# Why are glass light bulbs evacuated? **URL:** \ **Category:** Factual Questions\ **Created:** [September 2, 2002, 9:32pm UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720 "2002-09-02T21:32:56Z")\ **Posts on this page:** 12\ **Page:** 1
**Author:** ![cdhostage](https://avatars.discourse-cdn.com/v4/letter/c/c6cbf5/32.png) [@cdhostage](https://boards.straightdope.com/u/cdhostage)\ **Post date:** [September 2, 2002, 9:32pm UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/1 "2002-09-02T21:32:56Z")
Why go the trouble of surrounding a filament with vacuum surounded by a thin glqass barrier? Glass is fragile enough by itself, but why confound the problem by making an odd shape out of it only a few millimeters thick? Why not just surround the wire with a glass cladding and be done with it? It would be stronger, cheaper top make, and could even be made so as not to shatter into the horrible jaggies. Perhaps 1. Glass is an insulator, but vacuum is the best insulator. Of heat I mean. Thusly, a greater deal of heat would be conducted away from the filament, thus requiring a greater amount of electricity to continue heating the tungsten up to the proper temperature. The light bulb would be 1% efficient instead of 5%. Or some absurdly small number like that. 2. Again, glass is a worse insulator than vacuum. Glass might melt att the temperatures needed to make tungsten glow white, and would not lose that heat easily. This would be a dangerous situation. 3. Heating expansion problems. Glass is brittle, and might crack or explode if heated suddenly by a high wattage incandescent light. Any other ideas, or perhaps even the real answer? ---
**Author:** ![Mangetout](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/mangetout/32/19_2.png) [@Mangetout](https://boards.straightdope.com/u/Mangetout)\ **Post date:** [September 2, 2002, 9:39pm UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/2 "2002-09-02T21:39:12Z")
I believe that they are actually filled with an inert gas at low pressure; atoms of metal from a hot filament in a vacuum can evaporate off, but apparently when there’s gas present an evaporated atom stands a fair chance of being ‘bounced’ back onto the filament, prolonging the life. ---
**Author:** ![astro](https://avatars.discourse-cdn.com/v4/letter/a/9dc877/32.png) [@astro](https://boards.straightdope.com/u/astro)\ **Post date:** [September 2, 2002, 9:48pm UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/3 "2002-09-02T21:48:30Z")
Welcome to the SDMB! Per Mangetout your assumptions about the final construction (ie “evacuated”) of light bulbs is incorrect. Also FWIW a light bulb’s glass shell is considerably thinner than “a few millimeters”. [How things work](http://howthingswork.virginia.edu/incandescent_light_bulbs.html) > [@](#): > > How are incandescent light bulbs made? – SU > > The glass enclosures are made from a ribbon of hot glass that’s first thickened and then blown into molds to form the bulb shapes. These enclosures are then cooled, cut from the ribbon, and their insides are coated with the diffusing material that gives the finished bulb its soft white appearance. > > The filament is formed by drawing tungsten metal into a very fine wire. This wire, typically only 42 microns (0.0017 inches) in diameter is first wound into a coil and then this coil is itself wound into a coil. The mandrels used in these two coiling processes are trapped in the coils and must be dissolved away with acids after the filament has been annealed. > > The finished filament is clamped or welded to the power leads, which have already been embedded in a glass supporting structure. This glass support is inserted into a bulb and the two glass parts are fused together. A tube in the glass support allows the manufacturer to pump the air out of the bulb and then reintroduce various inert gases. When virtually all of the oxygen has been eliminated from the bulb, the tube is cut off and the opening is sealed. Once the base of the bulb has been attached, the bulb is ready for use ---
**Author:** ![NotMrKnowItAll](https://avatars.discourse-cdn.com/v4/letter/n/6bbea6/32.png) [@NotMrKnowItAll](https://boards.straightdope.com/u/NotMrKnowItAll)\ **Post date:** [September 2, 2002, 9:53pm UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/4 "2002-09-02T21:53:49Z")
Additionally, before the bulb is assembled, the filament is dipped in something called “getting.” Upon the initial in-factory lighting, this getting combusts, consuming any residual oxygen left in the bulb. ---
**Author:** ![Mangetout](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/mangetout/32/19_2.png) [@Mangetout](https://boards.straightdope.com/u/Mangetout)\ **Post date:** [September 2, 2002, 10:31pm UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/5 "2002-09-02T22:31:51Z")
Another point is that if you have a hard vacuum inside the bulb, then the pressure outside is more likely to force air (including oxygen) through any microscopic leaks or imperfections (possibly diffusion through the glass itself - can anyone confirm this or otherwise?) - oxygen inside the enclosure is bad because the filament will burn. ---
**Author:** ![AskNott](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/asknott/32/5790_2.png) [@AskNott](https://boards.straightdope.com/u/AskNott)\ **Post date:** [September 3, 2002, 12:04am UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/6 "2002-09-03T00:04:57Z")
I once worked in the “sealed-beam” headlights section of our plant. The tungsten wire came to us in the right diameter, and our machines wound it into filaments. The filaments were clamped into slots in the support rods. After the lens was welded on top, the air was sucked out and replaced with nitrogen. Then the little tube on the back was heated until soft and nipped shut. The lamp was test-lit, and if it still had air in it, the filament would burn, filling the lamp with thick yellow smoke. The whole process was much more complicated than that, and it would take quite a while to describe. By the way, here’s an answer. If the filament were simply encased in solid glass, the glass would absorb the filament’s heat. It would take an unacceptably long time to get hot enough to light up. ---
