, leading to the "photon" concept.

All gas-discharge tubes suffer from a common problem: the gas atoms tend to become embedded in the metal surface of electrodes, so over many years the gas pressure falls as well as the gas mixture being altered.

  • \$\begingroup\$ +1 for the wavelength dependant test suggestion to confirm the photoelectric effect. \$\endgroup\$
    – KalleMP
    Mar 4, 2018 at 7:55
  • \$\begingroup\$ Could this flicker have an effect on a CRT image? I have a power strip with neon illuminated switches on each of the six outlets. They all flicker in varying degrees, except for one. While my test wasn't extensive, my CRT was plugged into one of the flickering outlets and the image was jittery. I unplugged it and replugged it in the non-flickering outlet and the image stabilized. Could the neon bulb flicker cause the jittering image? \$\endgroup\$
    – user148298
    Mar 2, 2020 at 19:07
  • \$\begingroup\$ @user148298 I'd guess the indicator is close enough in the circuit that any power fluctuations through the lamp cause very slight fluctuations in the powered circuit. For most electronics these don't really affect anything, but if your CRT's timing was already borderline, those tiny fluctuations may have disturbed the synchronization,, resulting in a jittery image. Just my guess, though. \$\endgroup\$
    – Doktor J
    Mar 23, 2020 at 13:48
  • \$\begingroup\$ @DoktorJ I found out it had nothing to do with flickering of the illuminated switches. The monitor I was using was a studio monitor and I had it sitting on another studio monitor. The bottom one created magnetic interference with the top one. \$\endgroup\$
    – user148298
    Mar 24, 2020 at 14:38
  • \$\begingroup\$ @user148298 hah I had that a long time ago! Had a big 19" Sony Trinitron (one of the 60lb beasts); got a 17" Viewsonic, but if I put them too close together the Viewsonic's deflection plates interfered with the Trinitron's sync and made it all wobbly; since the Viewsonic was shadow mask based, it seemed much less susceptible to interference from the Sony, whereas the Trinitron was aperture grille based and seemed more susceptible to interference. \$\endgroup\$
    – Doktor J
    Mar 24, 2020 at 15:07

To quote from the great wiki - https://en.wikipedia.org/wiki/Neon_lamp

"When the current through the lamp is lower than the current for the highest-current discharge path, the glow discharge may become unstable and not cover the entire surface of the electrodes.[6] This may be a sign of aging of the indicator bulb, and is exploited in the decorative "flicker flame" neon lamps. However, while too low a current causes flickering, too high a current increases the wear of the electrodes by stimulating sputtering, which coats the internal surface of the lamp with metal and causes it to darken.

The potential needed to strike the discharge is higher than what is needed to sustain the discharge. When there is not enough current, the glow forms around only part of the electrode surface. Convective currents make the glowing areas flow upwards, not unlike the discharge in a Jacob's ladder. A photoionization effect can also be observed here, as the electrode area covered by the glow discharge can be increased by shining light at the lamp."

  • \$\begingroup\$ To concentrate on the first sentence of your quote: "When the current through the lamp is lower than the current for the highest-current discharge path..."--how do you propose that this happens in OP's case? The line voltage is not likely to fall significantly just by turning on some lights, much less a battery-powered telephone or laptop computer. Also, it is not photoionization (nothing is being ionized), but rather the photoelectric effect. \$\endgroup\$ Apr 25, 2016 at 2:48
  • \$\begingroup\$ @OleksandrR. I don't propose anything, I'm simply quoting from the article. I would agree that the term used in the article should have been photoelectric rather than photoionisation (as in a GM tube). If you wish to correct and update the wiki article please feel free \$\endgroup\$ Apr 25, 2016 at 12:50
  • \$\begingroup\$ @OleksandrR. The current will fall to zero 100 (129) times per second. The discharge path changes with age and at some point the peak line voltage will no longer achieve the minimum stable current path demands. \$\endgroup\$
    – KalleMP
    Mar 4, 2018 at 7:49
  • \$\begingroup\$ @JImDearden The GM tube does not use Photo-ionisation in normal function. It does not care about the work-function of the electrode surfaces. You are thinking perhaps of a Phototube en.wikipedia.org/wiki/Phototube \$\endgroup\$
    – KalleMP
    Mar 4, 2018 at 7:53

Responding to old post but it comes up at top of google search:

Neon lights can't work in complete darkness. This is the "neon lamp dark effect".

I know it sounds fake, but that's why neon lamp bulb makers sometimes add a little ionizing radiation inside the bulbs.

Dark Effect: All ILT Neon lamps are subject to a condition called dark effect. Dark affect is defined as a drastic increase in the amount of voltage required to make a lamp glow when the lamp is in a dark environment. Because the lamp is photosensitive, it may require many additional volts to start if no light is present. Neon lamps can also become erratic in total darkness.

