LED 'afterglow' time?

Discussion in 'General Electronics Chat' started by Hypatia's Protege, Dec 22, 2015.

  1. Hypatia's Protege

    Thread Starter Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    Kind friends:

    It occurs to me that stroboscopic tachometery represents a highly satisfactory solution to determination of rotor bearing condition (Re: Coolidge RA types) --- Moreover, owing to enhanced switching stability and favorable (i.e. compact) geometry, LED technology would seem a better choice than traditional ‘flash-tube’ systems…

    Hence my questions:

    1) Do ‘ultra bright’ white LEDs exhibit significant ‘after glow’? If yes, what is the antcipatable range in characteristics (i.e. luminance vs time following interruption) –- failing precise data - how do such LEDs compare to Xe 'photoflash tubes' in this regard?

    2) Are such LEDs 'tolerant' of rapid switching?

    Many advance thanks for any and all assistance!:)

    Best regards
    HP
     
  2. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    Frankly I have little idea about this, but I have a few things to consider. Does a forward biased led exhibit any capacitance? Also, do laser diodes have different parameters, since they are used at very high speeds quite often?

    Also, you definitely don´t want to use white LEDs since they use a luminofor and I guess will be very slow.
     
    Hypatia's Protege likes this.
  3. BReeves

    Member

    Nov 24, 2012
    412
    64
    Just had to look this up as I had never heard of this before. Had better luck searching for "Lumiramic" which is a technology developed by Phillips that places a Phosphor layer over the LED to control the color temperature of the light. Very interesting and can understand why you would suspect the afterglow time would be longer than an LED without this layer.
     
    Hypatia's Protege likes this.
  4. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    Oops sorry didn´t realize this is not used in english. But if you search for luminophore, you get much better results ;)
     
  5. Hypatia's Protege

    Thread Starter Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    That's just the sort of issue I was 'afraid of':( --- Clearly 'low intensity' LEDs have rather low 'persistence' (as evinced by 'antique' calculators, and their ilk) wherein the strobing of the display may be 'detected' via a photo transistor --- I might have guessed such performance from high output devices was too much to hope for:rolleyes:

    That's an interesting point!:) -- In my experience diode pumped YAGs tend to 'start' rather slowly - but then I expect there's a difference between application specific instrumentation and 'hacked' pointers:oops::D

    Many thanks four responses!!!:):):)
     
  6. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,320
    889
    Not necessarily. Flash tube systems dump an enormous amount of energy into the Xe plasma in a very short amount of time. LEDs can be strobed, of course, but I don't know that they can achieve the extremely high ratios of peak current to average current that a flash tube can withstand without being destroyed.

    I don't have any data on this, but my understanding is that these parts exhibit significant afterglow, lasting on the order of many microseconds up to several milliseconds. IIRC, I've read they have a complex afterglow "tail" since the white color is obtained by a mix of phosphors, each having its own decay characteristics. (Caveat: this is all secondhand info, unverified and cited from my somewhat dim memory.)

    Any LED, like all diodes, is going to be tolerant of rapid switching; what they will NOT tolerate is any significant reverse bias (>4V or so) for even the briefest of durations. With high-power LEDs driven by short pulses of many amperes, I imagine that paying close attention to lead inductance is necessary and it's probably a good idea to parallel the LED with a co-located Schottky diode to clamp any reverse "kickback."

    Hope this helps a bit...
     
  7. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,061
    3,823
    @Hypatia's Protege

    I've tested with photodiodes (photo-transistors are too slow). LEDs are extremely fast. Even with phosphors, white LEDs are very fast. "Phosphors" is a misnomer because most of the color-shifting coatings are actually fluorescent (nearly immediate quantum-mechanically "allowed" triplet to singlet transitions) that occur on the low nano-second timescale. Phosphorescence is a "forbidden" decay of an excited molecule and can occur in the high nano-seconds to minutes time frame (can be described as a half-life like nuclear decay).

    I don't know your definition of rapid switching or high power is exactly but 100k Hz is easily done and measured. In most cases, the LED will not be your biggest problem at high speeds.

