Seventeen milliwatts
Dec. 2nd, 2003 08:27 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
fivemackA blue LED, dropping 2.88V and consuming 5.93 milliamps, throws a visible circle of light a metre wide onto a ceiling two metres above it, and you get little spots in your eyes if you look directly at it.
This surprises me somewhat, particularly since I know the eye isn't all that efficient in the blue (I don't have a data-sheet for the LED, it came from ultraleds.co.uk and possibly their market is more "ooh, shiny blue thing to put on my tyre-rims" than the quantum fan, but hc/Ve gives 430nm); especially since the LED is specced to run up to 20mA power consumption.
I have a green one that consumes 5.27mA over 3.2V and lights the ceiling more brightly, though in a smaller circle; the red, 2.1V and something like 63mA, are a much brighter light. Also some infra-red LEDs, 11mA at 1.16V; obviously you can't see them lighting the ceiling, though I have a webcam that can detect them (to my slight surprise they appear white, so the colour filters on the webcam's CMOS must be uniformly transparent in the IR) -- but the webcam can't see the reflection on the ceiling, and I'm not sure if this is because the light is faint or the webcam sensor poor. And even a few UV LEDs, though I feel uneasy plugging them in (I don't have any obvious test for UV, and even 375nm light is not great for the eyes), so haven't.
Not quite sure why I picked now, rather than say fifteen years ago, to get interested in electronics again; I suppose I'm a bit less clumsy, significantly less prone to blowing components up to see the flash, and I feel more able to afford the components and the test equipment (though a multimeter nowadays costs £2.99). My vague hope is, with sixteen NAND gates, half a dozen 555 oscillators and a dozen LEDs of different bright primary colours, to be able to build an entertainingly flashy Christmas ornament -- though I probably don't have the craft skills to build a nice box to put it in.
This surprises me somewhat, particularly since I know the eye isn't all that efficient in the blue (I don't have a data-sheet for the LED, it came from ultraleds.co.uk and possibly their market is more "ooh, shiny blue thing to put on my tyre-rims" than the quantum fan, but hc/Ve gives 430nm); especially since the LED is specced to run up to 20mA power consumption.
I have a green one that consumes 5.27mA over 3.2V and lights the ceiling more brightly, though in a smaller circle; the red, 2.1V and something like 63mA, are a much brighter light. Also some infra-red LEDs, 11mA at 1.16V; obviously you can't see them lighting the ceiling, though I have a webcam that can detect them (to my slight surprise they appear white, so the colour filters on the webcam's CMOS must be uniformly transparent in the IR) -- but the webcam can't see the reflection on the ceiling, and I'm not sure if this is because the light is faint or the webcam sensor poor. And even a few UV LEDs, though I feel uneasy plugging them in (I don't have any obvious test for UV, and even 375nm light is not great for the eyes), so haven't.
Not quite sure why I picked now, rather than say fifteen years ago, to get interested in electronics again; I suppose I'm a bit less clumsy, significantly less prone to blowing components up to see the flash, and I feel more able to afford the components and the test equipment (though a multimeter nowadays costs £2.99). My vague hope is, with sixteen NAND gates, half a dozen 555 oscillators and a dozen LEDs of different bright primary colours, to be able to build an entertainingly flashy Christmas ornament -- though I probably don't have the craft skills to build a nice box to put it in.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2003-12-02 01:59 pm (UTC)http://members.misty.com/don/index.html
http://ledmuseum.home.att.net/
http://www.equipped.org/led_lights.htm
where there is a lot of information on LEDs.
For the UV LEDs, use a bit of fluorescent plastic or a new T-shirt, the optical brighteners fluoresce blue-white.
Something else that may interest you is the line of Toshiba Constant Current LED Drivers. These can help prevent blowing up the newer LEDs, which are not forgiving of overvoltage and overcurrent conditions.
http://www.marktechopto.com/catalog.cfm?Drill_Level=Series&DeptID=2100&SeriesID=1286
If you were to set up a shift register IC as a pseudo random number generator, and drove one of the Toshiba drivers you'd get a nice blinky light device with a lot fewer ICs.
no subject
Date: 2003-12-02 02:56 pm (UTC)I'm not seeing so much of a problem with over-current and over-voltage at the moment, because I've been carefully OTT with my current-limiting resistors - though I may have spoken too soon, one of my UV LEDs has no voltage drop across it either way round, I may have killed it. Any obvious source for fluorescent plastic? Would clothes-washing tablets work as a concentrated source of optical brightener?
[for IR, I had a known-working IR emitter which I could use to check that the webcam was an IR detector; I feel both chicken- and egg-less when it comes to UV]
(ah, I suppose I could simulate lots of tap-and-NAND configurations until I can find one with a reasonable period, and use the NAND chips I already have ... but that doesn't solve the non-existence of the shift register; and a single clocked flip-flop uses the whole of a 7400 quad-NAND)
And I have no idea whether the blinky-lights would look adequately random if I took them from several stages of the same shift register;
once you get to thinking of more than one shift register, I suspect I should go the whole hog and use a PIC instead. And, given that a PIC programmer costs as much as a low-end FPGA evaluation board, probably I go an even wholer hog and get the FPGA board. Gifts that costly to myself must wait until after Christmas, of course.
no subject
Date: 2003-12-02 03:57 pm (UTC)The wash tablets might work, if they list something like "optical brighteners". Fluorecent plastic rulers and toys, glow-in-the-dark stuff, like at
http://www.mcphee.com/
Shift registers - MM74C164 is CMOS 8 stage, 74164 is the TTL version. Many good, and even maximum length, feedback choices take no more than 4 XOR gates.
And here you can find some combinations for up to 32 stages. Longer than the minimum you need is OK, it makes it harder to see the pattern.
http://www.newwaveinstruments.com/resources/articles/m_sequence_linear_feedback_shift_register_lfsr.htm
How I found you? You are a Friend of Jon Singer, isn't that the way it always works?
no subject
Date: 2003-12-02 02:58 pm (UTC)no subject
Date: 2003-12-02 02:45 pm (UTC)Do you have anything in the way of large expanses of white fabric? Say, a sheet?
There's something in most commercial laundry detergents / brighteners that fluoresces very nicely. You've probably seen the effect on TV any number of times - someone's T-shirt glowing a nice bright barely-blue white? My white sheets do the same thing, as do all my other whites - t-shirts, dress shirts, socks, etc. ad nauseum.
no subject
Date: 2003-12-03 02:44 am (UTC)You have a webcam that does IR?! *boggle* *giggle*