To prevent dark affect an external source may be run near the neon lamps or in some cases, custom Neon lamps can be supplied with a radioactive gas, often Krypton 85.

(Source: International Light Technologies, Inc. - ILT)

Yes, it is bizarre and a trouble-shooting nightmare for the unaware, lol. But you can use a simple neon bulb as a photosensor. But also, they do flicker more and more as they age and die: but they are not really dead, they just need more power than the source device gives. If you ramp up the juice (in a different device) they will work for a while longer.

  • 1
    \$\begingroup\$ Michael Welcome to the EE stack exchange. Please go to the help button and take the instructional tour. Good answer though +1. \$\endgroup\$
    – RoyC
    Apr 1, 2018 at 17:39

There is natural background radioactivity around us at all times. Ionising particles may pass through the glass of the neon tube at random causing the gas to conduct more fully at various times. This would be seen as a flickering effect.

The is the principle of the Geiger counter where a doped gas conducts as ionising particles pass through it via a thin mica window.


There are some possibilities here.

1: You live in a house where somehow, weirdly, the power dips and unbalances when there is no use. That normally only happens with over-use, though.

2: You're like every single other person in the world and have working pupils and a normal human brain. (That may either be a relief to you, or a shock, I hope the prior).

Of course, there are some other options, but I'll explain the more likely number 2.

Step one to that is that our brains are amazing processors that know what is important and what is not in the context that they are seeing. We've trained them for millions of years to become good at that. So in normal conditions the flickering is unimportant and boring to your brain.

Step two is that in low light conditions your pupils widen, allowing a lot more of the light and differences into our eyes. Presenting, suddenly, a much stronger flicker to your brain in a dark environment with nothing else to process.

Step three is that your brain is also very good at fooling itself; you see what you think you see. If you start focusing on the Light your expectations exaggerate the actual flickering.

Step four is that any light source will create a huge amount of light compared to the Neon light, closing your pupils, removing the level of detail from the Neon light and also adding a lot of new information for your brain to focus on that seems more important. (Where the chairs are. Are there any feral cats under my desk? Did you get any new texts? Why has {insert cute girl [-or boy as far as I care-]'s name here} not responded yet?)

Step five is an actual thing that happens: The light turning off can happen once in a while through either the light wearing down, or having small moments of high levels of polarization in a place where there's little fluorescent material, so that all current flows there for a second or two creating nearly no light. This would logically also happen when the lights are on, but you may just not notice so much with a lot of light around you, since the event is not significant at all. Think about it: If you have 5000 Neon lights, and one pops off, how long would it take you to notice? If you have only one of those same lights, and it goes off suddenly, you're bound to see immediately: All your light is gone!

The fifth is helped by the Neon lights in power strips costing a negligible amount of money, as they are the cheapest kind of Neon light ever invented. 'You get the quality you pay for'.

All that said, of course, there may be a combination of several things going on. But I'm betting (heavily) on the flickering/no-flickering definitely just being a change in perception.

The Neon's in series with a load, such as in a wall light switch, can of course exhibit all kinds of strange behavior with capacitive loads, or inductive loads, or very light loads. But the Neon in a power strip is directly connected to the mains through the switch, in parallel with all other loads, so that's not happening here.

It might only 'spark a bit brighter or dimmer' just the once when you switch on or off a very heavy load, but that should be clear enough, in some buildings all the lights always do that.

  • \$\begingroup\$ +1. OP could try monitoring the neon through a cardboard tube, blocking out ambient light, while someone switches load to see if the effect is real or perceived. \$\endgroup\$
    – Transistor
    Apr 24, 2016 at 20:04
  • \$\begingroup\$ @transistor But the OP says that the ambient light is the cause. If you block it, then it has no sense for such experiment. \$\endgroup\$ Apr 24, 2016 at 21:28
  • 1
    \$\begingroup\$ @Asmyldof Thanks for your answer but this fact is not only sensible my eyes/brain/perception but also I captured it with my iPhone. There is no connected plugs to the sockets so maybe the load is constant. In addition, I saw this fact in other cities. So I think these steps 1-5 is not matching the fact. \$\endgroup\$ Apr 24, 2016 at 22:50
  • \$\begingroup\$ @transistor I think this is not related to the load. With a cardboard tube and without any ambient light, the flickering is observed. Even with a camera. \$\endgroup\$ Apr 24, 2016 at 22:58
  • \$\begingroup\$ @HamidRohani Why do you think the Auto-Gain on your iPhone is any different than your pupils? Who's looking at your iPhone, You or some omnipotent objective artificial intelligence? How did I ever imply any of these brain effects only happen in your house and not anywhere else in the world? \$\endgroup\$
    – Asmyldof
    Apr 25, 2016 at 6:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.