Why LED screw-base bulbs are a bad idea

[blasterman]

To mods: Another forum member here had a tag at the end of his response I found to be sort of an attack. It also shows the general ignorance and myopia of many people on the issue of lighting technology. My purpose of responding is not to defend myself not directed at anybody in general (not sure where the original statement came from), but hopefully will be educational and discussion worthy.

A tag at the end of LEDNinja read :

blasterman, stop whining about legacy fixtures. design a new energy star LED fixture that meets code and submit it to the powers that are considering alternatives to the good old Edison bulb.

First, who are these 'powers' ? The only powers I know of are VPs of marketing divisions of big box stores who in reality are really nothing more than Western Distributors for Asian factories trying to appease shareholders. There are already a jillion different types of MR-16 and standard base LED retrofits, and the concensus is that the cheapest ones found at Walmart are the best ones using the cheapest drivers that nobody knows if they meet code. Seriously, go look at the junk drivers on DealExtreme and tell me if you want this inside your ceiling cans. Guess what, most LED retrofits use this quality level of electronics. CFLs are actually worse because of the voltage ramp-ups required.

You quit your day job, and go to work at Walmart or target and design cheap consumer electronics for them with questionable U.L listings. The LED rods were throwing together here won't catch fire like a CFL will. The only reason LED retrofits exist is not because of some visionary at G.E. or Phillips wants to save the world, but because they know Western Consumers are basically stupid and will buy them.

Also, the science of fitting LEDs inside a standard base bulb basically comes down to eliminating heat with that package, and we've had that discussion resolved many times. There is nothing fancy here - LED retrofits are limited by how they can eliminate heat because of the format limitation, not how much light they can produce. That is universal knowledge.

However, every time a new LED bulb appears on the shelf at Walmart or K-mart with less lumens than a christmas ornament somebody here reviews it. Who cares? The bulb isn't going to get brighter because of the physical limitations above, and yet we get multiple responses from mensa candidates whining about how "LEDs aren't there" yet because that crappy bulb isn't as bright as their 75-watt Sylvania.

You could become the 'Bill Gates' of the new green lighting industry.
You can not expect 300 million Americans and 30 million Canadians to design and build their own LED setup.

That's like SUV owners complaining about the high price of gas and telling me to design a more fuel efficient engine as a response to me telling you not to buy a large vehicle.

I find it ironic that 30 years a typical high-school kid would have little problem wiring a simple DC circuit with LEDs, but now it's suddenly quantum mechanics. Says a lot about Western Education systems and why a 7th grade in India or Korea now has a four head start on your 12th grade honor student. Sounds to me like somebody just can't figure out how to wire 3-LEDs in series and what resistor to use.

They are going to go to a lighting store, buy something they like the looks of and hire an electrician to install and wire it.

Local furniture stores sell high end lamps that cost over $400, and they plug into a damn wall socket, and couples buy those things by the half dozen for their new condo. They then stick $1.00 light bulbs in them because CFLs look like garbage and can't dim. Track lighting that uses line voltage can also be installed without the assistance of an electrician, so I don't know what your problem is. Either format presents me with limitless possibilities of desiging LED lighting with lumen values as high as I want them.

Shoving LEDs inside 100year old light bulb formats is the problem, and a format I want nothing to do with because it's dead and beaten to death by big Box stores selling cat litter next to table saws.

The CFL industry is coming out with GU 24 fixtures and bulbs.
We need someone to do something similar for LED.

Yep, and they'll be cheap, built in Western China factories and almost certainly have the same problems as current CFLs.

Personally I could really care less what people buy, but perhaps the biggest slam against conventional bulb formats you seem to like so much is the fact corporate and industrial America stopped using them right after WWII. Seriously, go into an office building or warehouse or grocery store or factory anywhere on this planet and show me all the Thomas Edison era light bulbs, or GU / MR format bulbs being used. They almost entirely use Fl tubes, or HID, or increasingly solid state. Incan? Only if you run an art gallery. Now connect whatever brain cells are required to resolve that logic dilema.

There's a serious disjoint between true energy efficient technology adopated by corporate civilization and the whims of American consumers driven by laziness, horridly outdated home designs, and the Chinese backing your McMansion mortgage that looks just like 10 others on the street. The solution has nothing to do with what you buy in the aisle at Home Depot either even though marketing jerks are falling over themselves convincing you otherwise.

 

[jtr1962]

Re: Response to LEDNinja, and a few others

Great post, blasterman. I have to say I agree 100% here. I've long been saying that the insistence on sticking with the conventional screw-in lamps and/or triac dimmers is the biggest roadblock to adopting energy-efficient lighting of any type, whether it's LED, fluorescent, HID, whatever. I've been testing LED screw-in bulbs recently, and have come to the conclusion that the thermal envelope (i.e. maximum heat which can be dissipated comfortably) in the bulb form factor is perhaps 5 watts. So even with the best LEDs these days which emit about 1/3 of their output as light, that means no more than 7.5 watts driving the LEDs and ballast combined. A really good ballast might absorb 0.5 watts. That leaves 7 watts for the LEDs. 7x ~125 lm/W comes to about 875 lumens. In other words, right now we'll be lucky to match a 60 watt incandescent with LEDs. Even then it really won't be a great solution.

