LEDs as lamp replacements

Eeyore wrote in news: snipped-for-privacy@hotmail.com:

Tell that to NIST >:)

Eeyore wrote in news: snipped-for-privacy@hotmail.com:

That should satisfy you then. You challenged me to cite proof of 176 mW as if I was claiming too much. Are you now saying I'm wrong because it's not EXACTLY 176 mW? If so I'm done discussing the matter.

I doubt you'll do much better actually. 82% sounds about right (not 95%) however that is at max load.

I see they don't state efficiency for the lower power models.

Graham

Well actually I wonder about that. Especially replacements for those 'low voltage halogen' types like these.

There's not actually any easy way to lose the heat. The 'enclosure' is very small.

I'm not so sure about the quickly bit.

Graham

Eeyore wrote in news: snipped-for-privacy@hotmail.com:

No fair! >:) You try fitting a CFL in there. That's why you're pulling teeth out of the biting argument in favour of LED's isn't it? Most of the time in this thread you're advocating CFL's, so this is a spurious issue.

Actually, the Cliften Suspension Bridge in Bristol has its chains entirely lit by exactly the type of LED-based lamp that replaces little tunsten halogens. There are three emitters per lamp. The lowest is too high to get a good look at, but they appear to be Cree or Luxeons in small parabolic reflectors. Sure, a bridge doesn't have many cooling problems most times, but those 'chains' are mighty plates of iron, and on summer nights they are hot. Doesn't stop those lamps working though.

Lostgallifreyan wrote in news:Xns9968D2D4113A1zoodlewurdle@140.99.99.130:

That is so embarrasing I have to correct it. 'Clifton'.

How do you work that out ?

Ten emitters of the current best type account for about 34W ! You'll have a hell of a job keeping them below 70C or so in the space of a current lighbulb ! Forced cooling would be essential and that's going to be annoying. How many fans last 50,000 hrs too ? I'm sure they'd get end up overheating through dust build-up reducing fan efficiency too.

Graham

Eeyore wrote in news: snipped-for-privacy@hotmail.com:

Ok, I think it will run hotter than I was suggesting, but you're forgetting something too, a much greater chunk of that imput wattage will be emitted as light, not heat, compared with tungsten lamps.

However you did mention volts previously. It wasn't me who did that.

LEDS operate at lowish voltages. You tend to get lower efficiencies with at lowish voltages, largely due to rectifier losses.

I doubt you'll much better 85% power conversion efficiency overall from the AC line to the LED package actually. That would be a *very* good figure.

Graham

No, it's fine. I hadn't previously noticed anything above about 150 lumens.

Graham

Eeyore wrote in news: snipped-for-privacy@hotmail.com:

[snip]

I see now, that's just loose phrasing by me that led you to misinterpret.

We both know that LED's are current driven. What I meant was that the input can range between 5 and 32 volts. The conversion is to one of a variety of preset fixed currents.

Eeyore wrote in news: snipped-for-privacy@hotmail.com:

LED drivers chain LED's to limit that problem. The one I bought can chain 7 LED's.

To be honest, I'm not actually advocating any specific technology. I am however very interested on Philips new compact 'CHLi' halogens.

I have seen LED 'replacements' for that GU10 type touted, but it's clear they must have very much lower light output.

Graham

Eeyore wrote in news: snipped-for-privacy@hotmail.com:

It changes fast. :) I lost interest in LED's around the time of the 14 candela amber Toshiba 10mm types, because laser diodes were more fun. Things have changed big time. I really do think they'll overtake CFL's so fast that most of the public, especially legislators, will have no time to adapt except in sudden jumps. They'll be thankful to do it too, once they face the clamour of public dissent at having to throw away good light fittings to suit the intended ban of tungsten lamps.

While something could stall the development of LED's there seems to be no sign of that, if anything it will likely accelerate as technology further blurs the distinction between LED and laser diode.

Only about 10%. There'll still be 34W to get rid of. And I forgot the dissipation from the electronics too ! That'll be another ~ 5W so we're back to where we were.

Graham

Even then, the seondary rectifiers alone will consume about 5% of the power.

Graham

Eeyore wrote in news: snipped-for-privacy@hotmail.com:

I'm keen on the developments in halogen lamps too. They have a light quality I consider second to none. I hope broadband phosphor mixes will be improved, but they still don't work like a small efficient halogen does.

I need to stop now, want to watch Minder. :) I need a rest.

145 typical lumens at junction temperature 25 degrees C for the highest brightness rank in the datasheet. Now, what about with a heatsink temperature of 35 degrees C and for the highest brightness rank in the Future Electronics website? 30 watts into these means probably about 25 watts of heat. Hmmm, what if you mount a dozen of those onto a heatsink the size of the tip and heating element combined of a 25 watt soldering iron? What would the heatsink temperature reach then? What does the datasheet say performance is at that temperature?

With a good size heatsink to stay at a comfortable temperature with 25 watts, I don't see it looking like a universal incandescent or CFL replacement just yet.

That I believe!

- Don Klipstein ( snipped-for-privacy@misty.com)

No, they use a different phosphor made for LEDs. Color rendering of white LEDs is more like that of "old tech" halophoaphate fluorescents than like that of CFLs. At least the color rendering index is somewhat better than that of "old tech cool white".

- Don Klipstein ( snipped-for-privacy@misty.com)