Preserving the lifetime of a bulb ?

[f3nr15]

Preserving the lifetime of a bulb ?

My father claims that is is better do leave a light bulb on rather than switch it on or off when used often because he says,
"Electric charges are very fast and damages the filament inside with each flick of the switch closing the circuit, even more damage than leaving the lightbulb on constantly, within the same period of time", effectively increasing the lifetime if the bulb, I really want to know, is this true ?

While on the same topic..
Is it true the greater the current in a conductor the greater the velocity of the charges ?
What is the most effective way of preserving the lifetime of a bulb ?

 

[bdrosd]

Incandescent bulbs are more likely to burn out just as they are turned on, so there is some truth to the recommendation not to flick them on and off more then necessary. However, if you leave the light on long enough, you are guaranteed to burn it out, so you can't go too far in that direction either. I suspect there is some time interval where it is better to leave a light running rather than turn it off and then back on again, but I don't happen to know that value. Maybe there is some expertise out there??
With regard to your velocity question, the answer is "yes but not much". The average velocity that electrons need to have to conduct current in a wire is surprisingly small; something like millimeters per second. However, electrons in a metal travel at much greater speeds (10^6 meters/sec) in random directions due to their thermal energy. With no net current flowing, their average velocity is zero since just a many are moving left as right. It is true though that if the current increases, the net average velocity will increase (but it is a small value compared to the random thermal speeds of the electrons)

 

[mezarashi]

My father claims that is is better do leave a light bulb on rather than switch it on or off when used often because he says,
"Electric charges are very fast and damages the filament inside with each flick of the switch closing the circuit, even more damage than leaving the lightbulb on constantly, within the same period of time", effectively increasing the lifetime if the bulb, I really want to know, is this true ?

From the latest episode of "Myth Busters" on Discovery (but seriously the MOST BORING myth ever), they tried testing something similar: "Is it better to leave your lights on or turn them off when you don't need them?". It turns out that on the overall average, you need to literally turn it on and off 10-20 thousand times for it to burn out. Too bad they didn't have specific information for incandescents.

But from my own frustrating experiences, those darn incandescents always always burn out when you switch them on. Why don't you test this at home? It'd be fun anyway. Oh, but in the end, the Myth Busters' verdict was that you ultimately save more money by saving power so keeping them off is a better choice economically, although you'll need to buy bulbs more often.

While on the same topic..
Is it true the greater the current in a conductor the greater the velocity of the charges ?
What is the most effective way of preserving the lifetime of a bulb ?

While the electric field causing these charges to drift propagates at the speed of light, the charges themselves move at a dismal micrometers per second scale in most cases. Not convinced? Me neither, so let's try calculating:

Taking that there's 64 grams of copper in one mole of copper and there are 6.023x10^23 atoms in each mole, every 64 grams of copper will have 6.023x10^23 free electrons (1 free electron per atom). Considering the density of copper at 8920 kilograms per cubic meter, we get something like 8.4830x10^28 free electrons per cubic meter. Note: 1 Coulomb = 6.2415x10^18 electrons

Now push 1 A DC current (1 Coulomb per second) through a typical copper wire with a cross-sectional area of say 1mm x 1mm = 1 micro meter square, and we see that we need a net drift velocity of... 7.35 micrometers per second. Anybody want to verify?

 

[russ_watters]

Since incandescent bulbs cost much more to operate over their lifetime than they do to buy, it is a big waste of money to try to prolong their life by leaving them on.

Lets say that by not switching off a light bulb, you double it's life...

With a nominal life of 750 hrs, doubling that would be a 2 month of life at a cost of $.50 for the bulb and $9.75 per month for the electricity (at 13 cents per kWh).

If you had the light off half the time, it would still last 2 months and still cost $.50 to replace, but the electricity would cost $4.88 per month.

 

[lpfr]

I think that I can give some information about how a bulb burns.
First, in normal working, the temperature of the filament is about 2500 K. At this temperature the partial pressure of tungsten is enough for the filament to sublimate slowly. To slow this process heavy inert gas is added in the bulb (argon, krypton). But this do not stops the sublimation. If you look at an old but still working bulb you will see that it is darkened. This is due to tungsten deposed on the inner surface of the bulb. The filament gets thinner as it ages.

Second, and most important. What happens when you light-on? The electric resistance of tungsten varies strongly with the temperature. At working temperature, the resistance of the filament is about TEN times higher than at room temperature. This can be verified easily: take a bulb, measure its resistance with an ohmmeter and compare whit the value computed from the nominal power and the mains voltage.
This means that, when you switch-on, the current is about ten times the working current. In less than a tenth of a second the filament acquires the working temperature.
Now, suppose that the evaporation of the filament is not strictly uniform. There is a place where the filament has evaporated a little more than elsewhere.
This evaporation is a cumulative process. In the place where the filament is thinner, resistance is higher and power and temperature also higher. Then evaporation is also faster. To be true, there is a partial compensation. Hotter zones radiate more power (proportional to [tex]T^4[/tex]) and the temperature raise is lesser.
Whet can happen is a runaway process at switch-on. The thinner part of the filament heats more and faster than the rest which is still "cold". As it gets hot, resistance increases and power increases ([tex]I^2R[/tex]). The current is limited almost only by the (low) resistance of the rest of the filament and is, say, ten times normal current. the filament evaporates readily and can blow up this time or next one.
The situation is worst when the switch-on happens near a maximum of voltage in the AC cycle. If the switch-on happens when the mains is at a low, the heating of the filament is more progressive.
I agree with bdrosd that there is some optimum time to leave a bulb alighted. But I do not know what it is.
Maybe the better way to prolong the life of a bulb is to light it always with a variator (a dimmer). It allows the stating current to be low and lengthens the lighting process.