LED lamps vs fluorescent lamps?

Flyback said:

so they must be failing short.....that's interesting, LEDs don't usually fail short unless they suffer significant overcurrents.......if leds die from general wear and tear, then they usualy fail open (almost always)........the fact that these leds are failing short points to serious mis-design in the led driver.

Click to expand...

I find it interesting that you would assume that, since a product is not working the way you think it should, that it is poorly designed.

Isn't it more likely, since the 'design flaw' you see actually means that it is working better for it's intended purpose, that it is actually well designed?

no, a led driver should never under any circumstance deliver large overcurrents into a led...ie overcurrents that are so large that they cause a led to fail short.
Short circuit led failures are from massive abuse of the led...something in the led has actually melted so badly that the led has become a permanent short...that means a seriously badly designed driver.....unless of course the lights are mega old and something else failed such that a massive overcurrent into the led occurred, but that would be after years and years of use.

Do you think its normal for a led driver to sometimes deliver massive overcurrents into the leds?...overcurrents that are so massive that they melt up the led so much that it becomes a short circuit?

I take it that you mean that because the product is still working even though a few leds have failed then it must be well designed?

Maybe/maybe not... I think we'd have to see the schem & bom and make our opinions.

Of course , its only speculation, massive amounts of outdoor electronics like traffic lights fail due to water ingress into the enclosure.......................I remember working as an assembler for a architectural led lighting company once.....we used to put led pcbs into tubes and seal the ends up with silicone........only a small percentage of the finished products were water-tight tested.......sometimes, an assembler might not have wanted to get sticky silicone on their fingers and mightn't have bothered to apply the silicone......sometimes someone disturbed them to tell a joke, and they then forgot about the silicone......non water proof products then would get shipped.

But you are assuming that the reason for the behavior is that the LEDs are being run in an over current condition. I doubt that that is the case. I feel it's more likely that they have added some sort of fail-safe circuitry so that a failed LED doesn't take out a whole string of LEDs.

The point to my previous post is that you automatically assumed the worst possible reason for the behavior. Instead of wondering what they might have done to improve the reliability of their product.

right I see your point, they may have added those breakover things which become conductive if a led goes open and thus it conducts round the led.

https://www.electro-tech-online.com/custompdfs/2013/09/NUD4700-DPDF.pdf
looks good at just $0.2

The LED shunts will keep the remainder of a series string lit when an LED or two become open.

I have noticed that LED traffic lights 5 years old have a few LEDs in random positions that do not light. The traffic light has about 30 LEDs so if a few do not light then the average amount of light is not reduced much.

Nige, did you know CPC do them, 100 notes, you can get pure white and warm white.

I got some from City electrcal for £140 then got one of the satday morning flyers from CPC with them in, the rest came from CPC.

**broken link removed** From This Range&MER=e-bb45-00001003

No doubt most of you will be aware of this, but ESP have a great set of articles on the subject here:
**broken link removed**
According to this data (and it's not all from the same study, so it may not a fair comparison) the most efficient light source listed is low-pressure sodium, and I'm sure I read previously (although I can't see it now) that a standard tube is more efficient than LEDs.

Some things that come to my mind are:

1. We mustn't confuse intensity with power output - comparisons need to be done on a power out vs power in basis that is not affected by directionality.

2. We need to take into account embodied energy - that used in manufacture, transport and disposal. This, divided by the lifetime of the product, deducts from its efficiency.

3. We also need an accounting system for the material resources used and how effectively they can be recycled. LEDs have the great advantage of containing no toxic mercury, but don't lend themselves to recycling in the way that glass/metal tubes do.

4. We need to manufacture fixtures so that only the failed part must be replaced - a major failing of many LED and CFL lamps is that the ballast must be discarded when the lamp fails (and vice verca).

I must admit that while I think LEDs may well be the long term solution (and are almost certainly superior to CFLs) I do worry that we can easily find ourselves on a path where a new technology becomes very trendy and is seen as the solution to every problem.
I'm sure that there is a single answer as to which kind of lamp is best in terms of overally efficiency of energy and materials, but this will only be found by dispassionate scientific research, not by pre-concieved notions, fashion or commercial considderations.

Just my 2p worth anyway...