    LEDs have a datasheet just like every other component and the On lag, off lag (or some company-specific analogous measurement is usually given. High-nanosecond range is common for white, less for red and blue/UV. Many high brightness GaN green LEDs are actually made with phosphors so a bit slower.
     
    Last edited: Dec 22, 2015
    cmartinez, Hypatia's Protege and #12 like this.
  8. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,292
    1,255
    [​IMG]
    They make RGB ones that might be faster.
     
  9. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,061
    3,823
  10. hp1729

    Well-Known Member

    Nov 23, 2015
    1,957
    219
    On a quick search I couldn't find such things mentioned on an LED data sheet. It makes sense that they must be fast since they are used in fiber optics which can reach pretty high speeds. I also would expect the "phosphor" type white LEDs to be slower than other types.
    Can we devise a fixture to measure such a characteristic? A high speed counter that starts when power to the LED is removed and stops when a phototransistor senses decay of the light. A scope would probably be easier.
    Phototransistors have a rise and fall time in the microseconds, at least the one I checked on did. This would be more than suitable for an RPM sensor for a motor under 10,000 RPM if you just wanted to monitor the RPM instead of using a strobe light.
     
    Hypatia's Protege likes this.
  11. #12

    Expert

    Nov 30, 2010
    16,316
    6,818
    That looks like a 20 nanosecond tail...50MHz on the half cycle. Fiberoptic service for Internet use suggests faster methods are employed. Or am I being fooled by TV commercials for 30 megabyte (30MHz x 8 bits) performance?
     
    Hypatia's Protege likes this.
  12. hp1729

    Well-Known Member

    Nov 23, 2015
    1,957
    219
    I don't grasp what this graph represents. On one side is 0.05 to 0.50 ... what units?
    On the time scale I would expect a comparison between exciting voltage and LED light.
     
    Hypatia's Protege likes this.
  13. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,061
    3,823
    Listen carefully, they advertise 30Mbps - lower case "b", means Megabit per second, not byte. Scammers!
     
    Hypatia's Protege and #12 like this.
  14. Hypatia's Protege

    Thread Starter Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    Thanks for your informed responses! -- Sounds encouraging!:)
    Time for me to actually (gasp!) bench test it -- (an ounce of 'empiricism' and all that...;))

    Again, many sincere thanks for your knowledge, insight and suggestions!:):):)

    Best regards
    HP:)
     
  15. #12

    Expert

    Nov 30, 2010
    16,316
    6,818
    Is that the same meaning as, "One valid test result is worth a dozen theories."?
     
    Hypatia's Protege likes this.
  16. Hypatia's Protege

    Thread Starter Distinguished Member

    Mar 1, 2015
    2,800
    1,237
    I wouldn't go that far:D --- 'Twas merely an acknowledgment of my proclivity for the 'abstract' -- Now in possession of some knowledgeable, affirming, feedback -- 'tis time my mitts were 'sullied' with some real-world (i.e. laboratory) investigation:eek::D

    Best regards
    HP:)
     
    #12 likes this.
  17. #12

    Expert

    Nov 30, 2010
    16,316
    6,818
    Abstract don't feed the bulldog. Get in there and smoke some transistors!
     
    Sinus23 and Hypatia's Protege like this.
  18. RichardO

    Well-Known Member

    May 4, 2013
    1,235
    384
    I hear what you are saying but my eyes see an afterglow from white LED's. :confused: I have been meaning to verify this observation for quite some time usxing a photodiode...
     
  19. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,061
    3,823
    The point source is very bright. It is more a question of object permanence in your eyes than of the phosphor on the LED.
     
  20. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,061
    3,823
    If you don't have a photodiode (faster than photo transistor), make your own by simply using a old-school green LED - cathode connected to ground and anode connected to a 1M resistor and other end of resistor connected to ground.

    Then connected the resistor cathode node to an amplifier. Always make sure the absorbing LED is longer wavelength than your transmitter LED - but not too much. An IR LED will not detect flashing from a blue or green LED. As mentioned above, lots of newer high-brightness GaN LEDs are emitting blue or UV and red-shifted with phosphors. That means, you cannot use certain green LEDs to detect blue light.



    Good luck.
     
    RichardO and Hypatia's Protege like this.
Loading...