Please don't incriminate me with that all-inclusive "WE". Marketing types decide what will be on the shelves and they never ever listen to what I say.
Of course there is a valid reason for putting LED emitters and supplies in a bulb-looking package, which is that until very recently those bulbs were the only show in town, and the luminairs that used them were what there was. So there exists a massive installed base of edison based light sockets, and that is where a whole lot of the replacements must fit. While change delivering an improvement is good, not everybody is ready to scrap all their present lighting and replace every bit of it with new LED systems. So the present obstacle is the current installed infrastructure, and that will not go away rapidly, no matter what congres legislates.
While it is not reasonable to manfdate a lower price, it certainly should be possible to mandate a minimum lifetime, and simply not allow products of inadequate quality to be sold for any price. OF course, determining that a product has an adequate lifetime would be a bit of a challenge, no question about that. But assuring adequate product life is the best way to gain acceptance.
Yes, the board may go out, but the LED, not so. In this circuit, it looks like the fuse is before the rectifier. I would assume it to be large enough to remove or change. Every company that would deal with such a device should have someone adept enough to at least open a box.
Most likely, the "fuse" will be a tiny, surface mounted device the size of a Tic-Tac and buried inside a non accessable cavity. Even if you could mange to get to the fuse, the replacement is only available by the 1000 piece reel from a mainstream electronics distributor.
While safety and circuit protection are two important aspects in the overall design, todays mainstream LED drivers suffer from (IMHO) two BIG problems - low efficiency and bad location. Most of the drivers I've seen have horrible power factors and efficiencies in the 70% range. How in the world can we be promoting LED technology with its huge efficiency gains and think a 30% loss in power conversion is acceptable.
Furthermore, Why do we continually "shoehorn" this inefficient converter into a tiny little oven heated by the LEDs. All because we insist on making this "new" technology appear like a 100 year old A19 light bulb.
We need to get the electronics package out and away from the LEDs. It can then be larger, cheaper, operate cooler so it will last longer, and perhaps employ user replaceable circuit protection devices.
I think the assemblies are also susceptable to temperature swings - The traffic signals in North Charlston SC are LEDs, and most them have signifcant failures. In some cases, more than 60 percent of the LEDs in a single color have failed. Either that, or North Charleston got incredibly ripped off by whomever sold them the LED signals.
The dowside of using a fuse in a consumer application like this is that the fuse will never be replaced. Once the fuse opens the product will not be repaired by the consumer and the device will end up in the garbage. Yes, the fuse is reliable and inexpensive, but it really provides no protection for the rest of the circuitry because the entire device will be disposed of when the fuse opens.
Using Siemens NX software, a team of engineering students from the University of Michigan built an electric vehicle and raced in the 2013 Bridgestone World Solar Challenge. One of those students blogged for Design News throughout the race.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.