The science of flying molten tungsten

[MathYew]

Hi,

a friend of mine - a college physics professor, and me, have had a discussion about an experience he had in his early school days, and we haven't been able to prove or dissprove what he thinks he saw.
During an incandescent light burn-out (12V bulb on 220V =), he thinks he might have seen a drop of molten bulb wire (tungsten) fly (or melt its way) right throug the glass bulb housing and on to the floor, leaving the bulb intact. (but not the parquet :)
My gut feeling is, that what he saw, probably came from the outside of the bulb, but frankly, neither of us is able to make any foolproof physical calculations to rule out the "penetration hypothesis". :)

What's your oppinion on this?

bye
Matej

 

[Bob_for_short]

It's easy. If the bulb remained intact, then, apart from interieur damage with high current, there was an external one - on the bulb contacts. They also may melt and give droplets of flying metal. I do not recommend doing this. Any short-cut may produce unpredictable dammage.

 

[MathYew]

I agree, but his explanation at the moment is that the droplet, smaler than the thickness of the glass, might have made it's way through, sealing the hole behind it and not leaving a noticeable mark. (He didn't investigate thoroughly...)
Sounds incredible to me, but it'd be fascinating if it was true, don't you think?

 

[Danger]

it'd be fascinating if it was true, don't you think?

Yes, it would, but I really don't think that it happened. I believe that Bob is correct. On almost any bulb that I've seen there's a small deposit of solder at the top of the metal section right where the glass goes in. It connects the 'ground' side of the filament to the case, and could easily melt under such overload conditions.

 

[Vanadium 50]

There's also much, much less tungsten in a lightbulb than a typical drop.

 

[Borek]

There's also much, much less tungsten in a lightbulb than a typical drop.

That doesn't have to be a problem - if the tungsten wire breaks small piece can drop out - it doesn't have to be a "drop" size.

 

[MathYew]

I just thought of another clue, which I think disproves the possibility that the spark came from the inside of the bulb... As my college put it, it flew out in an arc. Glass even if it was molten, would still be quite viscous and so, would probably stop the droplet completely.
Well whatever. I think my colleague is a good professor and a good physicist... in his field... :) Had he given this thing any thought, he'd realize, his conclusion was probably false. But I'm not the one to say...

 

[Vanadium 50]

Well, I was thinking "drop size" was related to visibility. But now that I think of it, there's probably some complicated interplay between weight and surface tension that determines how big a drop is. It would be interesting to compare the amount of tungsten and the minimum drop size.

 

[MathYew]

I think the sizes of drops in this case would be at least partially determined by the fragmentation of the filament during the "burning-out" and would probably depend on the weak spots in material - thicknes and local resistivity. But also on a thousand of other more kinetic factors, I think. But what fascinates me personally is the physics of "melting - through" a layer of matrial. How much less (I suppose) energy is required for a hot sphere to pass through ice in comparison with the energy required to melt a column of ice of the same hight and radius... for example. Some frighteningly concrete physics. =)