LED atrobe

[snip]

And perfect isn't always simple :-) ...Jim Thompson

-- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

| 1962 |

Remember: Once you go over the hill, you pick up speed

to

Halted Specialties had a pile of small xenon photo strobes. [Might still have them.] I have half a dozen in my junk stash. Any good designs around for firing a photo strobe? By good, I mean one that someone actually built rather than just a google search.

I have two laying about. One uses an NE2 to self-trigger the SCR. I forget what the other one used, except that it wasn't the NE2. I can go open them up, again, I suppose. Of course, both used a 300V to 4000V trigger transformer to pulse the metal strap.

Jon

At this moment, I invite you to web-search for a schematic of astable

555 oscillator

Involving Pin 7 (discharge)

So as to be favorable to uneven duty cycle between high and low.

Perhaps, get a 555, conect Pins 2 and 6 to each other, and from these run a capacitor to ground with value of .01 microfarad (uF), tolerance of

+/- 10%, and if the capacitor is a ceramic one, choose the temperature characteristic of the X7R one of the choices for ceramic capacitors.

I think 50 milliamps for .08 second every .5 second makes a "1 watt" or "high power" LED work very well here, probably not even needing heatsinking...

If using a 9V battery, assuming 8V "under heavy-ish load after some aging" for 50m mA, assuming 4 volts combined voltage drop of the LED and the pulldown transistor connecting to Pin 3 of a 555, I work out a value of a resistor to add in series with the LED to be 200 ohms. Nearby usable common values are 220 and 180 ohms. Pin 3 of a 555 in this case is "low" when the LED is to flash. So, connect Pin 3 of the 555 to the negative lead of the LED, positive lead of the LED goes to positive power supply. But add in series with the LED a

180 to 220 ohm resistor. I would suggest a 1/4 watt one is sufficient.

If you use some very-miniature 12V battery for garage door openers or the like, I would make the resistornmin series with the4 LED being 330 ohms. I still seem to think that 1/4 watt power rating is enough here.

With "timing capacitor" of .01 microfarad, and desire for maybe .042 second off, .08 second on... At this point, I come up with 100 kilo-ohms ("100 K" /"100K") for a resistor to connect from Pin 7 to connected-together Pins 2/6,

Pins 2 and 6 of a 555 get connected to each otgher here, and to one lead of a .01 microfarad capacitor, with the other lead of this capacitor going to "ground" AKA "negative battery".

So far here, I would add to this a resistor of 470 K-ohms from Pin 7 of the above 555 to "B+".

If anyone needs help from here, I think I have already gotten halfway to sci.electronics.basics, IIRC name of such newsgroup, and I subscribe to that one. Please bring up in that newsgroup any requests for further help here.

--
 - DEon Klipstein (don@misty.com)

Opps might have undersold the battery

38maH. High power indeed! (there is also a smaller one 18maH)

but I had forgot how nasty the 555 was - 7555 much better or maybe even a bit of CMOS 74C14 + an output transistor.

A stack of low power LEDs is the OPs idea. I think 1 or 2 high power LEDs is a much better idea

Simpler if not perfecter

Hmm. Yes, by a mere factor of 10! ;)

Yeah. Since you pitched an IC into the mix, first thing I did was to check a datasheet for quiescent current since this was a battery application.

okay.

Well, up to the OP. Since the web site that the OP linked was announcing the part as "Lets have a sale. Price em low," I kind of figured that was part of the goal.

Hmm. I _do_ happen to have some number of nice yellows here.... ;)

Jon

I was hoping you didn't mean neon bulbs

I used a diffused type that has a wider viewing angle. It just wasn't bright enough for daytime use. I didn't have time to pursue it further, but I think a single higher-powered LED will be better than a string of low-power types.

Yes, of course. A xenon tube should certainly be easier to see. The one advantage of an LED is that it doesn't need a high voltage supply.

That circuit is MUCH more complicated than the ones I have around here and that work very well from one D cell. At a guess? Twice as many parts.

Jon

Le 25/01/2011 07:36, snipped-for-privacy@gmail.com a écrit :

Thanks again to everyone that answered. I think I'll first try a 555 switching a small (10A?) power transister that will dump a cap into a series/parallel of LEDs. Why go this way? Light weight; it will be lighter than a xenon strobe. Toughness; xenon strobe tubes are glass

- parachutes sometimes fail to open/open completely. They also generate RF when charging, don't want to interfere w/RF tracking bug. Audible alert; already use them - $1 personal alarm from Dollar Tree, even comes w/batteries. The problem is that you can't hear it when it's 4000 feet in the air. 555 design - don't need to google circuits (thanks though) - I have a copy of "The 555 Cookbook" by Jung. Will make circuit using diode(s) to seperate charging/discharging paths so I can get a small (value) millisecond ON pulse every second or so. If I get this working to my satisfaction, I'll put up a link to the schematic, construction, and parts list/source. I doubt there will be a circuit board. I'll probably wire it "dead bug" then pot it in epoxy or two-part foam.

LEDs also survive the landing bounce better.

Jon Kirwan Inscribed thus:

--
Best Regards:
                Baron.

snipped-for-privacy@gmail.com Inscribed thus:

Did you see the link I posted earlier.

Can't beat free !

--
Best Regards:
                Baron.

My flash is only two terminals, not three, so I guess there are two types of tubes.

So do you have the schematic?

If you are flashing LED, presumably using duty cycle to allow a higher peak current, the clock hat to be at least 1khz. That is the total pulse time, not the ON time. That is the rule used in chip design, though technically it only refers to duty cycling the metalization, not the semiconductor junction.

Those flash tubes I have used before use 300V or more between two terminals and there is a strap of metal or wire wrapped midbody that gets a 4kV kick to stimulate the flashover. The

4kV pulse is needed only for a moment.

Yes. Here's one:

Here's another:

Something like those. My notes are pretty old, now, and simply extracted by examining the two devices. I don't know whether a diac or sidac was used or what voltage it triggered on. At the time, I hadn't even heard of a sidac. The D1 in both cases was unmarked, I think. No measurements were made of the boost transformer T1. I assumed, but didn't test, that the trigger transformer was a 300V:4000V type, since those were in common use for triggering.

Jon

Slight correction below (ground.)