LED atrobe

Is this for a night launch? Would you be able to use Cyalume chemical lighting, instead? How much weight and volume can you allow? What height of flight? Or is this merely for recovery only?

Some references that came up immediately:

"High-Intensity Strobe Payload," Mort Binstock, Sport Rocketry, Apr '94

"Night Launch Light Payloads," David Sollberger, Sport Rocketry, Apr '94

It appears that back issues can be still bought from the NAR back issue service. 800-262-4872

But I don't see that particular one. So I'd call and ask about it, if interested.

I've picked up some emergency xenon strobe lamps in the past that operate off of a single D cell. I suppose they could run off of a AA or even AAA. Very bright. You might consider adapting them. If you can afford the room. I believe they cost me about $12.

The directional nature of LEDs seems to me to be a bit of a problem. You can diffuse that (wax paper?) or use randomized specular reflection (crumpled tin foil?) EL strips and wire?

But if you are planning anything above 2000' at night, I think xenon flash is your only proper alternative as you are usually required to be able to follow it through the entire flight.

Jon

Model rocketry - as opposed to amateur rocketry - will have a payload limit of about 60 grams for the largest rockets and much less for more normal sized rockets, and the body tubes are as small as 1/2" in diameter. A perfect job for an LM3909 and a lithium button cell - if you can get a 3909.

How about an off-the-shelf blinking LED:

I'm not sure what level you are at.....

Perhaps look for a circuit using a 555 timer that flashes a LED.

Try running it from a 2 or 3 AAA batteries in series.

Find an ultra-bright LED, pick the LED series resistor to suit.

A bright green LED may be easier to see at night.

If you go white keep an eye on volts needed for the LED.

You'll probably need at least a couple of LEDs.

Hmm. Well, the xenon tube, trigger transformer, and BJT/transformer might total up less than half that. Might work if everything were stripped away except what was necessary. The 1/2" diameter I remember, but I was imagining that there are small enough xenon lamps. Straight ones, too.

I can post up an LM3909 equivalent from discretes, if needed. Might need to adjust things (a lot, to make it a tripler) to handle a white LED -- they tend to require _more_ than twice the source battery voltage of 1.2-1.5V. Also, the battery is involved in the discharge of the LM3909, and a lithium button probably won't cooperate without a decent sized parallel capacitor present. Another part to add.

A 3-BJT SCR+trigger circuit and a 1-BJT/toroid booster is crossing my mind to handle one or two whites or more. Might be no more complex.

Jon

How about a small radio transmitter. Then you can use a directional antenna to find the rocket. Because you wont see a LED in many circumstances, bushes etc.

Part of the reason for a light is so that using time lapse photo or video will give you the acceleration. I'm not sure a LED would be good enough. Normally a xenon flash is used, and they have far higher peak powers than any LED

--
Dirk

http://www.neopax.com/technomage/ - My new book - Magick and Technology

Well, thanks for all the responses so far. Let me put a little more detail into the question. We're launching model and mid-power rockets (black powder and up to G composite motors). Body tube size is 1/2" up. It's not for night or low light launch. I've got a xenon strobe powered by a D battery, but am looking for something lighter/smaller diameter. We use a radio beacon for some flights. I was thinking of using a string of 5-10 high power (14,000mCd)* white LEDs. They would be oriented to point in several directions. I also know LEDs can be pulsed at a higher voltage than they are rated for in continuous use. That's why I'm looking for something more than a simple blinker. Length is not as much problem as diameter. Perhaps I could make a circuit that used a 555 to discharge a cap charged by two 9v batteries? It's also not a big promlem if the circuit eats batteries. Some guys use electronics (staging timers, parachute deployment electronics, etc.) that use a new battery each flight. As long as it works for 30 minutes, that would be good enough.

I'm thinking of these:

You might reverse-engineer one of the small strobe lights used by bicyclists and pedestrians at night. If you can determine a part number for the strobe LED, there should be data sheets and application notes available on the web.

Fred

We have a bit more detail now. No need for a white LED - Red or perhaps green would probably do.

From his now posted size requirements, he is probably better off looking at 2 or more different circuits. If he can loft a "D" cell powered xenon strobe, then just changing that to a AA or even AAA NiMH will give a good weight saving (it only needs to run for 30 minuets)

For the little ones maybe use a vacuum formed nose cone (rather than balsa) as a diffuser and a flashing LED and button cells. Stick it right up top and you might be able to chuck out some ballast for a near zero performance impact. Or perhaps a 1 watt LED without a heatsink and strobe it using a fresh A23 + cap and a 555.

