Re: Auto head light failure statistical distribution
- From: "TKM" <nomail@xxxxxx>
- Date: Mon, 02 Jul 2007 12:50:09 GMT
"TimR" <timothy42b@xxxxxxx> wrote in message
news:1183361888.228412.258610@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Recently one headlight on my BMW burned out, so I had them replace
both. My theory is probably the other one is getting close, and this
way I feel safe for a longer period of time. I live temporarily in
Germany and speak the language poorly, so I try to avoid long
discussions with the Polizei. I do the same with tail lights, turn
signals, etc. Even ceiling lights - I don't climb a ladder twice if I
can avoid it.
The garage that put the headlight in gave me strong resistance to
replacing both. Their theory was then they'll burn out at the same
time, and I wouldn't be able to drive at all.
This seems rather unlikely to me. Average life span is an average. I
can't imagine it's very likely to have two go bad at the same time.
Is there an equation that predicts how likely the next failure will be?
While your sample size is tiny (just 2), your idea is fine and is the basis
for what's called "group relamping" in commercial, industrial and outdoor
lighting. In a large installation of lamps of the same type with the same
burning cycles, the number of failures over time can be predicted; but the
failure of a specific lamp cannot. Failures typically follow a normal
distribution, but that depends upon the size of the installation and the
type of lamp. Google "lamp mortality curves" and you'll get some articles
that illustrate the curves themselves; but I couldn't find the equations on
line. They're not in the current IESNA Handbook either.
I do the same thing as you do when a headlamp or tail lamp fails in my car.
I buy two and either replace both or carry the extra lamp in the car until
the other fails depending upon the cost of the lamp. Since I've had the
same car now for 13 years and note lamp failures in the service log, I can
say that 3 headlamps have been replaced and two each of the brake and tail
lights. It would be very unlikely that both headlamps would fail at the
same time even if they are the same brand and bought at the same time simply
due to variations in lamp manufacturing.
Your idea should work fine for incandescent and incandescent/halogen
automobile lamps. It won't if the lamps are LEDs where lamp life is very
long -- similar to the life of the car itself -- and more of a function of
light output depreciation. Discharge headlamps are expensive and I don't
know enough about their life characteristics to know if buying in
anticipation of failure makes any sense. I doubt it.
Terry McGowan
.
- References:
- Prev by Date: Auto head light failure statistical distribution
- Next by Date: Re: power of a cold cathode bulb
- Previous by thread: Auto head light failure statistical distribution
- Next by thread: Re: Auto head light failure statistical distribution
- Index(es):
Relevant Pages
|
