[Closed] Are we reaching the limit of LED brightness?

I'd say it's still very much in its infancy. There are some big design issues to crack yet - the primary one being efficiency. Sure they are more efficient than filament lamps but LED's are still only 10% efficient which means 90% percent of the power you put in is turned into heat and not light.

The big change currently is the move towards larger die. About 5 years ago, the best LEDs such as Luxeon K2 all used a 1mm square die. The high performace stuff nowadays like Luminus SST90's use a much larger 9 square mm die. Interestingly the lumens per emitting area hasn't really increased that much but now the emmiting area is much bigger. As I said before though, the big issue is heat dissipation. The hotter the LED gets, the less light it emits.

This is know as 'droop' in the industry and it affects all LEDs but the red, red/orange and amber devices suffer most. Typically, if you drive an amber LED such that its junction temperature is 100 deg C, you will only see 25% of the light you would get if you could somehow manage to drive the LED to maintain a Tjc of 25 deg C.

I could go on and on...........

Nah, I doubt it.
Having gone through a fair few lights in the last 5/6 years I imagine clarity etc will continue to evolve. Particularly when we have got the likes of troutie challenging the big brands ...
Proper dark this week around Malvern and I had forgotten how astonishingly good my lights were allowing me to night ride much, much faster than I go during daylight hours (albeit with HiD) 1850lumens from Trailtech MR16.

I hope so

Let's have a little dark on our night rides?

stoatsbrother +1

I guess the two areas I'd like to see improvements in are price, and battery technology (the output to weight ratio)

TBH I don't think we need further advances in actual light power. At the Gorrick last weekend the power output on the majority of rider's bikes was astounding compared to a few years back. I just don't see that we need more light.

I just don't see that we need more light.

I remember someone saying something similar about suspension travel. 😉

True; but in those terms we're already riding the equivalent of 10" front and rear on sub 25lb bikes with the lights we have now.

If we can do 10" f&r and get the weight down further OK, but I don't think bikes are going to improve with 15" each end 😕

All my bikes have zero travel each end and my off road lights consist of a single helmet mounted lumicycle Nimh 2000 lampholder modified with a single 200 lumen LED. Personally, I find its more than adequate. Until I ride with someone who has a much more powerful light then it feels like I'm riding in shadow.

All my bikes have zero travel each end

Ahhh, pedant-tastic STW - marvelous. I would rephrase but you know what I meant. Or perhaps you didn't?

Isn't it just the normal cycle of development - exactly the same as suspension in fact - for years travel got bigger now the focus is on quality fo travel.
LED's have got brighter but now I'm guessing the focus will be on quality of light and function rather than just sticking more lumens in. Thats kind of where the guys like Luminous and Trout are ahead in my opinion as they have started with a product that is conceived to give good light and good run times and engineered it to also be very bright.

Lense design is also very import. It would be good to see more light available with different lenses option depending on how and were you were going to use your lights

Lense design is also very import. It would be good to see more light available with different lenses option depending on how and were you were going to use your lights

When the market is right Quantum dot micro-LEDs will take over - their efficiencies blow current led techs out of the water - 85%
BUT I suspect manufacturers won't start dripping the tech in for 5+ years.

Yes i did say 85% efficiency - The same quantum (Non-Radiative) Energy Transfer will also be applied to photovoltaic cells - they will be revolutionary.

"700lm from AA battery running for ten hours charged by a solar panel on my bald spot. I have seen the future" 😮

stoatsbrother +2

Quite frankly find it a bit boring when someone sticks daylight in front of me on a night ride. Much prefer the fun and feeling of a more limited vision on a night ride. But I know some people don't night ride for the sake of night riding, they do it because they can't get out in the day.

I just don't see that we need more light.

I don't care about more light, I'd love to have longer runtimes and have fewer batteries to lug around.

Aidy "I don't care about more light, I'd love to have longer runtimes and have fewer batteries to lug around. "

Big big +1 for that, the P7s took us to a point where "more than enough light" is easily accessible but smaller units and better lifespan/smaller batteries will always be worthwhile even when more light isn't neccesarily desirable.

fwiw: there is only one E in Lens.

HTH

I imagine clarity etc will continue to evolve.

what does that mean ? Light has brightness, colour and directivity. Clarity is more a consequence of conditions and the distribution of light. There's a limit to what can be achieved from what is effectively a point source, unless perhaps one were to use some kind of holographic projection...

