LED's

[SteveP]

Joe and others,

I'm interested in whether or not you've experimented with LED's. The industy's movement toward these is going full steam.

Regarding the MR16 type, there are numerous on the market. Here's a few notes and comments.

The brightest I've seen approximate 20W incandescent. It's unlikely they will ever exceed 35W levels.

Lamp life claims are all over the map. Since it's impossible to do accellerated tests (i.e. incr. voltage) and the lives are predicted for several years, LED mfgs are basing their ratings on lumen depreciation. For LED's, the standard is being set at lamp life being equal to a prediction of when lumen value has depreciated to 70% of initial.

This depreciation is largely dependent on temperature and the tests are conducted at room temperature. There have been no studies on the high temperatures and variations in temperature present in landscape lighting fixtures.

Various mfgs. are addressing temp control in different ways. Most of the new MR16's have built-in heat sinks in the form of fins or baffles. The effectiveness of these in a closed environment is in question.

Joe, I'm thinking that if you are testing the new LED's, the only applicable tool will be a light meter. And, for practical purposes, I'd think that designers would want to set their own standards to somewhere between 80 and 90% lumen depreciation. I think that if a lighting installation dropped to 70% of initial light levels, then the lighting would be unnacceptably dim.

I have nightmares that thousands of LED installations will go in with marginal light levels to begin with, then in the course of one or two years, the light levels will be abysmal. At that point, are installers going to go in and replace all the LED's? Incandescents depreciate no more than 5% throughout their life. One can blow and the new one will still match the others in light level. With LED's the whole system gradually dims so a relacement will stand out like a sore thumb. Also, the bullets (high heat) compared to path lights (low heat) will dim at different rates.

If you agree that 70% of initial lumen levels is too much depreciation for landscape lighting design then what about 85% as an acceptable limit. By my calculations, then under ideal calculations the life of a LED rated at 50,000 hrs. would actually need to be replaced at about 15,000 hrs. For high temperature applications, the life will be closer 10,000 hrs.

[Paul]

I agree with you Steve. From what I have seen, it looks like it's going to be awhile before LED is ready for our applications. I do think that it might work well in under water fixtures, as the water will help in the cooling.

[SteveP]

Good point about underwater applications, but you have to be concerned about corrosion of the LED circuit boards unless they completely encapsulate them.

[sitelights]

Starting in 2002 I began to examine LED technology as it pertains to low voltage outdoor lighting. Posts to date include "MisLED" 09.28.02 in the Lamps...Forum; "EasyTron LED" 09.14.04, "New Hadco LED Fixtures" 02.10.05 and "Vista's LED..." 03.24.06 in the Product Review Forum. If read sequentially one would detect a slow trend toward improved performance.

In the "Hadco..." review I postulated 20K hours before significant output drop; I will stick with that number which I grabbed out of very thin air. My reading up to the date of that post hinted consensus on the useful life of an LED array in outdoor lighting applications. I would question the 15K hour useful life estimate in SteveP's post only insofar as I would question the threshhold of the layman's perception of reduced light levels. It seems to me that if all the LED arrays are suffering diminished output simultaneously the change would be noticeable at a somewhat longer interval. This is a judgement that is based on perception not quantification.

I fully understand the "rush to market" with these LED applications since for the most part low voltage landscape lighting fixture manufacturers have very little if any technology-driven product differentiation at their fingertips. The situation puts me in mind of a saying that one of my business mentors was fond of: "You can always recognize the pioneers by the arrows in their backs."

If the manufacturers relied on foresight rather than hindsight they would be developing systems that would operate "off the grid" rather than scrambling for position in an immature technology as it now impacts the outdoor lighting business.

I make this statement because I believe that the emerging solar and LED technology will soon combine to form a stand-alone application in the realm of projected...repeat...projected light. After all solar is DC is it not?

[Paul]

Agreed, but that shouldn't be hard to do.

[sitelights]

I am not as confident as you are that this is an easy thing to do in the immediate future. My comments on the emerging solar/LED technology are based on my involvement with solar lighting of over a mile of trail in a parkland setting utilizing individual solar arrays powering separate light pole mounted luminaires equiped with LEDs. Very substantial advances will have to occur (efficacy, storage systems and miniaturization, etc.) before this can become a reality in landscape lighting applications similar to that being used now..

[Paul]

Agreed, but that shouldn't be hard to do.

This was in regard to Steve's post on encapsulating the LED circuit.

[sitelights]

I missed the connection on the topic of LED heat and I now understand your comment.

Since one of the characteristics of LED equivalency is substantially lower heat generation vs incandescent/halogen sources it would be interesting to compare the ratio of heat to light. Incandescent sources are about 95% heat and 5% light in terms of energy utilization (nominal figures) whereas the Hadco LED array consumed 12w for a 20w equivalency in output (see Review) or a reduction of 40%. Thus in even higher output circumstances an equivalent LED compared to a 50w halogen would be about 30w.

Perhaps I am missing something in SteveP's scenario unless his information leads one to the conclusion that LEDs are much more sensitive to temperature levels. In incandescent lighting the heat driven deterioration occurs at the socket but LEDs are hardwired (a very real advantage) and so there is no "socket" in the usual sense.

I am puzzled that the difference in watts consumed does not have a similar effect on heat production i.e. the LED should generate 40% less heat than an incandescent source. In any case it is the body cavity size and the conductivity of the body material and its ability to shed heat to ambient that determines the temperature level of the interior components.

