Experimental results for current/power surge in a lightbulb

[freefaller]

Hello everyone,

I have been working on my extended essay (IB) in physics these past few months, and my topic is the current surge that occurs when a light bulb is first turned on, due to temperature.
I have finished all the math/theoretical part for this, and long story short, as I expected, the graph (current*time) is supposed to be hyperbolic decrease which reaches an equilibrium.

Anyways, once I went started my experiments, using a dc power supply, light bulb + socket and voltmeter, this effect that I predicted only happened with certain odd conditions. When I had the circuit complete and then stuck the power supply's cable into the wall outlet, then I recorded the current surge which then settled at equilibrium (using vernier, fyi). But when the power supply was already plugged into the outlet and I completed the circuit, the current just jumped to a stable value without a surge.

I find this rather strange.

Does anyone have an explanation for this, because I certainly don't see the reason why this would happen? I tried fiddling with connections and all, and waited long enough for the bulbs to cool. Is it possible that there is a power surge only when the dc power supply is turned on, but shouldn't a light bulb always have a current surge since it isn't ohmic? Which condition applies more to a regular light bulb connected to a building's current supply?

Thank you for your help.

 

[marcusl]

A classic 110VAC bulb is connected to a voltage supply (power mains) of very large capacity (15A typical). The inrush current, which may be around 10 amperes, can easily be supplied by the supply without loading it down. What you see on the bench with a DC supply, on the other hand, will depend both on the characteristics of the bulb and of your supply. For instance, suppose you have a variable supply with a 1A rating that you set to voltage V0, the nominal operating voltage of the bulb. If the inrush current of your bulb is, say, 10A, then the voltage out of your supply may sag to the point that it is able to put out 1A. Only one-tenth the expected power is dissipated in the filament, so it heats more slowly than expected and not in the way expected. Your curve will be significantly off. Some supplies behave more badly when overloaded, and won't even supply the 1A that they are rated to in the face of a near-short circuit.

It is hard to diagnose your experimental results without detailed information. To proceed, you might measure the cold filament resistance with an ohm-meter, and calculate the expected inrush current. Make sure that your supply is overrated so that it can supply more than that inrush current. This may solve your problem. You can also provide us with details of what you did and measured, for more help.

 

[truesearch]

If I understand you correctly you are looking for the current surge that results from the lower resistance of the filament when it is cold.
When the filament is hot the resistance is higher and the operating current is lower than the current at switch on.
This change occurs very quickly in my experience (less than 1 second) and when I demonstrate it I need to use an oscilloscope or fast data recorder to make measurements... how did you make your measurements?

 

[freefaller]

Thank you for replying.

Truesearch: I was using a Vernier probe to graph the current and the frequency of measurements I think was 20 per second. I did try higher frequencies but then the graph was pixelated and fuzzy and with too much noise. Is 0.05s precise enough in your opinion?

Marcusl: are you saying that the power supply might not have been able to actually supply enough current? It seems like a pretty good point. The power supply I used gave up to 30V and drew from a power outlet. I used 25, 40, 60, 75 and 100 W bulbs but there wasn't much difference in the shape of the graph. Sorry, I don't have any experimental results with me at the moment, and I actually discarded most because I wasn't seeing the inrush effect. However generally the current never went higher than half an ampere, although I don't think I can directly control the current output, only the voltage.
Maybe what I should try then is very low wattage bulbs, like flashlight bulbs, and then the effect might show up?

Anyways, I don't know if it is any help but here is a sketch of the shape of the current flows I was getting. The first graph shows what would happen when I completed the circuit. The second is what I was expecting to get and what I actually got but only when I plugged in the power supply into the outlet when the circuit was already complete.

https://dl.dropbox.com/s/56ffb78mgp55n9a/sketch.png?dl=1

 

[marcusl]

Oh! You should definitely match the supply voltage to the rated voltage of the bulb. If you use a flashlight bulb, set your supply to 3V and try again.

BTW, what's a Vernier probe?

 

[freefaller]

I talked to my physics teacher, he thought the insufficient current explanation was valid for the problem I was having. I'll try with a 9V battery and a flashlight bulb. The temperature will be similar, but the different resistance and voltage will allow the smaller amount of current to go through.

Vernier makes a number of different probes that can be used all with the same interface and computer software. It is maybe more suitable for the classroom but I think it quite good.
Here's a link to their site I guess: http://www.vernier.com/products/sensors/

 

[marcusl]

Thanks.

I suggest matching the bulb and supply--use 2D batteries for a standard 2D flashlight bulb, for example.

 

[Redbelly98]

I recommend using a car headlamp and 12V battery that can supply the necessary current. The thicker filament used in these lights takes longer to heat up, so it will be easier to measure the current surge at 20 readings per second.