When a manufacture quotes 50,000 hours that does not mean any given LED will last 50,000 hours. It means in a population of lamps every 50,000 hours a failure occurs. So if you have 10 LED lamps in your house expect one failure every 5000 hours. (50,000/10).
Most LED bulbs being marketed to consumers through common retail outlets are just plain BS labeled; so don't expect anything remotely close to the rated hours of usage they imply. You get what you pay for. But the statistical odds of individual components in an large composite LED array going bad are probably beyond anyone's range of experience here. It would taking from ferreting out of the most reputable primary sources, and seeing is they have published documented tests or not. There are just so many variables of quality control, wiring, ambient heat, specific application. etc. I'd distrust any generic answer.
I'd say camera choice is a more relevant analogy, Alan. There have always been consumer cameras, right down to disposable cardboard ones, versus pro gear expected to last reliably for decades. Now its short-term cellphones versus just about everything else, although an amount of cheap plastic film camera mfg is still going on. Being a tool distributor, I was keenly aware of the distinction. I've known of workmen being fired on the spot for bringing a home center power tool to the jobsite. "how ya gonna get any real work done with that thing?" Often the most expensive option, twenty times more expensive, would pay for itself within two days due to its dramatically better efficiency. The fact a cheap substitute might be warranty replaced for free, over and over again if necessary, means less than zero if the labor rate is tied up running back and forth to Cheapo Depot or whatever getting replacements. The junk version ends up being the most expensive option nearly every time.
And in fact, many junk "warranties" have some very deceptive fine print, not in fact always honored. For example, you buy a tool with a "lifetime replacement warranty", and they change the color of the switch or some other silly thing every few months, and with it, alter the model number too. So you bring in your failed tool, and they tell you that specific model no longer exists, so cannot be warranty replaced. Happens all the time.
Likewise, I don't want to spend a whole lot of time and money and gasoline expense, let alone days of strenuous backpacking, guessing whether or not my camera is going to keep operating properly. Nor do I want enlarger bulbs going brown and popping during printing sessions, especially making big expensive prints. Needing to replace a bulb once a decade on average, for any of my enlargers, is certainly tolerable, and what I'm accustomed to. But the cheap versions of the same bulbs sometimes don't last half an hour. So it's important to recognize such qualitative distinctions, and whom sells which.
I'm used to it. A tube of caulking which guarantees lifetime performance costs $1.99. If you are unsatisfied with the product, simply return in person the unopened tube along with your original receipt to the customer service desk in Kabul, Afghanistan. They will gladly give you a free replacement tube. So it helps to bring your focus loupe with you when you shop, or preferably a microscope to read the fine print. You also need to know what "lifetime" means - life of the product? But if you choose the kind of caulk used for commercial construction, which costs five times more per tube, it comes with no warranty claims at all, but does have a lengthy tech sheet available with all kinds of distinct engineering and application specs. Anything so cheap that it's too good to be true ... well.... When it comes to consumer lightbulbs, I alway bring along a portable math calculator with a BS Coefficient button somewhere on it, to figure out the real number of minutes it might last.
What's the best caulking (in off-white color) for this application?
Thanks Don, I'll check it out.
50000/1000000 which mean a failure every three minutes.So in the total population of LEDs which may be say a million Does the expected failure rate become 50,000 /100,000,000?
I has always assumed that this was an average life before failure figure so 50,000 hours represented bulbs that lasted longer than 50,000 and some that lasted less
Thanks
pentaxuser
If the manufacture sell 50,000 lamps there will be one failure every hour of operation. No one operates a lamp 24/7. If most operate 4 hours a day, a lamp would fail every 24/4=:6x3 or every18 minutes.
Not really. Hardware failure statistics tend to follow what we call a bathtub curve: relatively many very early failures, then a long period of few failures, and then a high failure rate past a certain age. The curve looks sort of like \______/ , thus the name.
-tih
Second, specialty bulbs are exempt from the incandescent ban
Wiebold curve actually.
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?