Why is their such an aversion to purpose-designed fixtures in residences? They've been used in industrial and commercial applications for decades. You could make a fixture operating the LEDs so they last a lifetime, and thus need no facility for replacement at all (although someone handy with a soldering iron could replace them if push came to show)? Is it really that hard to change out a fixture? And it's not like the screw-in fixture being replaced is generally expensive or permanent anyway. Half the fixtures I see being sold these days are garbage which will need to be replaced anyway in a decade. So why not just abandon the bulb format altogether in the replacement fixture? And while we're at it stop shooting for these ridiculously low prices which are the cause of a lot of dissatisfaction with alternative lighting. People are willing to pay $400 for a fixture because it "looks pretty", but then refuse to invest any further in the parts that don't show. Funny thing is years ago these types of products generally didn't exist in large numbers. People knew a good tool (and that's what a light fixture is) was worth the investment. The option to buy ultra low quality/ultra cheap just didn't exist. We need to get back to that sort of buy once to last a lifetime attitude. Not just in lighting, but in many other areas.

I find it ironic that 30 years a typical high-school kid would have little problem wiring a simple DC circuit with LEDs, but now it's suddenly quantum mechanics. Says a lot about Western Education systems and why a 7th grade in India or Korea now has a four head start on your 12th grade honor student. Sounds to me like somebody just can't figure out how to wire 3-LEDs in series and what resistor to use.

I think the problem isn't just one of education although I'll agree 100% we're really lagging when it comes to science and math compared to elsewhere. Rather, a HUGE part of the problem is a reluctance on the part of many to do anything they don't already know how to do. Look how many actually pay people to mow their lawn, for example. Us and a neighbor across the street are the only two who don't have gardeners. And this isn't a wealthy area. It's solidly middle class. Nobody wants to get their hands dirty, so to speak, whether it's mowing lawns, wiring together LEDs, or replacing fixtures. And when you get straight down to it, this attitude is at the root of why the US is losing to the rest of the world. Americans used to love to create, invent, attempt new things. You don't even have to go back very far to see when this attitude permeated the culture. A scant 40 years ago the can-do attitude was alive and well. Now when we want something new, we'll sit back and hope some other country invents it so we can buy it off them instead of creating it ourselves.

Anyway, great post and great topic for discussion!

 

[jerry i h]

Re: Response to LEDNinja, and a few others

Hmm, beg to differ, big time.
1) Yes, we do have exclusively FL's at work, but only because that is what was installed when our company first moved in to it's current location. Upgrading/improving is not an economic option.
2) You are moving into a house/apt that is already constructed and finished. You are more or less stuck with the electrical/lighting scheme already in existence.
3) you are confronted with a forest of E26 sockets, like it or not, in your new residence.
CFL? Yes, but it buzzes at 60 hz, often has funny colors, and cannot go into an enclosed fixture. LED? Yes, but it is costly and often has a limited lifespan. In this case, you are limited to what will screw into a conventional light socket. So, any substitute that is cheap is something that will work commercially. In this case, a stupid, old incan is often the best choice.
If you have the $$$ or contractor wherewithal, there are many alternatives limited only to your imagination: track lighting, spot lights, indirect fixtures, chandeliers, light strips, light panels, etc., powered by CFL, LED, HID, fire flies, etc.

 

[AnAppleSnail]

Re: Response to LEDNinja, and a few others

There are exactly 4 'permanent' light fixtures in my 2 bedroom apartment:
2 in kitchen, track lighting
4 in each bathroom (2 bathrooms) - 3 'vanity' and one normal.
3 in dining room, one fixture.

Everything else is floor lamps (From Big Lots, think Ikea on the cheap). I'd like to get real lighting furniture for the rest of the apartment - I use a 30-LED bulb in my reading lamp, CFLs all over, and that keeps the bills down some. LEDs are hard to make bright enough for full-room lighting, but...it'd be nice to have the option for it.

 

[curby]

Re: Response to LEDNinja, and a few others

Hmm, my post has changed a lot as I wrote it. In the end, I think our DIY solutions are currently the only possible way to do LED lighting well.

Regardless of how easy it is to do things, some folks don't have the desire to DIY. Some people don't like getting their hands dirty. As mentioned, many don't have the knowledge nor the desire to get it. While we can lament that fact and what it shows about our society and our education system, it is nonetheless reality. We can and should work to change that reality, but we must accept it for now.

So let's say we have Company X, which is committed to furthering the adoption of well-designed, semi-custom LED fixtures in high end new construction. This is pretty much the best case solution: wealthy buyers combined with flexible design constraints due to the blank slate that new construction affords.

To begin with, I don't think it's going to be possible to do truly custom designs. Correct me if I'm wrong as I'm not in the industry, but I don't think that you can make a business selling uncertified home-grown solutions. Sure you make things as safe as possible within the limits of your knowledge and ability, but I anticipate that getting each and every installation vetted and approved would take more time and money than any customer is willing to spend, given the alternatives.

So let's say you make a family of products for semi-custom installations. These LED bars, LED plates, drivers, and accessories are made to fit together LEGO-style and interoperate whether they are mounted behind mirrors, under cabinets, washing walls and ceilings, etc. You go to the effort of getting everything UL certified and ready to go, but the industry is moving so quickly that your designs are outdated before they hit the market. That said, you want to maximize your R&D dollars and exploit economies of scale by selling as many of your design as possible. What results is the current situation: there are companies out there building great lighting bars with efficient designs, good emitters, etc., but they are horribly expensive and out of date compared to what you could do with a soldering iron.

To summarize, it's not currently possible to have a successful business built around selling viable high-quality LED lighting products to the general consumer. IMHO, at least. Please tell me how wrong I am (or show me, by creating such a business).

Perhaps the best thing we can hope for is a transition from enthusiast designs (ours) to high end installations for those who are willing to swallow the initial cost. That in turn promotes awareness and appreciation for custom-designed solutions, and the price can gradually come down to meet the expectations and needs of the average consumer with semi-custom LEGO-style products.

Having a common base would really do a lot to drive adoption, and competing against an entrenched standard makes paradigm shifts difficult. Unfortunately a lot of domestic lighting currently uses the screw-base, which is a reality we have to deal with in spite of it being impractical for LED-based designs. We can do our best to promote custom designs, and we can hope for the development of low-cost light bars that are easily installed by laymen, but true penetration will not happen without some transition design that can get us from screw-in bulbs to where we want to be. There are simply too many fixtures out there with screw-in lamp bases, and we have to deal with the laziness factor.

Star-mounted emitters are a decent start to getting a common format, but we need the design to be easily replaced (if not due to burned out emitters then at least for upgrades to better emitter tech), and simultaneously extract heat while providing electricity (whereas every existing interface only deals with electricity and not heat). By making things modular and easily replaced, you drive down costs. Big LCD displays are expensive because you want every pixel lit, and you can't easily swap out a dead pixel. Likewise, you want to be able to replace lemons and old tech without prying up epoxy and resoldering a new emitter, as 105 out of every 100 consumers out there will balk at such an idea, especially after the shock of the initial cost of LED-based lighting.

 

[DM51]

Re: Response to LEDNinja, and a few others

The thread title is inappropriate and in violation of Rule 4. Please change the title to reflect an acceptable topic for this forum, or the thread will be closed.

 

[RyanA]

Re: Response to LEDNinja, and a few others

I think it is a good idea and there is a market for it. Led fixtures are already becoming common in boating where power conservation is important. I think it will become more important in people who live off grid as well, as well as those who want to conserve or simply save money.

I'm not certain that this thread does the best job addressing these pros. There are people out there that will and do agree, place your focus there rather than with those who do not. There is about as much sense in arguing as there is in a Mexican standoff. Besides, life is too short.

 

[Neondiod]

Re: Response to LEDNinja, and a few others

Star-mounted emitters are a decent start to getting a common format, but we need the design to be easily replaced (if not due to burned out emitters then at least for upgrades to better emitter tech), and simultaneously extract heat while providing electricity (whereas every existing interface only deals with electricity and not heat). By making things modular and easily replaced, you drive down costs.

Yes, the "star" is kind of the new standard in the lightning industry aimed for leds. If the star was both electricaly and thermaly connected by screws it should be quite simple to change the star. Like this scetch I made..

BR //

 

[J_C]

Re: Response to LEDNinja, and a few others

First, who are these 'powers' ? The only powers I know of are VPs of marketing divisions of big box stores who in reality are really nothing more than Western Distributors for Asian factories trying to appease shareholders. There are already a jillion different types of MR-16 and standard base LED retrofits, and the concensus is that the cheapest ones found at Walmart are the best ones using the cheapest drivers that nobody knows if they meet code. Seriously, go look at the junk drivers on DealExtreme and tell me if you want this inside your ceiling cans. Guess what, most LED retrofits use this quality level of electronics. CFLs are actually worse because of the voltage ramp-ups required.

The thing is, you advocate designing and wiring up something yourself which definitely doesn't meet code, unless you want everyone to hire inspectors to come out and assess it too.

You quit your day job, and go to work at Walmart or target and design cheap consumer electronics for them with questionable U.L listings. The LED rods were throwing together here won't catch fire like a CFL will. The only reason LED retrofits exist is not because of some visionary at G.E. or Phillips wants to save the world, but because they know Western Consumers are basically stupid and will buy them.

Just because you like UL listings, UL is not an oriental governing body so what is their obligation? Have you witnessed the hypothetical, seemingly nonexistant products you vaguely refer to catching on fire?

There is a simple reason retrofits are made, it's a bit dumb to buy an entire fixture when you have one that will suffice, IF you can accept lesser light output and some people can. The only remaining issue is the absurd prices because they are still seen as a luxury product.

Also, the science of fitting LEDs inside a standard base bulb basically comes down to eliminating heat with that package, and we've had that discussion resolved many times. There is nothing fancy here - LED retrofits are limited by how they can eliminate heat because of the format limitation, not how much light they can produce. That is universal knowledge.

You're not saying anything significant. Take any random fixture design and it too has "some" kind of inherent limitation. That incan bulb sizes limit amount of power means very little, people were already accustomed to having multi-bulbed or multiple lamps for more light output.

Perhaps the main issue is you are trying to insist an LED bulb should put out the same amount of light as some particular wattage incandescent, but why? Did the incandescent bulbls try to make the same amount of light as a kerosene lamp, a torch or 1 candle? No. There is no problem at all making a screw in retrofit LED bulb, only your insistence that if it doesn't suit your needs it couldn't work fine for someone else to suit their needs.

However, every time a new LED bulb appears on the shelf at Walmart or K-mart with less lumens than a christmas ornament somebody here reviews it. Who cares? The bulb isn't going to get brighter because of the physical limitations above, and yet we get multiple responses from mensa candidates whining about how "LEDs aren't there" yet because that crappy bulb isn't as bright as their 75-watt Sylvania.

How will you ever know the state of existing products if people aren't reviewing them? Who cares is probably most people on the 'site since it revolves around lighting, or if you don't care then I can't recall an obligation to read any and every topic.

So what if a person writes "LEDs aren't there"? Is it not true, what is the problem with that? It seems you are just being anti-social.

That's like SUV owners complaining about the high price of gas and telling me to design a more fuel efficient engine as a response to me telling you not to buy a large vehicle.

Yet, we can't deny that today's SUVs get much better gas mileage than they used to, designing a more fuel efficient engine did work. Not buy as large a vehicle is not a solution because if they didn't want a large vehicle there existed other options at time of purchase.

I find it ironic that 30 years a typical high-school kid would have little problem wiring a simple DC circuit with LEDs, but now it's suddenly quantum mechanics. Says a lot about Western Education systems and why a 7th grade in India or Korea now has a four head start on your 12th grade honor student. Sounds to me like somebody just can't figure out how to wire 3-LEDs in series and what resistor to use.

Sure, a high-school kid can kind of half-assed throw something together with basic instructions, but you can't seriously recommend they make AC line powered lighting, the knowledge and experience of product engineers does make your life safer for you, let alone some high school kid doing it for the first time instead.

Further, some people, most actually, simply don't want to do it and why should they? That's why finished products exist, not just the better result but because it is terribly inefficient, a gross waste of time to reinvent the wheel and DIY from scratch on every item one owns so the line has to be drawn somewhere.

Local furniture stores sell high end lamps that cost over $400, and they plug into a damn wall socket, and couples buy those things by the half dozen for their new condo. They then stick $1.00 light bulbs in them because CFLs look like garbage and can't dim. Track lighting that uses line voltage can also be installed without the assistance of an electrician, so I don't know what your problem is. Either format presents me with limitless possibilities of desiging LED lighting with lumen values as high as I want them.

Don't know where to even start on the above block of text.

1) Most people don't buy $400 lamps obviously but if they did, all the more reason not to throw them out and pay again for a special LED fixture.

2) The lamp is room decor, there is no requirement to pay a lot for the bulb as the cheap incan bulbs serve the need. Nobody really needs a bulb to burn for many years in a row if it is inexpensive. Nobody really needs a bulb to conserve electricity, people invariably use as much electricity as they can afford to and money saved on bulbs is money that can earn interest or be invested, and most people do not run most of their lights a dozen hours a day, nor need an ultimate brightest light possible.

Lighting as high as you want costs more money. The goal is not max light, it's inexpensive practical solutions to just get the job done, it's a light bulb not rocket science and should not consume anyone's time unless they are designing as a hobby or for mass manufacture.

Shoving LEDs inside 100year old light bulb formats is the problem, and a format I want nothing to do with because it's dead and beaten to death by big Box stores selling cat litter next to table saws.

... then don't buy them, lots of people just as informed feel they are a fine alternative if LED lighting is desired at all, although it is a bit dishonest when manufacturers mislead about the brightness. Put simply, it is not a "problem" that someone makes or sells a product that you personally don't want to buy, the same could be said about any product and/or any person's preferences.

Yep, and they'll be cheap, built in Western China factories and almost certainly have the same problems as current CFLs.

It is merely a light bulb, it does not need to be hypothetically perfect or last forever. Trying to take something simple and make it complex for the sake of personal desires is fine so long as you can accept others aren't on a quest for any particular lighting, they just need to see when it gets dark.

Personally I could really care less what people buy, but perhaps the biggest slam against conventional bulb formats you seem to like so much is the fact corporate and industrial America stopped using them right after WWII. Seriously, go into an office building or warehouse or grocery store or factory anywhere on this planet and show me all the Thomas Edison era light bulbs, or GU / MR format bulbs being used. They almost entirely use Fl tubes, or HID, or increasingly solid state. Incan? Only if you run an art gallery. Now connect whatever brain cells are required to resolve that logic dilema.

Oddly enough, I have no problems whatsoever using conventional bulb formats. None, nada, zip, zilch, and recognize that the reasons corporate and industrial America stopped using them are reasons not so applicable to individual consumer use. My ceilings aren't 30 foot high and my living room isn't the size of a warehouse, and I don't pay a maintenance man to spend a half hour coming out to unscrew a light bulb and replace it in 10 seconds. I have already taken more time replying to this post than it took to change every light bulb I've needed to for the last few years.

There's a serious disjoint between true energy efficient technology adopated by corporate civilization and the whims of American consumers driven by laziness, horridly outdated home designs, and the Chinese backing your McMansion mortgage that looks just like 10 others on the street. The solution has nothing to do with what you buy in the aisle at Home Depot either even though marketing jerks are falling over themselves convincing you otherwise.

Don't fall for the hype, we do not need ANY energy efficiency improvement. The sky is not falling, we don't need to excessively pollute to generate power (go nuclear), we can't even prove our actions contribute to global climate change, and it is intelligent to pause and remember these things.

It is not lazy to use existing tech that works, it is intelligent to accept and adapt to your environment instead of always wishing for things to be different and seeking problems where others happily co-exist without thinking about their light bulbs at all, except to occasionally put them on a shopping list along with a zillion other things bought in the same stores everywhere that sell them.

The solution is that there was never a problem! SO many many Americans have become brainwashed by those in power and by the media. You can pry my incan light bulbs out of my cold dead hands. If you want custom LED fixtures I'm fine with that too, it's your money and your time.

 

[jtr1962]

Re: Response to LEDNinja, and a few others

You're missing the point of this thread, J_C. While blasterman admittedly could have made his point a little more delicately, this is a very valid topic. I understand the frustration which he vented here, the main point being that it makes more sense to move to purpose-built LED fixtures, rather than to try to adapt LED technology to the problematic archaic screw-base socket. As someone who HAS tested quite a few LED bulbs, I have to say that so far I've seen nothing which really excites me in terms of price/performance. Nothing I've tested so far in a screw-base format has exceeded 300 lumens. Quite a few of the lamps were in the low double-digits as far as number of lumens go. Outside of one manufacturer so far, the longevity numbers haven't been that great. Two lamps using 5mm LEDs dimmed to under 70% in only 300 hours. And the ones that are doing well so far (they're using power LEDs) would probably be too expensive for the average consumer. So the bottom line as far as LED bulbs go are that they will likely be limited to the output of a 40 watt or smaller incandescent without active cooling, and they will either be expensive, or they won't last. Mass production will help, but it isn't going to change this basic fact.

You talk about simply using multiple lamps to obtain more light if the output of a single lamp is limited. And many existing fixtures do in fact have multiple sockets. However, think about the economics of this with regards to screw-in LED retrofits. A decent one will cost $40, and even then you'll be lucky if it produces 400 lumens. Now stick 4 of them in an existing fixture. That's $160 for 1600 lumens, tops. And because ALL bulb-based fixtures lose quite a bit of light by the nature of the light source, perhaps you'll end up with 1000 lumens where you want it. For less than that same $160 you could sell a purpose-built LED fixture which puts out more light, and which will run the LEDs far cooler. Instead of 4 ballasts (one in each lamp), and their associated losses, you only have one. You can have as many emitters as you want, constrained only by the size of the fixture. The amount of surface area for cooling goes up dramatically. The overall efficiency is greatly increased. And emitter lifetime goes up dramatically (from perhaps 25,000 hours at most to potentially 100,000, even 200,000). You don't even need to make any provision to change the LEDs when they last that long. At 4 hours a night, 100,000 hours is over 68 years of use. The only downside is you have to change out the fixture. Once. And then you don't have to think about it for decades. The most difficult part of doing this IMO is convincing the average consumer why this approach is better. The logistics of designing the fixture itself are relatively easy by comparison. You need to change the mindset so people think of a fixture and lighting as long-term investments, just as they do of their home or their car. People won't balk at paying IMO ridiculous sums for a new car because of the perceived value that car has. We need to do this with lighting. And it should be far easier to do compared to cars because a decent fixture won't cost tens of thousands or even thousands of dollars. Remember that 50 or 60 years ago cars were a hard sell, and now they're almost ubiquitous. It's really the marketing department who has work to do here, not the fixture designers. And it's not like most of the fixtures which would be discarded here in favor of LED fixtures are family treasures. A lot of the fixtures I see in houses these days and in Home Depot are garbage. Quite a few even use plastic sockets which melt even when run at the rated wattage. Chances are in ten years they'll have to be replaced anyway. That's how bad many are.

This is the point blasterman is trying to make here. The primary argument here isn't about whether LEDs should replace incandescents. We can save that for another thread but suffice it to say that the primary reason isn't reducing pollution/global warming, even though that's what's fed to the general public. This thread is based on the operative assumption that energy efficient lighting of some type should replace incandescents. Whether this is or isn't the right thing to do is irrelevant to the rest of the argument, which is namely that if LEDs take over, then we should abandon the archaic screw-based format once and for all. The screw-based lamp was invented primarily because incandescents don't last long. You need an easy way to change out lamps. There's no valid reason to continue to use this format if we decide to adopt LEDs.

While I agree with you that many (most) will not want or be able to build their own LED fixture, I also feel that this doesn't render the rest of blasterman's argument invalid. Purpose-built LED fixtures make sense on economic and many other levels. LED manufacturers should focus on making quality LED fixtures for residences, rather than LED screw-in retrofits. My own testing and knowledge of electronics tells me that screw-in retrofits are suboptimal on many levels. Their output is low, their cost is excessive relative to their output, many don't even last as long as incandescents. You say a it's a light bulb, it doesn't have to be perfect or last forever. Well, the primary selling point being touted for LEDs is indeed the fact that they have very long life. So while they don't need to be perfect, they should at least live up to the claims of long life. If they can't even do that because we insist on making them compatible with screw-based lamps, and/or making them inexpensive, then what's the point?

We've been down this same road already with CFLs. Yes, they work. Yes, they allow you to do a quick and dirty replacement of an incandescent with something more efficient. However, they often fail to live up to lifetime claims on account of the form factor and downward pressure on prices. They also require you to throw away the ballast electronics every time you change out a lamp, which IMO is beyond wasteful. And they're nowhere near as efficient as fluorescents running in purpose-built fixtures. Often the fixtures don't cost much more than putting 3 or 4 CFLs in an existing fixture. The tubes usually last far longer than CFLs. You don't throw away the ballast when the tubes need changing. The ballasts themselves are usually higher quality with fewer problems. Replacement tubes cost less than replacement CFLs. Besides all that, you gain perhaps 50% in efficiency. We've gone almost 100% fluorescent here (save for 3 chandeliers which I'll probably convert to LED in the next year), but the majority of the lighting here is linear tube fixtures. When I did the economics of it, it didn't make sense to buy CFLs just to save myself the bother of changing out fixtures. With decent LED screw-in bulbs costing far more than CFLs, the picture is skewed even more in favor of purpose-built fixtures. Now it's a matter of waiting for the lighting designers and general public to catch on to this fact.

 

[J_C]

Re: Response to LEDNinja, and a few others

I am not at all opposed to LED specific fixtures for new installations, but disagree about "archaic" and implied assumptions there is a problem with incandescent fixtures and disagree that it is some kind of problem that they are making screw-in retrofits. If you buy a home with junk fixtures or chose to buy them later what can I say? Same applies to anything including an LED specific fixture.

The economics of screw-in retrofits are not so different as you suggest. The basic difference is easily seen, that custom products are usually not only $160,most often multiple times that. If you are instead suggesting someone DIY for cheaper, then you'd also have to recall that DIY takes time, time=money to some, and that DIY can include retrofit lights too.

I am not against LED fixtures, yet there is a quite valid use for retrofits... they just aren't priced where they should be yet.

Manufacturers should absolutely NOT reduce even a tiny bit their focus on LED retrofits. If they want to expand their product offerings to include LED specific fixtures too, great, but it is way beyond naive to think the majority of the market should be ignored - that majority having incan fixtures. You propose something that would only cause bankruptcy when focused on the homeowner market rather than contractors, businesses, and special applications.

 

[jtr1962]

Re: Response to LEDNinja, and a few others

I am not at all opposed to LED specific fixtures for new installations, but disagree about "archaic" and implied assumptions there is a problem with incandescent fixtures and disagree that it is some kind of problem that they are making screw-in retrofits. If you buy a home with junk fixtures or chose to buy them later what can I say? Same applies to anything including an LED specific fixture.

I'm a little closer to the problem here as I actually have hands-on experience with these things. And I'm apparently not the only one who sees things this way. The screw-in fixture IS archaic with regards to LED technology. Nearly all of the problems which people are having with LED retrofits (i.e. short life, low output, overheating) ARE due to the attempts to mimic the form factor of an incandescent lamp, and not due to any limitation of LED technology itself. Not to mention that duplicating a ballast for every lamp, and then tossing it in the trash when the lamp dies, is beyond wasteful. The general public will see all these problems, simply say LEDs stink, and continue to use what they've been using for the last 100 years. So the real question is do we want to get people to adopt LED technology or not? I think the real answer is the policy makers mostly do, the power companies mostly do (on account of the grid already being taxed with no cost-effective way to increase capacity), but those involved selling light bulbs really don't. Some of the products being sold are so obviously junk which you can tell won't live up to the claims on the package. The manufacturers know this but don't really care. After all, they are already selling billions of incandescents. If anything, they probably have a vested interest in seeing the LED and CFL products mostly fail to gain market share, as it would impact their incandescent sales.

And as for houses coming with junk fixtures, most do when you move in. They basically have contractor specials which many people replace with something more to their liking before moving in. Everyone I know has done this. So if the LED fixtures were out there, they might well represent a viable choice.

The economics of screw-in retrofits are not so different as you suggest. The basic difference is easily seen, that custom products are usually not only $160,most often multiple times that. If you are instead suggesting someone DIY for cheaper, then you'd also have to recall that DIY takes time, time=money to some, and that DIY can include retrofit lights too.

No, I'm mainly suggesting buying LED fixtures, not making them yourself.

The economics of a purpose-built LED fixture aren't a whole lot different than existing linear tube fixtures, and those can often be had for under $100. Let's forget the stuff Cree sells. It's great but it's also overkill for residential use. An LED fixture needs a ballast. That costs no more than a flourescent ballast. The rest of the fixture can be stamped steel just like a linear tube fixture. Space the emtters far enough apart and the heat dissipation is just fine. Again, no cost difference from fluorescent fixtures. The primary difference is including perhaps a dozen emitters prewired to the ballast. Maybe that adds $30 or $40. So figure an LED fixture might run $30 or $40 over what a similar linear tube fixture will run. Big deal. Not many multiples of $160 as you suggest. Think about what getting the same amount of light from an existing screw-base fixture with LED retrofits will cost. Same number of prewired emitters, possibly even more due to higher operating temperatures, but spread over a couple of LED bulbs. Roughly the same cost there. But now you have several ballasts (one per bulb), and you also need to make a relatively pricey aluminum heatsink for each. Those 3 or 4 bulbs will end up costing more than your purpose-built LED fixture. They also won't last as long. And the purpose of this is what? So you don't have to discard an existing fixture to switch to LEDs, even though in the end you'll spend more money? Not to mention you'll probably run through 3 or 4 sets of those LED bulbs by the time a purpose-built LED fixture would die. So that's even more money spent.

Also, if some people are so in love with their existing fixtures, I tend to think there might in time being a good business in converting these fixtures to LEDs. Based on what some here have done, that kind of seems to solve both problems simultaneously.

I am not against LED fixtures, yet there is a quite valid use for retrofits... they just aren't priced where they should be yet.

Yes, there is a valid use for retrofits. Based on the output of most they might be great for an area with a hard-to-replace light which is seldomly used (and doesn't need to be brightly lit). Maybe in the attic or a closet. Even if you buy a cheap retrofit and it dims considerably, the light may well still be adequate.

As for being priced where they should be, if you're expecting LED retrofits to come down to $2 and also be good quality, forget it. I couldn't even mass produce a decent LED ballast for $2, let alone an entire lamp. While it's true emitters will get somewhat less expensive, don't expect radical changes from today's pricing. Cost per lumen will only come down because emitters get more efficient. You'll be able to get more lumens from any given emitter. The cost per emitter probably won't drop much lower than nowadays. And it'll be a long, long time before emitters become efficient enough that you can obtain 1500 lumens from a light bulb form factor with passive cooling. So I do see decent $10 LED bulbs which can match 100 watt incandescents perhaps 10 years or more in the future. But forget anything much lower. And the same economies of scale will carry over to LED fixtures when that happens, so they'll still be the more cost effective solution.

Manufacturers should absolutely NOT reduce even a tiny bit their focus on LED retrofits. If they want to expand their product offerings to include LED specific fixtures too, great, but it is way beyond naive to think the majority of the market should be ignored - that majority having incan fixtures. You propose something that would only cause bankruptcy when focused on the homeowner market rather than contractors, businesses, and special applications.

Again, you missed the point. You say LED manufacturers should continue to focus on the residential market mainly with LED retrofits solely because the fixtures already exist. However, it's a FACT that this focus is what will doom them to failure. We've already seen it with CFLs! Do we really need to repeat history here? Despite years of being pushed via rebates, give-aways by power companies, speeches by politicians, etc. CFLs have failed to gain much more than 10% of the market. Why? Well, at first many consumers felt they were too expensive and/or didn't fit in many lamps. So they tried to make them smaller and less expensive. When they did, along came the problems of short life and inconsistent quality/color. That turned lots of people off to them who may have formerly been open to the idea. Nothing wrong with fluorescent technology here, only with the particular implementation of it which was foisted upon the general public. Nobody or almost nobody made decent purpose-built fluorescent fixtures suitable for residences because apparently they felt changing out a lamp was the best they could expect from most of the general public. Had these fixtures been made, and the motto been "change a fixture, save the world", instead of "change a light bulb, save the world", then we might well not even be having this conversation. Homes, especially new ones, would be filled with mostly fluorescent fixtures. And the home owners would eventually replace those with LED fixtures once those fixtures reached a certain point in development (say 5-10 years from now when LEDs reach 200 lm/W). The idea of a screw-base bulb might well be seen in the same way we see TV tubes.

I'd love to be proven wrong, but it looks like history is repeating itself here. I see the same comments I saw when CFLs first came out. Too dim, too expensive, let's make it cheaper. It's already been proven that you can't do that with fluorescent retrofits, so what makes you think we can magically do it with LEDs? Whoever tries to make LED lamps catering to the present trends of CFL pricing will end up going bankrupt instead. I know I couldn't make LED ballasts cheap enough to sell lamps at $2. And that's even if the rest of the lamp (heat sink, screw base, emitter) cost nothing. The electronics business is marginal for the prices people are used to paying. While not everyone will want or be able to make their own fixture, more people than you think can either change out a fixture, or know someone who can. It's a shame so many people think otherwise.

