Due to the apparently lack of data and my curiosity, I was thinking of
throwing together a crude experiment. Two lamp bases, two 40 watt
clear envelope incandescent lamps, two SSS (solid state switches), and
some kind of pulse generator. 30 seconds on should be enough to get
the filament hot enough for sublimation. 30 seconds off should be
enough for it to cool down for a cold start (I need to check this with
an IR thermometer).
However, I have no intention of running this test for 1000+ hours.
Instead, an accelerated life test can be done with higher than normal
voltages.
<
http://www.welchallyn.com/documents/Lighting/OEM_Halogen_Lighting/MC3...>
For halogen bulbs, they use:
Life = (Vdesign / Vapplied)^12.0 * Life at design voltage
For a 1000 hr lamp running at 120% of the rated voltage, the life
might be:
life = (1/1.2)^12 * 1000 = 112 hrs
which is more reasonable for my tinkering and for Mythbusters. With
power cycling, the life will be even less. I should have a Variac
somewhere in my junk pile.
--
Jeff Liebermann
[email protected]
150 Felker St #D
http://www.LearnByDestroying.com
Santa Cruz CA 95060http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Hi SEB. Well I sent an email to Don klipstein on this topic. And
have permission to copy his reply.
<from Don K. below>
My apologies - I have been off Usenet for a little over a year.
Hi Don, I was wondering (out loud) if you still lurk on the
sci.electronics.basic usernet forum?
(So I figured I'd just drop you an email.)
We have a question about the aging of incandescent bulbs from being
turned on and off. You seem like quite a lamp expert and I was
wondering if you've ever come across any real data on the subject.
As for real data on effect of switching causing wear on
incandescents:
I know some data.
1: It is true that incandescents often failon cold starts. However,
I
know a mechanism where an aging incandescent becomes unable to
survive
a cold start a little before it becomes unable to survive continuous
operation.
I explain this in:
http://donklipstein.com/bulb1.html#how
And:
http://donklipstein.com/bulb1.html#wbt
2: In incandescent traffic signals, the bulbs for yellow last longer
than
for red and green. So even after being switched on and off about a
million times, on-time is still a significant factor in life
expectancy.
However, traffic signal bulbs have a more vibration-resistant
filament
than standard incandescents.
3: Flashing and chasing marquee lights used to be incandescent until
cold cathode CFLs became economically available. Cold cathode CFLs
are special CFLs that are dimmable and blinkable, and flashing them
does not detract from their life. However, their efficiency is less
than that of hot cathode CFLs. Some marquees still use
incandescents.
4: I did an experiment to check for voltage drop in one of those
now-hard-to-get thermistor-type life-extending "buttons" to attach
to the tip of the base of an incandescent. The device claimed to
double the life of the bulb. I found enough voltage drop to account
for 50-60% life extension. Light output was reduced 13%, and power
consumption of the combo of the bulb and the thermistor was only
2.05%
less than that of a bulb connected directly to the line.
5: In my experience, incandescent holiday lights that blink last
longer
than those that don't. However, most of my experience is with low
voltage bulbs whose short filaments are probably sturdier.
6: Some incandescents make an audible "ping" when switched on.
However,
deflection of the filament does not necessarily strain the filament
past its "endurance limit" - the threshold of causing metal fatigue.
7: Some railroad crossing signals have a set-up where bulbs have a
resistor added in series with them for the first half second or so
that they are on. However, this may be done because of how serious
the problem is widely said to be, and how serious it actually is
appears
to me to be much less.
8: One of my friends had a bathroom fixture with a high wattage bulb
that
was constantly run dimmed by a dimmer. He experienced little gain
in
life extension. Since his bulbs significantly audibly buzzed, I
suspect
his filaments resonated at the power line frequency or one of the
power
line frequency's lower harmonics.
9: My mother had some incandescent nightlights that had diodes in
them
to significantly dim the bulbs - which should have made the bulbs
last
for decades. However, they did not.
One thing I noticed: These 4-watt bulbs had extremely thin
filaments,
and with a diode and therefore being off half the time 60 times a
second, their temperature varied greatly up and down 60 times a
second.
I could see that from rolling my eyes up-and-down while looking at
them.
Maybe the filaments at times resonated at the power line frequency
or
a lower harmonic thereof. That could easily produce sound too weak
to
hear from more than a couple inches away, because such low wattage
120V
incandescents have a vacuum rather than a gas fill.
Or, maybe those bulbs do not do well with DC due to high voltage,
vacuum, and the ends of the filament being close to each other.
Please see:
http://donklipstein.com/bulb1.html#dc
======================
Overall, I am seeing generally that cold starts are not nearly as
bad as many say they are, but in a few bulbs they can be. The data
appears to me to be majority in favor of "little to generally no
problem from cold starts", but it is incomplete.
I would also suggest reading:
http://donklipstein.com/ltrouble.html#i
====================
Hope this helps!
- Don Klipstein (
[email protected])