I've read a lot of people saying colour matters, like using yellow tinted glasses to see better, but I'm hugely sceptical that this is anything other than delusion and suggestibility.

I think when Troutie comes up with something that is actually brighter than the sun, and melts little pug dogs, someone will have to have a word. I suspect that day won't be too far off now....

[i]"All my bikes have zero travel each end ... a single 200 lumen LED. Personally, I find its more than adequate."[/i]

There you are see. If you had suspension you'd be going faster and would need better lights. 😉

No disrespect to Troutie, but isn't he just buying in components and connecting them up with some fancy circuitry and nice CNC machining.
I would expect any future big improvements to come from the LED manufacturers, not the light manufacturers.

Stuey, what do you mean "When the market is right..."
If 85% efficient Quantum dot micro-LEDs are available now, why are they not being sold ?

700lm from AA battery running for ten hours charged by a solar panel on my bald spot. I have seen the future

Solar panel technology - that has a loooong way to go yet as well!

Graham - I think he meant that the QD_LED technology isn't yet mature enough and will need to be properly productionised before it can be marketed. Given that QD-LED is being heavily developed for displays etc it will be another leap that would give us QD-LED to use as a light source for biking. From what I've seen anyway.

Edit - found this
[url]www.qdvision.com/quantum-light-platform[/url]

Seems like they might be getting close - but not there yet.

[i]LED's have got brighter but now I'm guessing the focus will be on quality of light and function rather than just sticking more lumens in. Thats kind of where the guys like Luminous and Trout are ahead in my opinion as they have started with a product that is conceived to give good light and good run times and engineered it to also be very bright.[/i]

Very perceptive of you, and I agree. I am of the opinion that at the moment, riding with between 1500 to 2000 lumens should be adequate for all types, and we achieve that kind of performance now with current, productionized, LED tech.

However, its not just a lumen count-fest, as most here realise.
Instead, we can now turn our attention to harnessing the light for desired results.

This is where testing comes in, testing combinations of LEDs and optics, along with logic controlled, multi-level output electronics and battery options, in order to create a desireable light system.

Some of my goals are to reduce weight, increase runtime, and maintain a useful level of light.

In fact, I was discussing this plan with Trout a few weeks ago, and told him I was going to introduce an "[i] Enduro [/i]" version of the 601.
This version will actually have the output wound back a bit, in favour of an extended runtime from a lighter battery.
The prototype is on my workbench now, about to undergo runtime testing.
I think Trout now offers an "Endurance" option too, seemingly deciding to use a larger battery.

Its good to see different approaches to a common goal.

So yes, perhaps we have peaked on the Lumen / Lux count.
As Trout has pointed out, when we double up the lights attached to our handle bars, ie, 2 x 601s, we don't necessarily get double the result.

So now the focus turns from how much light we can actually squeeze from a light design, to matters of light temperature, beam pattern, runtime, etc.

As we know, lights have lots going on, and anyone who is insane enough to try to hand-make and sell lights, has to try to keep quite a few plates spinning.

Not to mention offering a quality product.

I was very surpirzed to see the innards, or lack of them, during a strip-down of an Ay-Up light recently, by Trout, while again, he brought another light upto date / back to life.
Surprized also, to see the mode of failure.
.

I've recently had a Lupine Wilma 4 in for repair, and thought the quality could have been better, for a £400 light, especiallly when you consider that lights such as the 601 can match it on the quality of the housing, sealing, finish, and exceeded it on quality of other components.

And for all those who discuss the "too much light" scenario.
Most current lights can be dimmed, which only goes to satisfy the need by some people to ride with less light, and to extend runtime, again.

Which is all good stuff.

😉

I see a lot of research being done in both industry and academia. Mostly on the glues and adhesives rather than the light output etc. (that's mostly because I work with the chemicals though) but people still care and are still lokng for improvement.

Cost, quality of light and energy consumption could all be improved. I imagine the manufacturers are looking to break the household market with a cheaper product than currently available. I also imagine they are looking at very high powered lights for more industrial and heavy duty applicatios like vehicle headlights.

While I don't want 6000lm lights the trickle down to lower levels would offer advantages in run time, weight etc.

Yes, work continues on improving LED performance, I for one would like to see LEDs produce less heat for a given output.

As the LED companies will continue to press for more lumens per die, we might take those advances to a) reduce the number of LEDs in a bike light, b) to reduce the physical size of a light, as newer LEDs produce more lumens and less heat ?.

🙂

There was the same arms race with camera manufacturers as well, trying to squeeze more and more mega-pixels in before finally settling with making the rest of it better.