This leads me to suspect the the LED source currently has only a minor advantage in terms of energy use compared to the incandescent equivalent. It is energy used that determines heat production for equivalent outputs and the apparent ratio for LEDs at 60% heat and 40% light vs 95% and 5% shows how far the LED has to go to demonstrate any striking advantage in efficacy. I state here that this is essentially an empirical approach rather than one based on scientific methodology.

Of course since the incandescent technology (as halogens) has reached maturity one can only expect improvements in the efficacy of the LED while incandescent sources are left in their own dust.

Perhaps SteveP can shed some light (no pun) on this aspect of the discussion.

[Paul]

Joe, Steve is talking about the heat that is created in the LED circuit board. Remember that light emitting diodes are a bi product of the circuit board that it is attached to. Though these boards don't produce heat that is anywhere near what an MR has, it does decrease the life of the LED.

In talking to people at GE, this is their biggest issue besides color. They would like to find a way to keep that circuit board cool and therefore the LED would last as long as they tend to claim.

[Hadco]

If you have specific questions about leds, what a junction temp is as well as what steps there are to control lamp life and heat, ask away i will try to answer them all.

I can tell you this LEDs in the near future will produce wattage than you would exspect, and the LED bi-pins, are very low wattage and will remain that way due to the heat factor.

Also i do Work for Hadco Lighting & Bronzelite and will be more than willing to answer questions about our products.

cheers

Sturgis

[sitelights]

Hardwired LEDs are of interest to me; retrofit types that require a socket simply replace one problem with another: socket deterioration.

In an earlier post I expressed puzzlement that there seemed to be no linear relationship with watts consumed and heat generated. Since a 12w LED array is the approximate equivalent of a 20w MR16 (see my Hadco LED review in the Products Forum) it would seem to me that the heat factor should work in favor of the LED array since fewer watts (heat) are consumed.

Within an identical cavity the Hadco 5016 with a 20w MR16 lamp and its LED counterpart the buildup of heat would seem less critical since the LED array is rated as 12w and the MR16 20w. What is there about the LED array that results in more rather than less heat? Is the LED array and its components more prone to heat failure at a lower temperature?

The elaborate heat sinks that I associate with these 20w MR16-equivalent LED arrays are obviously necessary (and expensive); one can only posit that the energy saving aspect of LEDs vs halogens of an equivalent lumen output is chimerical ("imaginary") rather than substantive. The price point (no need to get specific) difference is brutal and the resulting improvement in efficacy is almost nil.

The modest performance gains (if any) for LEDs in terms of efficacy argues (for me at least) that LED technology in landscape lighting applications is not only immature but disappointing. You (or other readers) may be inclined to look at "LEDs" posts via the search function.

As is uniformly the case, the technological advances of lighting improvements depend entirely on markets other than low voltage landscape lighting applications. In other posts I suggest that the applications of low voltage lighting technology are driven entirely by the automotive and interior lighting display market. Nothing has changed in 60 years. Give my regards to Lew.

[SteveP]

I am also very curious about the actual heat energy transferred to the circuit board. Does anyone have hard numbers for the amount of heat generated in the boards of LED's? Do we need to be concerned about heat transfer out of fixture bodies in tropical climates?

Interesting note, I heard of a contractor in the midwest who has been experimenting with LED's in landscape lighting. He complained, "They don't melt snow! I can't use them!" This is a valid consideration, much of the northern regions are blanketed in snow and fixtures must melt through snow to be effective in the winter.

Of course, there is global warming that will solve this problem. . .

[Hadco]

Heat is a huge factor in LEDs! The cooler you can get the junction temperature the longer life and light output a led can have. An example of this might be an LED with the junction temp of 90f an LED might have 100,000 led life, however if it goes up 30 degree to 120F you know have 50,000 hours led life, if it goes up another 30 degree to 150F it will then go to 25,000 hours. Also as the temp goes up the light output is less and less, I don’t have an formula of how much less though if it quite a bit. If I have a chance today I will take a picture of a demo kit we have here that shows 3 different junctions temps and their light output.

With the Hadco inground LED we have a module that combats higher degree ambient temps and keeps our junction temperature around the 120F degree mark and that is why we don’t market a 100,000 lamp life. Also if you have our inground led module out of the housing and run it for several hours you can not pick it up, the heat sink gets that hot, however the top where the LEDs are is cool to the touch to the point that it barely feels like it has ever been on.

Onto the plug in LEDs, in my opinion they are a joke. Yes you can say you have an LED but they are so low in wattage they don’t really light anything up. That is the reason we have designed the modules, and to touch on how hard it is to maintain junction temps we had a custom order for an I2 inground that the customer wanted the 12.4 watt LED in it. We tried the module out of the IUL516 LED and it worked but it still had too high of a junction temp, so we actually had to have a new module made to lower the junction temp.

Hahah, you know if we ever got snow here in the northeast I will see if the LEDs melt snow, my guess is the inground will not but our BL5016 version of the LED will, I will have to update you if we ever get snow. Yes, I do agree the global warming will not only melt the snow but destroy the LED market so maybe we should quit while we are ahead, haha

I'm sure I missed some questions on here but feel free to email me @ skyle@genlytegroup.com and I will try email or call you back to answer your questions.

Thanks

Sturgis