Is flipping the switch on an old light bulb more expensive than letting in run for a short time?

Thread Starter

c627627

Joined May 18, 2011
63
In the old days, did switching an average old light bulb consume more energy vs. keeping it running?
In other words, is running an old light bulb for X amount of (short) time cheaper than turning it OFF then ON again, meaning, is it a myth that the act of flipping the switch costs more than having the old bulb running, if we're talking a relatively short amount of time?
 

crutschow

Joined Mar 14, 2008
34,285
the thermal shock to the filament could shorten the life of the bulb.
I think that's a myth also.
Why would the thermal shock shorten its life?
The filament may open from the thermal shock, but that's only after it has been weakened to the end of its life by the evaporation of the filament material, making it more fragile.
 

oz93666

Joined Sep 7, 2010
739
Hmm...Searching gets a lot of hits on this , many to do with lighthouse lamp filaments ..

"1,000-watt, 120-volt, 20-5 lighthouse lamp with tungsten ...... In addition, the thermal shock of switching is quite high for these heavy filaments and, therefore flashing would seriously shorten the lamp life .."

A 1,000w lamp has thicker filamant and is probably more prone to failure because of this .... the main problem seems to be the big surge of current when switching on "10 times normal" since the filament resistance when cold is much lower ....

The consensus seems it would not save money to keep it on ... the obvious solution is to fade in the voltage . If using DC just put a capacitor across the lamp ... If AC a dimmer switch.
 

MrAl

Joined Jun 17, 2014
11,396
In the old days, did switching an average old light bulb consume more energy vs. keeping it running?
In other words, is running an old light bulb for X amount of (short) time cheaper than turning it OFF then ON again, meaning, is it a myth that the act of flipping the switch costs more than having the old bulb running, if we're talking a relatively short amount of time?
Hi,

I think what you are referring to is a question about the surge current when the bulb is switched on.

The surge current is higher, much higher, than the normal run current. That is because the bulb filament is cold and the cold filament resistance is much lower than the hot resistance and therefore consumes more energy per unit time than when running normally.

However, the time duration of the surge is limited, therefore there will be a break even time period for the turn on energy vs the run time energy.

Let's look at an example but let us also keep this simple.
We have a bulb that takes 10 times the current to start up when switched on than when running normally. The power at the line voltage would therefore be 10 times the normal running power. Therefore the break even energy use point is when the bulb is run normally for 10 times longer than the surge duration. This would mean that keeping the bulb ON and leaving it on for more than 11 surge duration times would start to save energy and therefore we assume money saved.
Now that means if the surge duration is 0.5 seconds the light would have turn on and stay on for more than 5.5 seconds in order to start to save money. If the surge duration was 1 second, it would have to stay on for more than 11 seconds to start to save money.
Since most regular household bulbs start up in less than 1 second (we could look into this for other bulbs) i would think we would always start to save money by keeping it on for more than 11 seconds otherwise we would lower the usage efficiency.

However, if we only intend to use the light for 2 seconds, then it makes no sense to keep it on for a full 11 or more seconds because we obviously use more energy the longer the bulb runs whether it surges or not. So the only way to save money is to alter our usage pattern. Consolidate the chores being done using the bulb into time frames that run at least say 12 seconds but no less. This would mean less energy expended over long time periods.
Note that the break even point could be much shorter than this too so the time frames may only have to be 5 seconds or even less.
 

Tonyr1084

Joined Sep 24, 2015
7,853
As already mentioned, flipping on a lightbulb will draw more current than when it has been on for any period of time. Also as mentioned, it has to do with filament temperature and resistance. IF you switched a light bulb on and off really fast, lets assume half the time it's on and half the time it's off - if the filament doesn't have time to cool then its resistance is much higher. Therefore, it's almost the same as just leaving it on all the time.