Flyback said:

....................
Regarding the comparison between led and CFL......having heard you guys out, maybe the little curly CFLs are less efficient than the leds(?).......but id be surprised if a led bulb hanging in the middle of the room illuminated the edges of the room as well as a CFL (curly bulb).

regarding office lighting and industrial lighting, the prevalence of fluorescent tubes shows you what is more efficient and cheaper......they wouldn't use them if leds were cheaper or more efficient................think Tesco's supermarket in UK.....Tesco scrimps and scavenges after every single penny......and Tesco use flu tubes......so you can bet flu tubes are overall cheaper than led.
I hear 100lm/w for flu and on average ~65lm/w for leds.

Click to expand...

Of course one LED is not going to give the diffused light that a CFL does. But most LED bulbs designed to replace incandescents or CFLs use multiple LEDs pointing in different directions to give more omni-directional lighting.

Yes, most office and industrial lighting is presently fluorescent but most of those were built in the past, well before LEDs were even available and, the cost to convert the fluorescents to LEDs is high, so that will not be done anytime soon. But I think that commercial construction in the near future will start to use LEDs since the higher cost of the fixtures will soon be balanced by the long life of the LEDs which won't require the expensive (for the labor) periodic replacement that fluorescents do.

But most LED bulbs designed to replace incandescents or CFLs use multiple LEDs pointing in different directions to give more omni-directional lighting.

Click to expand...

How often do I find myself trying to invent a simple, cheap, LED PCB structure that enables a ceiling hung led light to shine uniformly over all the room below it.....there are ways to do it, but they are not cheap........I have seen bulbs that do a compromise of having multiple side facing led pcbs which end up illuminating the walls high up, and then they have one downward facing pcb below, but that one is too much like a downlight.

So really, the directional property of leds really is an unfortunate problem, which is only properly soluble by expensive constructions......sorry...the omnidirectional flu's win the day....they don't need diffusors to make them shine over the whole room.

dr pepper said:

Nige, did you know CPC do them, 100 notes, you can get pure white and warm white.

Click to expand...

We didn't pay that much!

never the less , the price of leds is totally ridiculous.
Also, LEDs are'nt getting any more efficienct or efficacious , what ever the word is. Of the power going into a led , still only about 16% comes out as light.
Its always been like that.
The only difference today is that scientists are able to produce bigger led chips, which therefore can handle more power than before.
But the amount of light per Watt today for a led is still the same as it was in 1960.

Theres been no progress, other than the "progress" of investment money onto the bank accounts of those who keep telling us that if they are payed more money then they'll make the "breakthrough" into wonderfully efficient leds.
Theres been no progress from 1960 to today. none at all.....its one of the only areas of electronics that has seen no progress (except for the led chips being able to be made bigger)

Flyback said:

never the less , the price of leds is totally ridiculous.
Also, LEDs are'nt getting any more efficienct or efficacious , what ever the word is.

Click to expand...

So? - they are still greatly more efficient than florescent lighting, and far cheaper in the long term.

However, I would disagree with your claim that there's been no progress since 1960, modern LED's are far more efficient than early ones.

So? - they are still greatly more efficient than florescent lighting, and far cheaper in the long term.

Click to expand...

Sorry but theyre not mre efficient than flu.
And another poster here has told that induction lamps are more efficienct than led for >40W

No progress since 1960......in other electronics areas theres been progress in silicon products....eg better fets, diodes etc......but no progress in LEDs........still no more efficient than they were in the 1960s.

Flyback said:

Sorry but theyre not mre efficient than flu.

Click to expand...

Well the rest of the world would disagree with you, but I suppose as in everything else you post here you know more than everyone else

But I suppose you don't consider better lighting, smaller bills and vaster lower maintenance costs as 'more efficient'?.

Interesting link
**broken link removed**

Also checkout OLED Organic LED's

E

Eric that link is from a company who received their figures from one of the big led manufacturers.....read Wikipedia on LED for a real truth on where leds are really at.........the best one is when they say they've got 200lm/w....then they don't tell you that that was in a freezer where the led die, even when flowing 1 amp, was kept down at 25C....even then they don't know how to consistently manufacturer this "2oolm/w" led, and its too expensive to do it for common manufacture...read about "auger recombination"...a fact of nature which stops leds being very efficient.

Just look what it (wiki) says about cree MK-R leds....then search it on digikey and see that its only at ~50lm/w.

Flyback said:

Sorry but theyre not mre efficient than flu.
And another poster here has told that induction lamps are more efficienct than led for >40W

No progress since 1960......in other electronics areas theres been progress in silicon products....eg better fets, diodes etc......but no progress in LEDs........still no more efficient than they were in the 1960s.

Click to expand...

Hi there Fly,

No progress you say? I dont follow every color but i do follow white LED's, and i did follow them quite closely over the past several years. I've seen progress in the Luxeon Star white's and also the small Nichia type whites which went from something like 500mcd to 5000mcd or something like that.

Lighting efficiency is not a simple matter of comparing the number for one type to the number for another type either. It's actually a three dimensional problem. You'll notice that many LED's have a directed light pattern whereas most fluorescents have more or less a 360 degree pattern. So some of the fluor's require a special fixture to direct the light to the subject. Without a special fixture (just screwing one into a lamp) we can loose almost half the light output. And with a special fixture we might see strange shadows. LED's direct more of the light to the subject and so we might see a brighter light on a surface even comparing LED's that put out the SAME total light output as a fluor lamp. That's not even considering electrical power input. If the power input was the same though, then the LED would appear to put more light on the subject. I have found this to be very true firsthand because i had fluor lamps and went to LED's.

Also, LED's are more versatile. You can get them in low voltage 3.5v and run them off of a transformer/rectifier whereas you cant do that with fluor's.

Another point is that LED's are easier to control. It's hard to control the light output of a fluor lamp, but LED's are fairly easy to dim given the right driver design. And once dimmed the effciency is still good.

Do you remember the "Bright Stick" fluor lamp? I used to use them and they kept burning out and getting hard to turn on. Also, with a lower voltage line (like during the summer here) they dont turn on at all, but keep blinking when they try to turn on. I am happy to be rid of them

USA dept of energy (DOE) testing of commercial LED lamps designed to replace incandescent lamps or CFLs showed that average efficacy was still about 46 lm/W in 2009 (tested performance ranged from 17 lm/W to 79 lm/W).[

Click to expand...

..this is a quote from "led" in wiki.

46lm/W is very poor ..Have things really improved that much since 2009?

Flyback said:

never the less , the price of leds is totally ridiculous.
Also, LEDs are'nt getting any more efficienct or efficacious , what ever the word is. Of the power going into a led , still only about 16% comes out as light.
Its always been like that.
The only difference today is that scientists are able to produce bigger led chips, which therefore can handle more power than before.
But the amount of light per Watt today for a led is still the same as it was in 1960.

Theres been no progress, other than the "progress" of investment money onto the bank accounts of those who keep telling us that if they are payed more money then they'll make the "breakthrough" into wonderfully efficient leds.
Theres been no progress from 1960 to today. none at all.....its one of the only areas of electronics that has seen no progress (except for the led chips being able to be made bigger)

Click to expand...

.

Now you are talking rubbish Flyback.

I finally realize now you have not a clue about LED's. You are a good SMPS person maybe?

Seriously. Don't say or type another thing here. The little reputation you have gained here will be at risk.

Tomorrow is another day. Save it for now. And sleep it on rather.

Regards,
tvtech

What I meant was that *Power* LEDs have barely improved in efficacy since 1960...I admit that *power* leds weren't really even availalable in 1960-1999.

I worked for an LED company, and the engineer told an invester that LEDs are getting 50% more efficient every year. This is not true. People (not those reading here) are quoting the improvement in *indicator* leds from 1960-1999, and that is irrelevant for leds for general lighting.

https://en.wikipedia.org/wiki/LED
..under the "Efficiency and...." section you can see that since power leds came of age in 1999, they have shown very little improvement in efficacy since them...Auger recombination has held them back.
In fact , over the last 6 years, LEDs have barely improved at all in efficacy.

From 1960 to 1999 the improvement in small indicator leds was very good, as you see in the wiki graph near section "The blue and white led". (Haitz law)

If you want to see how poorly efficienct leds still are, just take a look at the size of the heatsinks on cars with led headlights.....you wont find a picture of one on www as the manufacturers don't want people knowing they have to use heatsinks that are that enormous and complex in terms of multiple protrusions.
At another place, When we used to show our sales guys the internals of a led lamp (~3 to 8W) , they all bemoaned the size of the metal heatsink it contained, this always being their first observation...."does it really need that big chunk of metal in there?"