Connect with us

Conversion efficiency of White LED backlights

Discussion in 'Electronic Design' started by PeteS, Aug 17, 2006.

Scroll to continue with content
  1. PeteS

    PeteS Guest

    Well, my LCD supplier has decided they can put a white LED backlight
    onto the panel size I use. Now I'm trying to get a reasonable thermal
    budget, and something I am trying to find out is the typical conversion
    efficiency ( electrical energy -> light ) of the backlight.

    Details: 3 x string of 8, Vf 27V typical at 25mA

    I currently don't have a panel here (it seems to be taking a slow boat
    from Taiwan) and I'd like to get a head start so I can do some rough
    placement planning and I wondered if anyone here had any knowledge of
    said conversion efficiency.

    Obviously, energy not converted to light gets converted to heat and I'd
    like to get a handle on what I need to deal with. This is a handheld
    XDA device, so thermal management is sorta important ;)

    Cheers

    PeteS
     
  2. Initially, while you are waiting for the slow boat, assume zero
    efficiency and calculate on that. If you don't find a problem, you don't
    need to know the efficiency. It's called 'engineering'.(;-)
     
  3. Iwo Mergler

    Iwo Mergler Guest

    Energy not *escaping* as light gets converted into heat. If the display
    shows a black picture, almost nothing escapes. So, for the worst case,
    practically all energy gets transformed into heat.

    Kind regards,

    Iwo
     
  4. PeteS

    PeteS Guest

    Snort

    I did that ;)

    Total power consumption for a 5.7 inch panel at full brightness comes
    out at just over 2W, which isn't *that* bad - provided I am careful
    with everything it will maintain the distributed nature of the heat and
    keep the temperatures low enough so I can charge My Li+ batteries.

    Cheers

    PeteS
     
  5. The black screen and the enclosure allow some heat to escape, by
    convection and radiation. The latter is not negligible at low
    temperatures.
     
  6. PeteS

    PeteS Guest

    I might have noted in the first place that I can live with 2W of
    dissipation and there is, of course, some heat escape via convection
    and radiation. I simply wondered if anyone knew the conversion
    efficiency so I could tweak the heat dissipation map for the device :)

    Cheers

    PeteS
     
  7. BobG

    BobG Guest

    =====================================
    I heard a rumor they were 'almost as efficient as compact
    flourescents', but since I don't know how efficient THOSE are, I guess
    I can't help.....
     
  8. The information is out there; I thought you wanted a way to progress
    without waiting for it to appear. Won't the manufacturer tell you?
     
  9. PeteS

    PeteS Guest

    The answer was 'we are still characterising the device' - which is of
    course a euphemism for 'we don't know'.

    I have already designed assuming zero efficiency, and I'll find out
    when I measure the damn thing in about a week or so.

    Cheers

    PeteS
     
  10. PeteS

    PeteS Guest

    The bright side of this (no direct pun intended) is I will actually
    report my findings once I test the units ;)

    Cheers

    PeteS
     
  11. Kevin White

    Kevin White Guest

    White LEDs are only about 3% efficent, 15-20 Lumens/watt - about the
    same as incandescent lamps. CFL produce about 60 Lumens/Watt. A lamp
    with 100% efficiency would produce 600-700 Lumens/Watt.

    For your application nearly all the electrical power would be converted
    to heat - a large proportion of even the small amount emitted as light
    would be converted to heat within the LCD display.

    See for
    example:http://en.wikipedia.org/wiki/Light_bulb#Luminous_efficacy_and_efficiency
    http://www.otherpower.com/otherpower_lighting.html.

    kevin
     
  12. colin

    colin Guest

    It is rather dissapointing how inneficient they are given the hype about
    their efficiency.

    Colin =^.^=
     
  13. GregS

    GregS Guest


    Kevin WHITE !!

    Thanks for the info.

    greg
     
  14. James Waldby

    James Waldby Guest

    ....

    Several DOE-funded research projects are aimed at 50-80 lumens/watt
    LED's. Eg, Color Kinetics Inc. aims at 80, Eastman Kodak aims at 50,
    Osram Sylvania aims at 80. (See page 4 of Summer 2006 LED Journal.
    Magazine's web page is http://www.ledjournal.com/ .) Also, CAO
    Group Inc. claims to be getting 50-70 lumens/watt in Dynasty 640
    white LED's on page 8 of same issue -- perhaps fueling the rumors...

    -jiw
     
  15. GregS

    GregS Guest

    Edmund Scientific made an anouncement some time ago about
    increasing the lamp market share with high intensity LED's using some kind
    of fiber optic conversion. I think mostly for projection type lamps.


    greg
     
  16. GregS

    GregS Guest

    Phillips 3 W K2, manufacurer says up to about 47 lumins from 1 watt.
    They are hard to get right now.

    greg
     
  17. last time I looked Luxeons were about 40 lumens/watt
     
  18. 100% efficiency of producing a yellow-green at a frequency of 540 THz
    (wavelength approx. 555 nm) is 683 lumens per watt by definition. Human
    photopic vision peaks there.

    100% efficiency of producing white light varies, depending on the
    definition of white light. But for "white light" to be the 400-700 nm
    portion of the spectrum produced by most white artificial white light
    sources, 100% would usually be somewhere around 250-300 lumens/watt.

    - Don Klipstein ()
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-