# Melting a lightbulb filament

1. Jan 8, 2014

### elegysix

Does anyone know an approximate voltage needed to melt the filament of a typical 120V/60W incandescent bulb?

I'm doing a research project and I've calculated 211V. However, I don't have a power supply capable of delivering it - yet. So for now I can't test my theory.

Any help is appreciated.

2. Jan 8, 2014

### .Scott

Incandescent light bulb filaments don't fail by simple "melting". From what I can tell, the tungsten sublimates under too much heat creating spots along the filament that are thinner and therefore hotter. In a regular modern light bulb that sublimation process is offset by the gases in the glass envelope - even if the bulb is not a halogen.

I can tell you from many reports that when the neutral in a home circuit is disconnected from ground and the load on each side is unbalanced, light bulbs on the lightly loaded (and therefore higher voltage) side will start to pop. From that, I would guess that a standard 60W bulb will not long survive 211 VAC.

3. Jan 8, 2014

### elegysix

Thanks. I've been modeling their spectra, voltage, current, resistance, and temperature characteristics.

I want to cause the filament to fail, and I figured raising the filament temperature to the melting point of tungsten (~3683K) would do it. I calculated that I would need 141W using the stefan-boltzmann law, which translated into 211V after modeling the power as a function of voltage.

As you pointed out, I realize there are many more factors involved which I have not yet taken into account. But it's a starting point.

4. Jan 8, 2014

### davenn

not an accurate comment, considering I have lots of 40 and 60 and 75W globes in my house operating at 240V

what is missing is that globes are designed for a specific wattage at a specific voltage

cheers
Dave

5. Jan 8, 2014

### sophiecentaur

I believe that photoflood lamps used to be more or less normal bulbs but operated at about 50% over their normal voltage. They would not last more than a few hours at full volts but produce about twice as much light (nice and blue, too)
I just found this link which may be of interest.

6. Jan 10, 2014

### elegysix

Thanks sophie, that plot is interesting.

I did specify 120V/60W. I think his reply was in context.

Moving on, there was a microwave taken apart in our electronics lab this evening. I suspect that tomorrow I'll have the power supply I need to do this test. I'll update with my results if so.

7. Jan 10, 2014

### davenn

Sorry, Yes I missed that in your post
I'll go stand in the corner

Dave

8. Jan 10, 2014

### sophiecentaur

Amazing that those photoflood lamps have a colour temperature of 3200C (which must be more or less the filament temperature in a frosted envelope) whilst the melting point of Tungsten is 3422C. Not a lot of room for error in the supply volts before the filament just gloops.

9. Jan 11, 2014

### elegysix

Since I didn't get my power supply yet, here's some pictures of our work so far to keep you interested...

Here are the data sets and models plotted: (disregard the title, as we took data between 174mA and 550mA... I just haven't gotten around to updating that)

and here is a table with the numbers: