Lets compare LEDs against one of the best flourescent

[Handlobraesing]

They compare the best available LED against an older fluorescent system in many articles. I find it rather unfair as the fluorescent technology hasn't stopped growing yet.

Let's look at perhaps the best fluorescent out there.

Philips ALTO extra long life.
http://www.nam.lighting.philips.com/...pdf/p-5794.pdf


F32T8. 2950 lumens(after 100 hrs of use) 2800lumens (after 20,000 to 23,000 hrs) - 32watt

At 12 hours a day cycle (hallways, stores, and most commercial lighting)

Using the latest available ballast available, the programmed rapid start technology, the lamp lasts an average of 46,000 hours. 40,000 hours on the common instant start.

95% of initial lumens maintained after 20 to 23K hours, a figure yet to be seen for LEDs.

[jtr1962]

It's been pretty much an accepted fact that linear fluorescent tubes will be among the last things LED replaces since they're hard to beat in terms of efficiency, cost per lumen, and lifetime. The only downside is that for whatever reason linear tubes have never been widely accepted in residential lighting. I've always found this puzzling, especially now with the popularity of recessed lighting like high hats. Linear tube fixtures can be just as easily recessed while offering greater efficiency, longer life, and less heat. I guess one of the reasons is the entrenched "light bulb" mentality of the general public.

Most of the comparisons I've seen have been of LEDs versus incandescents, or LEDs versus CFLs. That makes the most sense as LEDs are closest to replacing both these light sources. In the both cases, LED wins hands down in all areas except cost. The cost problem should be solved as mass production ramps up. I still think it'll be a long time (at least a decade) before commercial establishments start replacing their linear tube installations with LED.

BTW, LED can easily exceed the rated life of those tubes if underdriven. They have the capability to run for a million hours or more if that is a design goal. You even have a substantial efficiency gain by underdriving them. The only downside is increased cost for more emitters.

[Handlobraesing]

Originally Posted by jtr1962

BTW, LED can easily exceed the rated life of those tubes if underdriven. They have the capability to run for a million hours or more if that is a design goal. You even have a substantial efficiency gain by underdriving them. The only downside is increased cost for more emitters.

Do you suppose the entire lamp would maintain 90%+ output after 45,000 hours of use? While it might last n million hours, lumen depreciation is a big problem with LEDs. These fluorescent tubes maintain around 95% at 20,000 hours and 90+% after 45,000 hours.

If the output falls much below 70%, it's just as good as bad for general illumination.

Whats amazing about these T8 is that lighting system can be designed without a regard for lumen depreciation since 5% loss does not affect much.

When fluorescent had a mean lumen of 70% of initial lumen, it was often necessary to over illuminate by as much as 42% of desired lumens when the lamps were new.

[2xTrinity]

Originally Posted by Handlobraesing

Do you suppose the entire lamp would maintain 90%+ output after 45,000 hours of use? While it might last n million hours, lumen depreciation is a big problem with LEDs. These fluorescent tubes maintain around 95% at 20,000 hours and 90+% after 45,000 hours.

If the output falls much below 70%, it's just as good as bad for general illumination.

With phosphor LEDs, you're right that lumen depreciation is bad -- mainly because the phosphors get extremely hot when in close contact to the junction. With direct RGB LEDs, that are all underdriven, lumen maintenance could be close to 100% however -- that is certainly the case for many indicator LED lamps that have been on for over 10 years in my house. However, the system would be extremely expensive.

Whats amazing about these T8 is that lighting system can be designed without a regard for lumen depreciation since 5% loss does not affect much.

When fluorescent had a mean lumen of 70% of initial lumen, it was often necessary to over illuminate by as much as 42% of desired lumens when the lamps were new.

What's more amazing is that 90% of the fluorescent fixtures and lamps sold in stores to consumers fall into the latter category -- line frequency, 60CRI T12s -- utter garbage in every repsect.

The other thing that could be done though would be to separate the LED from the phopsohr, or to use blue LEDs to backlight a phosphor panel. The panel itself would contain the phosphor (and give more even illumination than most fluorescent ceiling fixtures), which would convert the blue light to red and green. Such a fixture could certainly keep high efficiency, and if needed, the phosphor panel itself could be replaced separately from the emitters. (This option of separating the phosphor from the emitter is not possible with fluorescent, as having a clear shortwave UV lamp with no coating would be dangerous)

[jtr1962]

Originally Posted by Handlobraesing

Do you suppose the entire lamp would maintain 90%+ output after 45,000 hours of use? While it might last n million hours, lumen depreciation is a big problem with LEDs. These fluorescent tubes maintain around 95% at 20,000 hours and 90+% after 45,000 hours.

If you design the system to last 1 million hours, then you can easily have 95% lumen maintenance after even 200,000 hours. It's simply a matter of how many extra underdriven emitters you'll need compared to a full-current solution. I'll grant that the T8 solution is currently far less expensive, and will be for quite some time, at least for general lighting. However, there are specialized applications where the extra cost of LED is more than saved in reduced maintenance. Traffic lights are one example. LED street lights will be another. I think 1 million hour life, with 90% lumen maintenance, is a stated goal of LED street lamp designers. The idea is to eliminate relamping entirely by having the light source last as long as the fixture (~200 years).

BTW, 90% lumen maintenance after 45,000 hours is indeed impressive. It's a pity that linear tubes aren't designed into new homes. Assuming that the lights are on 8 hours per day, that's one relamping every 15 years.

[Handlobraesing]

Originally Posted by jtr1962

If you design the system to last 1 million hours, then you can easily have 95% lumen maintenance after even 200,000 hours. It's simply a matter of how many extra underdriven emitters you'll need compared to a full-current solution. I'll grant that the T8 solution is currently far less expensive, and will be for quite some time, at least for general lighting. However, there are specialized applications where the extra cost of LED is more than saved in reduced maintenance. Traffic lights are one example. LED street lights will be another. I think 1 million hour life, with 90% lumen maintenance, is a stated goal of LED street lamp designers. The idea is to eliminate relamping entirely by having the light source last as long as the fixture (~200 years).

BTW, 90% lumen maintenance after 45,000 hours is indeed impressive. It's a pity that linear tubes aren't designed into new homes. Assuming that the lights are on 8 hours per day, that's one relamping every 15 years.

Granted, just like everything, LEDs have their place. In single discrete color application, almost nothing else can touch them.

The focus of this thread was meant to be general illumination.