quartz halogen lamp melting in projector

[mr_big]

Has anyone heard of this happening before I work at an elementary school as an electrician and I went to replace a quarts halogen 36 volt lamp I put the new lamp in and turned it on it started to turn completly black and silverish and the bottom of the quartz was pushed out any help would be great

[pSlawinski]

I have seen this before.  Was this a cheap Chinese lamp?  I think the black deposits on the lamp from the filament caused a hot spot on the quartz.  Halogen lamps are filled to positive pressure in order for the halogen cycle to work properly.  Clearly the hot spot was hot enough to cause the quartz to soften, and bulge out from the inner pressure.

[mr_big]

This was not a cheap chinese lamp it was made by the GE Company

[pSlawinski]

Oh, I was just curious.  My explanation still applies to the GE lamp.

[Mónico González]

I've observed this same phenomenon two years ago in a 16mm movie projector.
These machine has a ELC type 24v/250w reflector lamp that have been working properly for a couple of years.
As I said, two years ago, before a movie session, I was setting up the projector and threading the movie. When I turned on the lamp before the film were threaded, the light output over the screen was normal and uniform, but after a half-minute or so, I observed that the bright of the empty frame (without film) was slowly and inexplicably fading out.
I turned off the lamp and motor very quickly and opened the lantern of the machine. When I watched the lamp I was too surprised, because the bulb was completely bulged and internally blackened. The filament did remains intact and showed continuity when measured with an ohmeter, but the lamp was very near to explode.
It seems like the quartz envelope was starting to devitrify, perhaps due to an insufficient cooling, however, the air flow from the fan turbine over the bulb was very strong.
I did change the lamp by a new brand one that did work right to date.
I never understood very well why a such lamp could fail in so way.

Regards.
M. Gonz?lez.

[pSlawinski]

Perhaps over time the tungsten is deposited unevenly on the filament.  That would leave thinner spots on the filament.  The thinner spots on the filament run hotter causing more and more tungsten to evaporate from the filament.  Eventually the rate of evaporation becomes too great for the halogen cycle to keep up with.  Tungsten is deposited on the quartz causing a hotspot that heats up the quartz to its softening temperature.  The positive pressure inside the lamp causes the softened quartz to bulge out.

[James]

The bulb of a halogen lamp will begin to blacken any time when the speed of evaporation of tungsten from the filament begins to go faster than the rate at which the halogen cycle returns this tungsten back to the coil.

In a very high proportion of Chinese made lamps this occurrs because the manufacturers have no understanding at all of the chemistry in the halogen cycle, they invariably guess at what gas filling to use, and often get it wrong.

However the problem will also occurr in every good quality Western lamp if the lamp is run outside the specified conditions.  The halogen cycle will stop functioning if the bulb wall temperature drops below around 250 degrees celsius, or rises above about 800 celcius (the precise limits depening on the chemsitry of the halogen fill that is employed).  Thus, if air vents in a projector are obstructed and the lamp overheats, you will see this effect.  Similarly if the lamp is dimmed to the point that it cools and and the halogen cycle stops, it will begin to blacken.

The blackened region then begins to absorb the radiant heat from the filament, consequently heating the quartz to its softening temperature, when it will swell up due to the positive internal gas pressure.

Best regards,

James.

[mr_big]

Is it possible for this to happen in a lamp that had just come out of the package

[pSlawinski]

Yes.

The bulb of a halogen lamp will begin to blacken any time when the speed of evaporation of tungsten from the filament begins to go faster than the rate at which the halogen cycle returns this tungsten back to the coil.

[mr_big]

bad transformer I should have checked the voltage at the sockets before replacing the lamp