Same with lights, we know they're plenty bright enough, lets have lighter weight, longer run times, more adjustability/customisation and better weather-proofing.
NiteRider are on the right track with their user-programmable lights.

I'm sure it will be the automotive industry that pushes LED's to the next stage in development.

[i]I'm sure it will be the automotive industry that pushes LED's to the next stage in development.
[/i]

Thats what you'd think, but while there is an appreciable price difference between LEDs and conventional bullbs, the OEMs will make their decision based on cost and homologation requirements.
If all it takes to get your vehicle into a market, is the use of a (cheaper than an LED) bulb, then why go to the expense of using and LED ?.
Thats their thinking for the mainstream car.

Lighting efficiency, as well as other advantages that arise from using LEDs won't be at the top of the list of must haves for auto OEMs.

I think the home lighting market presents companies with the kind of business potential that attracts them to invest in LED R&D.

And us MTB'ers will just have to ride the wave of LED advancement
😉

What a thread!!!

The reality is that MTB lighting future is acombination of so many factors, and the stark diversity of customer requests (needs/wants) to the direction of lighting companies is ever changing

LED technology will have another big change in around 3-6 months when the new cree's will start to get produced. In the mean-time, battery technology is constantly being explored, and is a wonderful avenue to explore.

One thing I know we spent time on, and acheived, was that we don't emit any RF. So for all those that are using GPS units on there bikes, will still maintain acurate readings. This becomes important for adventure races.

Light output is one thing, but so many more factors are constantly being improved on (light spread, heat output, unit weight, RF, custom light outputs, burn time, charge time.....)and the list goes on. It exciting times in the light industry.

A lot said, can't be bothered to read it all.

Just have to say I've yet to see an LED that produces the same [i]quality[/i] of light as the first Halogen VistaLites I owned. If the technology advances to produce quality rather than quantity, then great.

AyUpRhino.

It seems that AyUps use only a resistor to limit current to the LEDs, is this correct ?.

Interesting point about the RF and EMC aspects. Technically I would have thought that anyone selling lights should have to have them CE marked and this would ensure compliance with the relevant EMC standard. A resistor is very inefficient compared to a buck/boost type driver but it is very 'quiet' as its a passive component.

I've lost count of the amount of hours spent at the EMC lab trying to get some of these switching drivers to pass.

Colour rendition is the thing for me.

I've not played with an eneormous amount of LED bike lights, but the DX P7 torch I had, despite being very bright had tuly appalling colour rendering, and at speed it was very easy to miss important detail (stuff like roots!), just because it blended in to the background. Riding the same trail with my old halo Lumis or an HID, that detail is seen much easier.

Now my understanding is that an LED outputs light at a specific wavelength (or at least a very tight band), and "white" LED is actually a very cold blue - hence objects with a large red or green content don't show up so well. Within the entertainment industry, we use colour mixing LED systems a lot, and many manufacturers are adding an amber LED into the mix of their RGB fixtures to get a good white, and also allow colour temperature to be matched to other light sources. Combining the existing (blue)white LED emitters with some carefully chosen output amber ones could bring dividends in terms of the quality of light output.

(FWIW my current Hope LED4 is better than the P7 in this respect, but only when on full output. The lower settings still seem to be quite "grey")

Agree they don't need to be any brighter though.

[i]A resistor is very inefficient compared to a buck/boost type driver but it is very 'quiet' as its a passive component.[/i]

Not sure about that, I would have thought that the two weren't really comparable.

🙂

EDIT:

A driver allows for multi level output and other features, its a control system.

A resistor, well, its just that, doesn't do anything else.

solar panels that work at night?, where do i send my money?

Rocketdog I got your mails but am in Tenerife and having great difficulty replying to mails
Will try again later

I'm sure it will be the automotive industry that pushes LED's to the next stage in development.

you'd be surprised who the automotive industry turns to for advice on led lighting.

and that is all i will say 😉

hi troutie, stop showing off! 😉

A resistor is very inefficient compared to a buck/boost type driver but it is very 'quiet' as its a passive component.

Is it still very inefficient when you're driving a 7.2v pair of LEDs with a 7.2v rated battery, so you need hardly any voltage drop, and the battery has a very flat discharge curve, so you don't need much of a safety margin on the resistor?

Joe

Last night I had a Wilma on my bars and Exposure on my bonce and still caught my bars on a tree a tree that whipped me sidewards into another tree like a ragdoll.

Now that nights are darker, its still warm and trails are dry its fanny-tatsastic. 😛

theotherjonv - Member

stoatsbrother +1

I guess the two areas I'd like to see improvements in are price, and battery technology (the output to weight ratio)

TBH I don't think we need further advances in actual light power. At the Gorrick last weekend the power output on the majority of rider's bikes was astounding compared to a few years back. I just don't see that we need more light.

I read this thread with interest last night - I am a lighting designer/engineer and I work for a company that sells the most advance LED technology on the market - this is far more advanced than stuff your seeing in bike lights or automotive uses etc.

It is interesting to read the opinions on here - obviously bike lights are a niche market, but in general the public opinion of LED technology is very skewed - how many of you have bought LED GU10 'replacements' from B&Q for your kitchen or Bathroom? And I bet your now rightly very sceptical of anything claiming to be 'good' LED technology! There is an awful lot of junk and misinformation out there.

As I dont actually do much night riding I don't have a lot of experience of LED bike lights but certainly some of the stuff out there is impressive. Its still far from the forefront of technology though.

Just have to say I've yet to see an LED that produces the same quality of light as the first Halogen VistaLites I owned. If the technology advances to produce quality rather than quantity, then great.

DEzB sounds like one of the people I described! DezB trust me there is LED technology that blasts Halogen out the field, but you wont see it in the cycling industry.

Colour rendition is the thing for me.

JonEdwards you have the right idea. A very important factor is what we call "Colour rendering index". You have also noted the very important difference between 'bright' light, and actuall good quality light.

The reason your Halogen/HID was better than your P7 was the colour rendering was higher - your P7 I am guessing has a colour rendering of around 70, which is pretty poor.

(Remember what its like walking down a street with Sodium (Orange) street lamps - where its quite 'bright' but all the colours look wrong, all the cars look black, etc etc? Thats because of the poor CRI of the Sodium street lamp. )

Something you wont have seen in the cycle industry is remote phosphor. (just like a flourescent lamp) This is one of the technologies that is improving LED lighting - it enables control of many factors such as colour temperature, colour point consistency, CRI, etc etc.

I'm pleased to see that there are some really nice LED stuff out there being used as bike lights - but there not quite at the forefront of technology!

Described where? In your bit about B&Q LEDs? I don't get you I'm afraid.

As in - you seem to be disillusioned with the LEDs you've tried!

My point is theres a lot of rubbish out there (whether that by cyle lights or things to fit in your kitched) that people have been sold on the basis of 'its as good as the halogen your replacing' when it clearly isnt!

There are LEDs out there that can genuinely match Halogen for light quality, but you wont have seen them on bicycles yet.

[url= http://www.ultraleds.co.uk/index.php?osCsid=91ac577234cf249249948abe66cafc7e ]Ultra LED[/url] have stuff that will not be in main stream production for a few years yet.

We use this [url= http://www.ultraleds.co.uk/mr11-acdc-blue-bulb-p-2111.html?osCsid=91ac577234cf249249948abe66cafc7e ]MR11 led lamp[/url] for interior lift lighting.

It is true 18w compatible and runs cold as ice.

[i]I'm pleased to see that there are some really nice LED stuff out there being used as bike lights - but there not quite at the forefront of technology!
[/i]

You might consider what is commercially available and applicable to the bespoke requirements of off road cycling.

Constraints are aplenty.
😉

As for cheaper lights. Not sure thats my field, quality will always come at a premium over absolute basics.
I prefer quality components that I'm confident will do the job very well, and having those components included into a good design solution.

The good thing is, there seems to be a light to fit most budgets now, which gets people out on the bike, when otherwise they might just stay home.
😉

@snotrag - fair enough. Not really though - for the size and weight they are fantastic, but 30W of Halogen for £80 was better light than £200 of LEDs.

Luminous your quite right - and as I said, as I dont actually do much night riding, then I'm not to experienced with the various systems available - certainly itsn ot bad thing with all this development going on and more people riding bikes can only be a good thing!

I really should get round to having a mess with one of our units and trying to fab a bike light - my problem is I know Zilch about battery technology, and everything I deal with runs off mains!

I work for a company that sells the most advance LED technology on the market - this is far more advanced than stuff your seeing in bike lights or automotive uses etc.

LED's are still only 10% efficient which means 90% percent of the power you put in is turned into heat and not light.

Yeah, LEDs are moving on, I'd still like to see more light and less heat, but wouldn't we all.

