I'm interested about whether or not a conventional neon sign such as this one could emit x-ray that could be harmful.

enter image description here

The sign that you can see in the picture is just below my office window so I'm basically sitting like 8 hours a day about 2 meters away from the sign which is functioning the whole day. Just curious if this can cause any harm of any kind because of the long exposure.

  • 2
    \$\begingroup\$ No, not harmful. \$\endgroup\$ May 19 at 17:24
  • 5
    \$\begingroup\$ If you are concerned, a Geiger counter does respond to x-rays and you can buy them where they are pretty sensitive. They won't tell you what energies -- they just "click." But they can provide a "measurement" of sorts. You could set one up where you sit and let it count for you for a day or so. Then, the next day, move it twice as far away from the sign (or further) and let it count for exactly the same period of time. If you see a significant change between the two summed count readings, and the closer one is much larger than the distant one, then you may have a more valid concern. \$\endgroup\$
    – jonk
    May 19 at 19:11
  • \$\begingroup\$ Thank you everybody. So I guess that if the sign doesn't emit xrays is some other kind of radiation that it can produce that I should be concerned about? Just asking because I sit close to it for a long time every day \$\endgroup\$
    – Raaas
    May 19 at 22:38
  • 1
    \$\begingroup\$ You are probably closer to neon tubes on the office ceiling that are the same, just white in color not green. Millions of people spend many hours a day for years with neon lights over their heads and there is no difference with others living in other conditions. \$\endgroup\$
    – Joan
    May 20 at 14:26
  • 3
    \$\begingroup\$ @Joan green "neon" tubes are probably argon-filled, while the lamps in an office ceiling use tiny amounts of mercury (mixed in to other gases) that then excites a phosphor coating on the glass. So chemically and spectrally very different, even though your conclusion holds \$\endgroup\$
    – Chris H
    May 20 at 15:46

2 Answers 2


Neon signs do not produce X-rays. Although they operate at high voltages, the gas within the tubes prevent electrons from gaining enough energy to produce X-rays.

However, if one wants to split hairs, all emission of electromagnetic waves is probabilistic. Even an incandescent bulb will produce a negligible amount of X-rays. Maybe one photon per second, or maybe one photon per year. I don't know the rate but it is very small compared to the more or less astronomical number of photons produced by an X-ray machine. I would guess that the amount of X-ray radiation you would receive from a neon light might well be less than what you would receive sitting under a beach umbrella on a sunny day.

Edit: I just did some back-of-the-envelope calculations to estimate the power of the X-rays emitted by an incandescent bulb.

Assume an incandescent bulb emits black body radiation according to Planck's Law, and is at 2700 Kelvins. Assume further that X-rays consist of photons with frequency 3 x 10^16 Hz.

To find the relative intensity of X-rays to yellow light, I divided the Planck's Law result for X-rays by that of yellow light and get this approximation.

$$B_{rel} = e^{\left(\frac{h}{k_{B}T}\right)\left({f_{yellow}-f_{x-ray}}\right)}$$

where \$h\approx 6.6 \times 10^{-34} joules\$ is Planck's constant. \$k_{B}\approx 1.38 \times 10^{23} [complicated-units]\$ is Boltzmann's constant, \$T\$ is the temperature in Kelvins.

$$\frac{h}{k_{B}} \approx 4.8 \times 10^{-11} sK$$

$$\frac{h}{k_{B}\times 2700K} \approx 1.778 \times 10^{-14} s$$

$$e^{\left(1.778 \times 10^{-14}\right)\left(3 \times 10^{16}\right)} = e^{\left(5.333 \times 10^2\right)} \approx 4 \times 10^{231}$$

So, (with a bit of further approximation) there is somewhere around \$4 \times 10^{231}\$ times more yellow light emitted from an incandescent light bulb than x-ray radiation. AFAIK, that is more than the number of atoms in the visible universe.

This was just a fun calculation and not meant to be precise.

  • 3
    \$\begingroup\$ Come on, the second paragraph is on par with a probability that vacuum can spontaneously generate pairs of particles, including X-ray level photons. \$\endgroup\$
    – Eugene Sh.
    May 19 at 17:26
  • 3
    \$\begingroup\$ This calculator says that a 1 cm^2 5000K thermal source generates an X-ray photon about once every 10^100 seconds. Not sure I trust the floating point precision there, but seems very unlikely you will ever see an X-ray photon from a light bulb. Link: opticsthewebsite.com/OpticsCalculators \$\endgroup\$ May 19 at 18:02
  • 1
    \$\begingroup\$ @EugeneSh. Lol. I just did a back of the envelope calculation for the x-rays emitted by an incandescent light bulb compared to the yellow light. The amusing result is found in an edit I just made. \$\endgroup\$ May 19 at 18:42
  • 2
    \$\begingroup\$ @Raaas Although I don't know the science well enough to make an airtight case that the UV radiation from a neon sign is nothing to worry about, I am neverthess confident that the UV radiation from neon lights is negligible. If you want an airtight case, I suggest you ask about UV radiation from Neon lights at Physics Stack Exchange. There are no other forms of harmful radiation that could come from a neon light. Some people, however, find electrical hum and the flickering of neon lights (not always noticable) to be low grade psychological stressors. \$\endgroup\$ May 19 at 22:38
  • 1
    \$\begingroup\$ Thank you. Lets say I cover the window so the neon light doesn't get to me. Will that stop the UV radiation coming to me? I mean, the only way to receive UV radiation is to being directly exposed to it as we are exposed to sunlight? \$\endgroup\$
    – Raaas
    May 19 at 22:48

High energy electrons produce x-rays when they strike atoms, usually metal. That's how a scanning electron microscope can detect elements. Basically, you need electrons accelerated with a high voltage and a long mean free path to get x-rays. Cathode ray tubes (CRT) produce x-rays when the electron beam strikes the phosphor. As Math said the noble gas in neon lights makes the mean free path of the electrons too short to gain the energy for the electrons to produce x-rays.

As shown with the black body radiation equations given by Math, to produce x-rays based on the temperature of the object takes temperatures well above our normal capabilities. That is not how we produce x-rays, nor is it how light is produced by neon lights.

Neon lights produce light by electrons interacting with the electron shells of the noble gases atoms. When the electrons lose energy, the interaction with the electron shells produces a photon with the energy equal the amount of energy lost by the electron.

We produce x-rays in a similar fashion, but consider this.

  1. An electron accelerated across a field of 1V has an energy of 1 electron volt (eV). Thus, the upper limit of the energy in eV produced is the voltage of the power source.

  2. Electrons cannot lose more energy than they have. Thus, cannot produce photons with a higher energy than they have.

  3. X-ray photons have energies in keV (>1keV).

Figure 1. Photon energies of the electromagnet spectrum (Taken from "Μελέτη της διακριτικής ικανότητας μηχανής ακτίνων-Χ με τεχνικές προσομοίωσης Monte-Carlo," p. 16)

enter image description here

  1. Thus, the power source to produce x-rays needs to be greater than 1000V.

Neon signs can be operating with 2V to 15kV, but for an electron to gain the energy of the power source, it needs to be accelerated from the negative electrode and strike the positive electrode without colliding with atoms, causing energy loss. In a neon sign electrons collide extensively with noble gas atoms. As the electrons lose energy in the atom's electron shells, they produce visible light photons (1.75 to 3.1eV). The unobstructed path between the electrodes to produce x-rays isn't there for neon signs.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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