White LED strobe

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
rshayes
Posts: 1286
Joined: Tue Mar 04, 2003 1:01 am
Contact:

Re: White LED strobe

Post by rshayes » Tue Dec 23, 2003 3:15 pm

An LED is basically a semiconductor junction. This has an exponential current vs. voltage characteristic. This means that a step increase in voltage will multiply the current by a constant factor. In a silicon diode, increasing the voltage by about 60 millivolts will increase the current about 10 times. A 120 millivolt voltage increase increases the current about 100 times, and so forth.<p>Practical diodes have some series resistance, depending on the diode design. This is mostly in the contacts and bulk resistance in the substrate. The voltage drop across this resistance is proportional to the current. At low currents, the diode behaves in an exponential fashion, changing to a resistive characteristic at high currents. In signal diodes, this change occurs at about 1 milliamp. Higher curent diodes with larger junction and contact areas will enter the resistive region at higher currents, on the order of 100 milliamps to amps.<p>In the short term, the diode will most likely fail by overheating. The plastic package used for LEDs seems to be capable of dissipating about 100 milliwatts. In the exponential region, the power dissipation will be slightly more than proportional to the current, since the voltage remains approximately constant as the current increases. In the resistive region, the power dissipation is proportional to the square of the current, since both the voltage and current both increase together. The light output is proportional to the current. In the exponential region, the efficiency (in terms of light output per watt input) will be nearly constant. In the resistive region, the efficiency will be lower as the currrent increases, since more energy will be lost in the resistive areas, which do not emit light.<p>In the exponential region, shorter pulses at higher current will result in the same heating effect with higher peak light output. When the current enters the resistive region, the decrease in efficiency will limit the peak light output even though the pulses are shorter and the peak current is higher.<p>The highest peak brightness will probably occur at about the current where the diode is entering the resistive region. This assumes that the duty cycle is reduced to keep the junction temperature constant. I would expect that the pulse width at this point would be in the 1 to 10 millisecond range as a rough guess.

User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: White LED strobe

Post by Chris Smith » Tue Dec 23, 2003 11:18 pm

LEDS like the first lasers are capable of handling current in excess of 50 to 100 AMPS*<p>*Amps, are the result in this case of a division of time, in that you can pulse a led at nano seconds of time, times hundreds of volts, and due to resistance and time frame, you accumulate a total 50 or 100 amps flow for only a few nano seconds, times those nano seconds, and still only equal the same milli *watts as a constant current at a constant voltage. 20 milliwatts at 3 volts, for example.<p>Once you turn on a diode junction, you must also clamp it just as fast, lest it burns out.

csbaker77598
Posts: 1
Joined: Thu Dec 25, 2003 1:01 am
Location: Houston, TX
Contact:

Re: White LED strobe

Post by csbaker77598 » Thu Dec 25, 2003 8:17 pm

Cheap White LEDS
I've ordered from SuperBrightLEDS and they're good. <p>Even cheaper white LEDS from LSDIODES.COM.
For example,
6000mcd .60/each
8000mcd .75/each
12000mcd .90/each
Shipping is $2 if total is under $20.
Packaging is not great, just all the LEDs taped together, but they work and are cheap.
Their prices are not the cheapest for other colors, especially red.<p>[ December 25, 2003: Message edited by: csbaker77598 ]</p>

rshayes
Posts: 1286
Joined: Tue Mar 04, 2003 1:01 am
Contact:

Re: White LED strobe

Post by rshayes » Fri Dec 26, 2003 3:18 am

If you are lucky, the quantum efficiency (photons out per electron in) is constant as current increases. The data sheets for LEDs often don't give very much information on voltage vs current characteristics. The slope of this curve at high current will give you an idea of the series ohmic resistance of the device. It may be in the neighborhood of 1 ohm, which would be consistent with a voltage drop of 100 volts when driven with 100 amps.<p>The junction area probably heats and cools in under 1/10 second. If the LED is operated at 30 mA with a forward voltage of 3.2 volts, in 1/10 second, the diode will receive 9.6 millijoules (3.2 volts x .03 amps x 1/10 second). The total charge transferred will be 3 millicoulombs (.03 amps x 1/10 second).<p>Now drive the diode at 100 amps. For the same pulse energy, the time must be shortened to 960 nanoseconds (100 volts x 100 amps x 960 nanoseconds = 9.6 millijoules). The total charge transferred will be 96 microcoulombs (100 amps x 960 nanoseconds). The total light in the pulse will be about 3.2% of that emitted in 1/10 second. The peak light output would be about 3000 times higher, but the human eye will tend to respond to the total energy rather than the peak power for pulses this short and the total energy is much less.<p>The pulse of white light probably won't be that short anyway. White LEDs are usually blue LEDs exciting a phosphor mixture which converts the blue light to longer wavelengths. Phosphors emit light with time scales in the microsecond to second range. The yellow phosphor used in the P7 CRT screen would emit light for several seconds after it was excited. This was used for Radar displays, where an image was scanned in about 30 seconds. At the other end of the scale, the blue phosphor used in the P15 screen for flying spot scanners emitted its light in a microsecond or two. The phosphor mixture used in the white LEDs probably falls somewhere in the middle, possibly about 10 milliseconds or so. Driving the LED with shorter pulses than this would not produce shorter output pulses, but would reduce the efficiency of the LED, and thus its total light output.<p>It is also possible that the phosphor may saturate and be unable to handle high light intensities.

