HV glows in incandescent lamps

[klugesmith]

Most burned-out incandescent lamps can be given new life by applying high voltage.
Here's a semi-burned-out lamp that came out of my car this morning.   Type 9003 "Quartz-halogen", 12 V, 55 + 60 W.


What if we apply high voltage, for example from NST on variac, between the apparently open terminal and the other two.
Will we get an arc between broken ends of filament, or between exposed metal parts outside the fused-quartz envelope?
If inside, does anyone want to guess the voltage at 30 mA or 1 mA?

Would the electrical behavior of low-beam circuit be substantially different if globe is heated, by operating 60 watt high beam for a while?

[klugesmith]

Does this really belong in Electronic Circuits instead of Light, Lasers, and Optics?

In a quick test with variac and NST, a tiny white arc developed between two turns of tungsten filament on opposite sides of the break.
Here shown at about 1 mA;  I have no voltage measurements of starting or running voltage.

We might guess that the fill gas includes xenon along with the titular halogen element.
As current increases, microscope shows an arc glow spreading over larger areas of the electrode surface. Shown below at about 5 mA.  I need to get a HV rectifier, and see which polarity gets the glow and gets the heating. Any guesses?
I will get a video up that shows irregular changes in metal incandescence temperature on one side.   Might be that hot cathodes with mismatched temperature make a self-rectifying glow lamp / arc lamp.

This tungsten wire surface has a crystalline appearance at the end of its lifetime, while the other filament is almost as smooth as in a new lamp.
Coil on left side of the break ends near bottom of picture, facing away from camera, with a tungsten ball that must have formed at the moment lamp burned out with about 5 amps of current.

These experiments can be conducted (no pun intended) by anyone with a burned-out lightbulb and a high or low frequency NST.  I have had cases where arcing happened between external electrodes before any action where tungsten filament used to be.    Clear bulbs with plain argon fill can make very pretty stationary arcs shaped like, uh, arches.

[Twospoons]

I think xenon is  too expensive to use in filament bulbs.  Usual fill in tungsten halogen lamps is argon plus iodine.  The iodine is what allows the lamp to run hotter, by binding to any evaporated tungsten then depositing it back on the hottest part of the filament, which is usually also the thinnest part.

[davekni]

I think xenon is  too expensive to use in filament bulbs.  Usual fill in tungsten halogen lamps is argon plus iodine.

Small flashlight bulbs used to be available with krypton fill for improved efficiency (lower thermal conductivity).  I don't know if halogen bulbs ever used krypton.

[johnnyzoo]

I think xenon is  too expensive to use in filament bulbs.  Usual fill in tungsten halogen lamps is argon plus iodine.  The iodine is what allows the lamp to run hotter, by binding to any evaporated tungsten then depositing it back on the hottest part of the filament, which is usually also the thinnest part.

I think xenon is  too expensive to use in filament bulbs.  Usual fill in tungsten halogen lamps is argon plus iodine.

Small flashlight bulbs used to be available with krypton fill for improved efficiency (lower thermal conductivity).  I don't know if halogen bulbs ever used krypton.

I have read that most halogen lamps have xenon as the bulk gas, but I'm not sure how reliable my sources were.

Some high performance incandescent flashlight bulbs were also advertised as "xenon" bulbs.

[klugesmith]

It would be telling to see the emission spectrum of that tiny white arc. Here's one reference whose simulated spectra include the noble gases.
https://www.astronomy.ohio-state.edu/pogge.1/Ast350/Labs/Lamps/index.html

Now where did my handheld diffraction-grating spectroscope go?  From study of near-infrared LED's a few years ago, using 2nd order diffraction of visible lines as wavelength reference.
 https://highvoltageforum.net/index.php?topic=832.msg5632#msg5632

[Twospoons]

you would need to run much lower current to get a good emission spectra, otherwise it will be swamped by black body radiation.  I've seen the same blueish-white arc in a short gap with high pressure air.

xenon is roughly 100x the price of argon.   This is probably one of those things where high-performance or specialist bulbs get xenon, where ordinary, cheaper bulbs use argon.

[klugesmith]

Thank you for warning about current.   I will try some multi-megohm ballast R's in series with primitive NST on variac.

Yes, xenon is the most expensive gas distilled from air.   On the order of $10 per liter.   $0.01 per cubic centimeter.  Maybe more in small quantities, as for hobbyists and neon artists.

It does matter when you need hundreds of kilograms. 
When DAWN spacecraft was launched I figured it carried more than 1/2 million dollars worth of Xe.

Here's one report from 2019 about xenon global production, usage, and price trends.   
https://ntrs.nasa.gov/api/citations/20150023079/downloads/20150023079.pdf
The writers say that if electric space propulsion will demand 10s of tons of xenon per year, they need to start planning today with the air gases industry.
New large air separation plants (O2 capacity > 2000 tons/day) should incorporate rare gas units, or at least be built with the appropriate hooks.
If that report mentioned countries of origin, Russia and Ukraine would have been high on the list.