**Author:** ![ftg](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/ftg/32/2801_2.png) [@ftg](https://boards.straightdope.com/u/ftg)\ **Post date:** [September 3, 2002, 1:25am UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/7 "2002-09-03T01:25:23Z")
**cdhostage** , I think all 4 of your reasons are valid. But there are also more. Note that the bulbs in overhead projectors are quite small and of course generate a lot of light. The glass of these bulbs are quite pure and transparent. Also, you are NOT supposed to touch your finger to the bulb while installing. The grease from your fingers will cause a slight variation in heating of the bulb and it will break. Note also the bulb gets so hot that you would burn yourself touching it but it is safely enclosed when running. (And with a fan.) Imagine a 100w bulb’s heat and now concentrate it in 1/10th the surface area. Ouch. Presumably, in mass production, the glass of regular light bulbs is really cheap stuff and you can’t expect Joe Consumer to understand “don’t ever touch the bulb”, hot or cold. As far as solidly encasing it. Everything would crack. The bulb, the filament, the connection with the base. ---
**Author:** ![Finagle](https://avatars.discourse-cdn.com/v4/letter/f/b19c9b/32.png) [@Finagle](https://boards.straightdope.com/u/Finagle)\ **Post date:** [September 3, 2002, 2:09am UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/8 "2002-09-03T02:09:35Z")
They have to evacuate the bulbs – too much atmosphere and people screw in them. ---
**Author:** ![Triskadecamus](https://avatars.discourse-cdn.com/v4/letter/t/b19c9b/32.png) [@Triskadecamus](https://boards.straightdope.com/u/Triskadecamus)\ **Post date:** [September 3, 2002, 5:31am UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/9 "2002-09-03T05:31:11Z")
The shape itself is pretty much legacy. Folks will buy the kind that fit the lamps they have, and folks will make the type of lamp that most people want, and most people want lamps that they can get bulbs for easily. The shape of the bulb hasn’t changed all that much since Edison. Tris ---
**Author:** ![bbeaty](https://avatars.discourse-cdn.com/v4/letter/b/c0e974/32.png) [@bbeaty](https://boards.straightdope.com/u/bbeaty)\ **Post date:** [September 3, 2002, 7:17am UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/10 "2002-09-03T07:17:33Z")
And red-hot glass is a conductor (white hot glass being even more conductive.) Old physics classroom demo: a glass rod is wrapped with two wires with a couple of cm between the two wrappings. Connect the two wires to a high voltage power supply. Heat up the exposed glass with a torch. Suddenly the red-hot glass spontaneously glows yellow, then incandescent white, then it dribbles away and breaks the circuit. It became conductive and temporarily formed a high-power resistor. Molten glass, being full of mobile ions, is actually an electrolyte. Hardened glass has just as many ions, but they cannot flow along as an electric current. ---
**Author:** ![SmackFu](https://avatars.discourse-cdn.com/v4/letter/s/53a042/32.png) [@SmackFu](https://boards.straightdope.com/u/SmackFu)\ **Post date:** [September 3, 2002, 2:54pm UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/11 "2002-09-03T14:54:50Z")
Another point… how would you make a soft-light bulb that way? And [more](http://invsee.asu.edu/Modules/lightbulb/meathist4.htm) on the lightbulb design: > [@](#): > > Inert gases such as nitrogen and argon were later added to bulbs to reduce tungsten evaporation, or sublimation. While these gases reduced evaporation and increased filament life, they also carried heat away from the filament, reducing its temperature and brightness. Winding the wires into fine coils, as used in modern incandescent filaments, reduced convective heat loss, allowing the filament to operate at the desired temperatures. ---
**Author:** ![CurtC](https://avatars.discourse-cdn.com/v4/letter/c/ce73a5/32.png) [@CurtC](https://boards.straightdope.com/u/CurtC)\ **Post date:** [September 3, 2002, 3:42pm UTC](https://boards.straightdope.com/t/why-are-glass-light-bulbs-evacuated/126720/12 "2002-09-03T15:42:29Z")
However, vacuum tubes, believe it or not, do have a near-vacuum inside. Vacuum tubes are what we used to use instead of transistors for electric amplification. Also, the computer screen that you’re looking at right now is one. They need to have a vacuum because the stream of electrons would ionize any gas inside screw up what it’s trying to do. I’ve seen CRT manufacturing, and they use “getters.” A getter is a little pot of some dark metallic material, located inside the tube, facing the glass wall. After the tube is sealed, the getter is heated using electrical induction from the outside, causing the material to boil off. The material then hits the glass wall and sticks. But on its way there, it will carry with it any air molecules it encounters, making a more perfect vacuum inside. After that’s done, the tube glass near the getter will be mirrored in that spot because it has metal on its backside. If you ever see a bare CRT, you’ll see mirrored spots on it. A light bulb filament couldn’t be encased in glass because the glass would be molten at that temperature. Glass gets runny as it gets hot - it’s made when it’s just barely orange-hot. At white-hot it would be much runnier. And white-hot glass is the same temperature as white-hot tungsten, by the way. You can tell the temperature of a material by the EM radiation coming off it without knowing what the material is.