And when I'm saying "more light", thats only cos then we might run the LEDs at lower currents and stretch-out battery runtime
😉

Things probably will get much better in the future, but the nights are getting longer [b]now[/b], and as an earlier post pointed out, at this time of year, we're still benefiting from relatively mild nights and dry trails.

🙂

Given bike lights are using the most up to date LEDs produced by Cree/SSC how exactly is what you do more advanced? I'm fairly sure there aren't more advanced versions of these LEDs available to other companies which we haven't heard of. Or is there some other way by which you mean "more advanced"? I suspect from what you say that you're simply not aware of how bike lights are at the cutting edge regarding use of LEDs.

Apologies if I've come across in any way forceful, it wasnt meant that way! I am quite impressed with what seems to be out there in terms of bike lights, I havent really been following it much but this thread prompted me to do a bit more research and digging
Its hard to compare though as the stuff I deal with is for very different purposes/markets where there are very different aims aswell, obviously.

Part of the improvements that are coming with LED is not the actual LED units themselves (where CREE etc are obviously the leaders in this field) but in control and selection of them, and how they are manufactured into a light source...

I'm far from bad mouthing the bike light scene - Solid state lighting is something I've been learning about a lot recently through work and I've just been applying knowledge learned at work to my hobbie, and as you've said its really interesting that the bike industry is right up there.

All I'm saying is that there are other developments in other fields aswell that could become relevant to bike lights - 'Brightness' is already well covered, its Colour rendering, efficacy etc that matters now, and thats where these other developments such as remote phosphors will come in to play.

Given 100% efficiency producing white light is only ~250lm/W, and current LEDs manage well over 100lm/W, they're doing far better than that.

100% efficiency would be 680 lm/W if all the electrical energy were turned into light. If they can do better than that then they have a perpetual motion machine and free energy 🙂

100% efficiency would be 680 lm/W if all the electrical energy were turned into light.

Only for a 555nm (greeny blue) monochromatic source. Given the lumen factors in the sensitivity of the human eye, and that the eye is less sensitive to other frequencies, then a 100% efficiency ideal [b]white[/b] light source is what I said before (admittedly a white LED is far from a pure white source, with significant amounts of energy in the higher sensitivity part of the spectra, so the theoretical 100% efficiency for that is probably a little higher).

Part of the improvements that are coming with LED is not the actual LED units themselves (where CREE etc are obviously the leaders in this field) but in control and selection of them, and how they are manufactured into a light source...

I agree there are issues with colour rendering, but if you wanted to you could improve that with warmer coloured LEDs (which are readily available). I'm assuming your remote phosphors effectively do something similar but better by broadening and flattening the spectra? The trouble is that such things decrease the brightness, and whilst I'm sure for fixed applications you just throw more power at it, with bike lights you don't have that option - personally I'll take poor CRI and bright every time (it's not like I'm wanting to admire fine art using my bike light).

I don't disagree that there are developments which could be incorporated into bike lights - it's just your "far more advanced" I have an issue with, when bike lights still appear to be fairly near the cutting edge (personally I got some of the first ever batch of Cree XR-E LEDs, and had a working bike light using them within a couple of weeks of their commercial release).

aracer,

I guess the idea snotrag is on about is to produce a high brightness blue LED, and use remote phosphors to downconvert and tune the colour output.
A good way of choosing the exact colour rendition you want.
New "phospor like" materials (OLED's for instance) can have very high conversion efficiency and this may be higher than trying to do the converison within the LED itself. Only a guess as I don't know anyting about the manufacturing process of these LED's.

As for who will drive the development. I suspect it will be a variety of nich markets (like bike lights, "designer lighting", etc) along with the automotive industry.
I know the Merc, BMW, Audi, etc that make big heavy cars are looking into any way of reducing emissions in the fear that tough new euro legislation will hammer then for tax. LED's are more efficient, an so ultimately less fuel, and light units can be made lighter too. As an example I used to work for a company that was close to a big deal with merc to replace about 10 of their dahboard switches for a single "smart switch". They said the main driver efficiency (electrical) and weight.

I've got 2x 240lumen (well, a claimed 240) Q5 head torches.

Brightness is excessive, everything appears white in the center of the spot. You end up pointing the spot past what your looking at so the trails lit by the dimmer bit and can retain some contrast between mud/water/rocks/roots etc

aracer

I think you're the one who doesn't get the numbers. Lumens per watt is a measure of efficacy - the amount of light that an LED emits for a watt of input power. Efficiency relates to how much of the input power is turned into light, which for current LED's is about 10%.