 

[bshanahan14rulz]

Re: Response to LEDNinja, and a few others

I say we make friends with the people on an architecture/design/artist forum. Then teach them about LEDs. They realize what an amazing new light source it is, and design us specialized "artsy" fixtures. We collect the underpants and profit!

 

[jtr1962]

Re: Response to LEDNinja, and a few others

I say we make friends with the people on an architecture/design/artist forum. Then teach them about LEDs. They realize what an amazing new light source it is, and design us specialized "artsy" fixtures. We collect the underpants and profit!

That's exactly what I'm getting at. The design possibilities with LEDs are virtually limitless compared to any other light source (although lately I'm finding glow wires nearly as fascinating, at least from an artistic point of view). I think lighting designers could have a field day once they removed the constraint of having sockets. We get the artists/architects to design the lamps, someone like me takes care of the electronics, and someone else wires it all together. Do a few of these to show people, and I think they would be all over them.

 

[rodfran]

Re: Response to LEDNinja, and a few others

First of all-very interesting thread. Thanks for the input.

I have been testing two edison base bulbs side by side and here are some of my impressions.

Microbrite 5 watt cfl-200 lumens, cool white,25,000 hrs.-$11.40

Superbriteleds-E27-G50-W6-250 lumens, cool white,50,000 hrs.-$36.95

The metal base on the led gets very, very hot. Too hot to touch. The Microbrite cfl (called cold cathode tube) runs cool enough so you can easily touch it without getting burned.
I agree with jtr about the 5 watt led limit on the edison base without active cooling.

I have run the led for 20 months and run the microbrite for about 8 months, both averaging 2-3 hrs. per day use.

Even though the led is rated at 250 lumens(vs 200 for the cfl) the cfl is easily 1.5 times brighter(estimated by my puzzle solving wife).

Now, these are just my impressions of both. I have no electronics background. I just wanted something I could screw into the lamp and get the best bang for the buck.

My wife uses the lamp for reading and doing jig-saw puzzles in the evening.

I agree that the led in the edison base is not a good combo. The cost of the led is 3 times the cost of the cfl and as hot as the led runs, I doubt it will last 3 times as long.
I am most interested in longevity vs. cost.
It looks to me like the high output leds will need to be spaced in custom fixtures for adequate cooling and longevity.

 

[usLEDsupply]

Re: Response to LEDNinja, and a few others

i agree and disagree

first off there should be some changes to the star design to allow easy replacement of the chips, (there are some options out there for about $10 by Tyco that do a nice job of this but you are limited to the size and shape of their design) tho most of what we sell now doesn't use the star chips
we are still designing a fixture to use them that should be out sometime this winter

second: manufacturers should and are designing fixtures with led lights and you are seeing them more and more (the down side to this is they generally have no way to replace the emitter with out replacing the whole fixture, so lots of waste)

third: i do quite a few jobs that use only LED lighting for inside and out and yes some of the products are expensive but they will come down in price
and you just have to determine if you are going to be using that outside flood light enough to justify spending $200 for an LED fixture or just get a $20 500w halogen and replace the bulbs every year or so.
it is amazing how many people are willing to spend a little more $ for the long term savings

 

[J_C]

Re: Response to LEDNinja, and a few others

... but there isn't always long term savings, let's look at a real-world example over 10 years of use.

Consider a 60W incan bulb ran 10 continuous hours a day that needs replaced every two years and costs $1 in small volumes, so $5 total bulb cost over 10 years.

Consider two LED bulbs (or a fixture) equivalent in light output could be about $120.

Considier an electric rate of 0.07/KWH. That is a little lower than US average, but it is my rate so I'm using it.

0.6KWH/day * 365 days/year * 10 years = 2190KWH
2190KWH * $0.07/KWH = $153

$120 LED cost - $5 incan cost = $115... @ 5% interest over 10 years is $187.

Over 10 years the LED bulbs cost ($187-153 =) $34 more than the incan bulb and this wasn't even counting any power the LED bulbs used. For a business you have to add maintenance costs but for a homeowner you don't.

 

[J_C]

Re: Response to LEDNinja, and a few others

...$200 for an LED fixture or just get a $20 500w halogen...

Which $200 LED fixtures have equal light to a 500W halogen? I'm not suggesting there aren't any, just curious.

 

[DM51]

Re: Response to LEDNinja, and a few others

blasterman, I think this thread needs a new title, preferably a less personalised one that better reflects what it is actually about. A better title will also assist in more accurately focusing discussion on a topic.

When considering what the title should be, please imagine someone chancing on this sub-forum for the first time, and try to give a title that will inform them what the thread is about.

 

[usLEDsupply]

Re: Response to LEDNinja, and a few others

i bought a dozen or so from some company in china for a job where i was installing outdoor area lighting, they were 36x 1w led's with a 15deg lens (witch i removed most of) to make it a wide angle and while up close a 500w was brighter when you mounted it in the top of the eves it looked quite similar i think there are some pictures in the slideshow at usledsupply.com the one with the waterfall toward the end is one of the spots with the lens still on but the light is 125' from the waterfall and it is still putting out some impressive light i think i have one of it on the eves somewhere but i will have to look for it