So, no, switching a light bulb on and off will not save energy. The bulb will operate at a lower temperature and lower average current; but you won't save money. It would be almost the same as putting the bulb on a dimmer. Reduce the current and you reduce the cost.

To address the "Fatigue" question, a bulb being turned on and off very fast will not give thermal shock to the filament because it never cools to room temperature. Essentially it's on half the time (assuming equal amounts of time on and time off), drawing half the current. Yeah, the filament will not be up to full temperature, therefore the current draw WILL be higher, but during the time period it's off - it draws no current, but also does not cool to room temperature.

Attended a lecture where the speaker used xenon strobe lights to illuminate his lecture. The talk was about 30 minutes. The strobe rate was about 30 strobes per second, each strobe lasting 5 milliseconds. So 5 mS x 30 per sec, x 30 minutes - the lamp was "LIT" only 4 1/2 seconds. The initial impression is "WOW! WHAT A COST SAVINGS". However, during the time the lamp was off, energy was being stored to fire the strobe. There was no cost savings by having a light flash that short and that fast for that period of time. In fact, when it comes right down to it - the cost was a little higher because of inefficiencies within the circuitry. Energy lost to heat in the form of resistance within the circuitry and within the wiring.
 

MrAl

Joined Jun 17, 2014
11,396
As already mentioned, flipping on a lightbulb will draw more current than when it has been on for any period of time. Also as mentioned, it has to do with filament temperature and resistance. IF you switched a light bulb on and off really fast, lets assume half the time it's on and half the time it's off - if the filament doesn't have time to cool then its resistance is much higher. Therefore, it's almost the same as just leaving it on all the time.

So, no, switching a light bulb on and off will not save energy. The bulb will operate at a lower temperature and lower average current; but you won't save money. It would be almost the same as putting the bulb on a dimmer. Reduce the current and you reduce the cost.

To address the "Fatigue" question, a bulb being turned on and off very fast will not give thermal shock to the filament because it never cools to room temperature. Essentially it's on half the time (assuming equal amounts of time on and time off), drawing half the current. Yeah, the filament will not be up to full temperature, therefore the current draw WILL be higher, but during the time period it's off - it draws no current, but also does not cool to room temperature.

Attended a lecture where the speaker used xenon strobe lights to illuminate his lecture. The talk was about 30 minutes. The strobe rate was about 30 strobes per second, each strobe lasting 5 milliseconds. So 5 mS x 30 per sec, x 30 minutes - the lamp was "LIT" only 4 1/2 seconds. The initial impression is "WOW! WHAT A COST SAVINGS". However, during the time the lamp was off, energy was being stored to fire the strobe. There was no cost savings by having a light flash that short and that fast for that period of time. In fact, when it comes right down to it - the cost was a little higher because of inefficiencies within the circuitry. Energy lost to heat in the form of resistance within the circuitry and within the wiring.
Was that lecture about the human eye vision persistence or similar?
 

Reloadron

Joined Jan 15, 2015
7,501
Has anyone actually ever measured the time for inrush current on an incandescent lamp? One of those things I have thought about but never actually looked at. I have a 100 watt incandescent lamp laying here. I actually plugged this bulb into a Kill-A-Watt meter and when lit it is actually a true 100 watts. The cold filament resistance is right about 10 Ohms so doing the math with a measured 120 VAC 60 Hz applied the bulb will draw 120 Volts / 10 Ohms = 12 Amps. Once the filament is hot the current will be 100 Watts / 120 Volts = 0.833 Amps. Keep in mind the current is changing direction 60 times per second. I guess we could power the bulb with DC pretty easily if we wanted to but the question is how long before that 12 Amps cold start drops to 0.833 Amps?

Need to consider also the bulb is doing what and being subjected to what it was designed for.