N9AOK
Posts: 22
Joined: Fri Aug 01, 2003 1:01 am
Location: Worthington, IN
Contact:

Re: White LED strobe

Post by N9AOK » Fri Dec 26, 2003 10:24 am

Great Stuff, Thanks.
My experience falls right in line with what you said...shorter, higher current, pulses dont seem to give me much brighter flashes.

User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: White LED strobe

Post by Chris Smith » Fri Dec 26, 2003 10:54 am

"shorter, higher current, pulses dont seem to give me much brighter flashes" <p>Shorter higher current pulses do give out much greater flashes. Yes, there is a limit based upon the eye, and the phosphor. My standard over the counter flasher unit, can be seen at a mile or more, with the naked Eye. Pretty good for a single Led, and a 2 battery supply charged to higher values, and pulsed. <p>However if you wish to transmit and recieve a led at several miles, its the only way you will aclomplish it.

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: White LED strobe

Post by MrAl » Fri Dec 26, 2003 9:36 pm

Hello,<p>Another idea is to use multiple LEDs.
Multiple LEDs are always more efficient then
one single LED operated at the same total current
no matter how they are driven (pulse, continuous).<p>The drawback to using more then one LED is of
course you have to buy more, and also that
you have to deal with the multiple focus points.<p>Since efficiency is a prime motivator with
battery operated equipment,
if it needs to be portable multiple LEDs
will run longer with the same set of batteries
for the same light output, pulsed or not.<p>If it's not battery operated, then driving
the LED at a higher current (and thus losing
efficiency) might not matter and will save
on the purchase price of the LED(s).<p>
Take care,
Al
LEDs vs Bulbs, LEDs are winning.

toejam
Posts: 253
Joined: Wed Dec 05, 2001 1:01 am
Location: n.c.
Contact:

Re: White LED strobe

Post by toejam » Mon Dec 29, 2003 3:30 pm

ok now her's one. Has anyone ever put a white led through a prisim? If so, what is the result?

User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: White LED strobe

Post by Chris Smith » Mon Dec 29, 2003 5:58 pm

No, but expect the same/simalar as a fluro light.

chessman
Posts: 292
Joined: Tue Jan 14, 2003 1:01 am
Location: Issaquah, WA
Contact:

Re: White LED strobe

Post by chessman » Sat Jan 03, 2004 10:12 am

I'm not going to go into the whole technical aspects of the project.<p>What I want to say is that a conventional strobe is not instantly full brightness and then off again as an LED is. An LED does not take nearly as long as a flash tube to "light" to full brightness.<p>Jolly Roger, the reason the LED looked more like a strobe might be because PWM could have been used on the unit to vary the fade-in/fade-out times. Granted, these fade times would be extremely short (not going to throw out a number, I'd be off anyway), but the brain is extremely precise in these things.<p>Just thought I'd throw that out there :cool:

Bernius1
Posts: 349
Joined: Sat Sep 21, 2002 1:01 am
Location: NY
Contact:

Re: White LED strobe

Post by Bernius1 » Fri Jan 23, 2004 6:01 am

I can't get this thread out of my head !!! Two things come to mind; I remember a short-pulse switching circuit without a load resistor, where pulse duration limited avg.current. The 'active turn-off' clamping side sounds familiar. The other thing is; I saw keychains with SMD LED 'strobes', and thought,
What about 1 or 2 blinking SMD LED's with a hearing-aid battery, all on a round 'band-aid', that can be stuck to a ballon? I got the idea while chasing my nephew around,blinking a flashlight into a balloon. In a dark room, the balloon glows like an orb. They now sell glow-in-the-dark adhesive stars to stick on the ceiling. If these "blink-aids" sold for under a buck, how many moms would try them?? Happy Birthday.
Can't we end all posts with a comical quip?

Post Reply

Who is online

Users browsing this forum: No registered users and 65 guests