Check out some datasheets.

So has the lumems per watt and general efficiency flattened out a bit in the last year or so? I know that Fenix still havn't got a 2xAA torch better than the one I bought 2 years ago (Q5 or something), I accept that batteries and optics are being optimised, but are we anywhere near 500 lumens from a pair of double AA cells with a few hours run time??

I think you're the one who doesn't get the numbers. Lumens per watt is a measure of efficacy - the amount of light that an LED emits for a watt of input power. Efficiency relates to how much of the input power is turned into light, which for current LED's is about 10%.

You should also try reading what I wrote: "Given 100% [b]efficiency[/b] producing white light is only ~250lm/W". I could have said "Given 100% [b]efficiency[/b] producing white light is only ~250lm/W [b]efficacy[/b]", but I didn't really think that would add much in the way of illumination (see what I did there?) You see you can convert between efficacy and efficiency quite easily by using the theoretical maximum efficacy as a divisor. ie if 100% of the input power is turned into light then you'll get ~250lm for every Watt of input power for a white light source if you need it clarifying further.

Where exactly does your 10% figure come from? As I said before, current white LEDs are better than 100lm/W efficacy, or to do the conversion, better than 40% efficient. Maybe you should check out the datasheets!

I'll dig a bit more if I may?

Where does your theoretical maximum figure of 250lm/W come from?

Well it's mentioned [url= http://en.wikipedia.org/wiki/Luminous_efficacy ]here[/url] (see "ideal white source") - any number is only ever an approximation, as it depends what you call "white light". Some more numbers [url= http://members.misty.com/don/lfunfac2.html ]here[/url] - I note the figure there of 331lm/W for white LED light, which is about what I'd expect for the reasons I gave before, though whilst I was GTFY I also found mention of Cree being through the 200lm/W barrier in the lab, so over 60% efficiency at turning electrical power into light. Also http://www.candlepowerforums.com/vb/showthread.php?t=250994 which has some discussion about this issue.

are we anywhere near 500 lumens from a pair of double AA cells with a few hours run time??

Thanks for the info. Having, ahem, checked the datasheets of some now obsolete Luxeon K2 green, amber and red devices which I used previously against the white Luminus SST90 that I have just used to replace a Welch Allyn Solarc HID, I see things have improved dramatically, particularly for the white devices.

How illuminating.... 😳

2 questions... How does an LED actually work and why does it stop my Cateye computer from working when i turn my Hope Visions on?

aracer

Hmmm......I do see where you are going with your effciency argument but it doesn't feel quite right. Below is extracted from a Luminus thermal application note. Apologies for poor formatting!

Consideration of Optical Power
In the previous examples, for simplification purposes, it was
assumed that all electrical power was converted to heat. Obviously,
this is not the case and a portion of the input electrical
power is converted to light. The amount that is converted into
optical power depends on the color of the LED and the drive
conditions of the LED. Since different color LEDs have different
efficiencies and the efficiency of LEDs degrades as a function
of input power, each use case will be different. For the most
up to date information on the optical power emitted from
PhlatLight devices, please consult the latest product data
sheets.
Once the optical power emitted from the LED is known, it can
be factored into the calculation of junction temperature. The
thermal resistance equation (2) can be modified as follows:
(4)
Thus when optical power is taken into account, the demands
on the heat sink are reduced.

Variable Red Green Blue Unit
Current 8.1 8.1 8.1 A
Voltage 2.3 4.3 4.1 V
Optical Power 2.3 1.8 3.5 W
Power Dissipated 16.3 33.0 29.7 W
Tamb 30 30 30 ºC
Tjmax 80 120 120 ºC
R? j-hs 1.32 1.32 1.32 ºC/W
?Tj-hs 21.6 43.6 39.2 ºC

So, in terms of efficiency when comparing input electrical power to output optical power using Watts, the above efficiencies range from about 5 to 15%. Now I know these figures are for colours but I believe a white LED is based on a blue device with a phosphor doping so its probably fairly representative?

What do you think?

2 questions... How does an LED actually work and why does it stop my Cateye computer from working when i turn my Hope Visions on?

In response to the second question... I imagine that your Cateye computer is wireless? If so, your Hope lights are likely creating sufficient EMI (Electromagnetic Interfearence) to jam the comms to the computer. The most likely cause is the switch mode driver that is used to supply the LED with the correct current. The noise created is likely overloading the front end of the computer receiver.