In the old days, did switching an average old light bulb consume more energy vs. keeping it running?
I would say no simply because the time it takes for the filament to reach temperature is negligible. That even considering I haven't a clue how long it takes for the filament to reach temperature. Maybe someday I'll get around to measuring that. :)

Ron
 

Reloadron

Joined Jan 15, 2015
7,501
See my link early in the thread. The Mythbusters did measure the inrush current and the energy dissipated during 'start up'.
Thanks, on the first read I got too interested in parachutes. Now I see where it gets to lamps. OK so they came up with Incandescent: 0.36 seconds or 360 mSec. or about 21.6 cycles at 60 Hz US power. Just over 1/3 second, actually longer than I would have thought.

Ron
 

MisterBill2

Joined Jan 23, 2018
18,179
As already mentioned, flipping on a lightbulb will draw more current than when it has been on for any period of time. Also as mentioned, it has to do with filament temperature and resistance. IF you switched a light bulb on and off really fast, lets assume half the time it's on and half the time it's off - if the filament doesn't have time to cool then its resistance is much higher. Therefore, it's almost the same as just leaving it on all the time.

So, no, switching a light bulb on and off will not save energy. The bulb will operate at a lower temperature and lower average current; but you won't save money. It would be almost the same as putting the bulb on a dimmer. Reduce the current and you reduce the cost.

To address the "Fatigue" question, a bulb being turned on and off very fast will not give thermal shock to the filament because it never cools to room temperature. Essentially it's on half the time (assuming equal amounts of time on and time off), drawing half the current. Yeah, the filament will not be up to full temperature, therefore the current draw WILL be higher, but during the time period it's off - it draws no current, but also does not cool to room temperature.

Attended a lecture where the speaker used xenon strobe lights to illuminate his lecture. The talk was about 30 minutes. The strobe rate was about 30 strobes per second, each strobe lasting 5 milliseconds. So 5 mS x 30 per sec, x 30 minutes - the lamp was "LIT" only 4 1/2 seconds. The initial impression is "WOW! WHAT A COST SAVINGS". However, during the time the lamp was off, energy was being stored to fire the strobe. There was no cost savings by having a light flash that short and that fast for that period of time. In fact, when it comes right down to it - the cost was a little higher because of inefficiencies within the circuitry. Energy lost to heat in the form of resistance within the circuitry and within the wiring.
The turn-on surge of an ordinary light bulb lasts about 3 cycles, about 42 milliseconds. So the higher current time is very small. This tale came from a lazy jerk who made up a story as an excuse for his lazy habit. It persisted because it got written in a neat page with no spelling or grammar errors. That is the way many folks still evaluate what they read.
Xenon strobe lights are very much different in almost every aspect.
 

Reloadron

Joined Jan 15, 2015
7,501
The turn-on surge of an ordinary light bulb lasts about 3 cycles, about 42 milliseconds. So the higher current time is very small. This tale came from a lazy jerk who made up a story as an excuse for his lazy habit. It persisted because it got written in a neat page with no spelling or grammar errors. That is the way many folks still evaluate what they read.
Xenon strobe lights are very much different in almost every aspect.
Now that number is much more inline with what I would have thought. That is a far cry from the 360 mSec Myth Busters came up with. I would also think the bulb wattage would figure in a little but not sure how much.

Ron
 

MrAl

Joined Jun 17, 2014
11,396
Thanks, on the first read I got too interested in parachutes. Now I see where it gets to lamps. OK so they came up with Incandescent: 0.36 seconds or 360 mSec. or about 21.6 cycles at 60 Hz US power. Just over 1/3 second, actually longer than I would have thought.

Ron
You can actually see some bulbs gradually light up, although it is fairly quick. The bigger ones take longer though.
 

MrAl

Joined Jun 17, 2014
11,396
All sorts of tid bits in there. Interesting stuff.

Ron
Just one caveat here when it comes to making bulbs last longer.
Bulbs that are hard to change are often replaced with bulbs that would actually have a higher voltage rating if used as a normal bulb would be used. This makes them last longer so they dont have to be changed as often.
 
Top