Maybe?

David Eather wrote: :

............[snip]..........

Huh? What does this have to do with music?

Ok, ok, sorry. Couldn't resist. :-)

I'm also interested in this. Some time ago, an aeromodelling buff complained that he often finds it difficult to tell if his plane is facing away or towards him. I thought pulsed LEDs, green and red facing forward and rearward respectively, would be useful. Ground trials with a hastily rigged up strobe using a low-power red LED pulsed at ~100mA indicated that it wouldn't be bright enough. A highly directional super-bright LED wouldn't suitable.

On a sunny day (Wed, 26 Jan 2011 01:46:47 +0530) it happened "Pimpom" wrote in :

For a bit more money you can put in a small PAL camera like this for example: ftp://panteltje.com/pub/PAL_camara_module_in_circuit_img_1584.jpg and a modulated oscillator at some analog TV channel. Then use a portable TV to get a real cockpit view. It can be done digital too, for example some 4 rotor model helicoper uses an iphone for control:

I think the only limitation is thr weight.

Yes, I like the xenon strobe concept. A straight-line xenon bulb could be used (or found in it already) together with an existing circuit picked up for between $10 and $20, in 1's. Strip all the enclosure stuff out and reduce it to the basic circuit and tube and power it from a AA. Simple, very effective, etc. I don't have a datasheet on those, but those I have here set up about 300V before triggering the xenon pulse on what I think is a 400V 4.7uF cap. Joules would be about a quarter-Joule per flash. Figuring 50% efficiency (I've no idea, but why not?), that's half a Joule from the battery per pulse. Their rate is something like 1-2 flashes per second.. so even at 2 it is 1J/s or 1W. That's going to push the 30 minute period, perhaps. But that's all hand waving at this point here with near-random estimates at efficiency and I think it might be just fine in the end.

Assuming five of his LEDs (3.6-4V@20mA) stacked up and using the worst case of 4V each works to 20V@20mA = 400mW. Assume only 80% efficiency, then it's 500mW from the battery.

From a 1.5V AA battery, let's assume discharge only to 1.2V, sourcing 1/2 watt will yield about 70 minutes operation. Down to 0.9V, maybe 200 minutes or more. All of which is more than 30 minutes. So I think there is solid margin there for continuous operation, without blinking. Pulsing might be more attention-getting and would, of course, yield longer periods still.

From a 1.5V AAA battery, down to 1.2V, it might only run for

25 minutes or so. Down to 0.9V, maybe 55 minutes, typical. So it _might_ be possible to run continuously with a AAA. And certainly so with pulsing.

The advantage of AA alkaline is that they are available everywhere in the world and provide very good, all around value.

The Sanyo eneloop rechargeables would last about the same time, or perhaps longer if the datasheet is right.

Half a watt trapped in a 1/2" tube (some of the OP's use) might be something to consider. But the OP can test it and see.

Why exactly he might prefer those particular white LEDs, I'm not sure, unless it is cost.

If continuous mode is okay for this use, then a high frequency boost using one BJT and a tiny toroid would be sufficient (along with at least one resistor.) The volt- seconds could be kept down with high frequency, so the size of the toroid also kept small. (Basically, all he has to do is look up Joule Thief on google, or look up some longish discussions in sci.electronics.basics with a similar search.

If it needs to be pulsed and still cheap and still use a single 1.5V battery smaller than D (say, AA) and still drive a few cheap white LEDs (stacked, too), then he still needs a boost circuit but filling a cap that is attached to a trigger (single BJT plsu diode comparator trigger) and an SCR of some kind that may be built from two BJTs and a diode, so that the LED pulsing is triggered when the cap reaches some threshold where the booster is designed to take the right time to reach. The booster would still need to operate at high frequency to keep the core volume and mass down, I think.

I don't know if there is an IC that already does that.

Jon

Hemispherical diffuser? Perhaps by roughing the plastic up with some ether, xylene, acetone mixture? Or by sand blasting or rough sand papering the darned thing? Or buy "frosted" ones. Or use more than just the two you used to get more distribution.

I still like a flashing xenon tube, though. Hard to miss. Limited life may be an issue.

Thanks for talking about that problem, though. Nice to hear actual experiences like that.

Jon

Fred McKenzie wrote in news:fmmck- snipped-for-privacy@5ad64b5e.bb.sky.com:

better to modify an AUDIO personal alarm for his model rocket,as the LED flasher would only be useful at night.there used to be instructions for that on Rocketry Online,which I think is now defunct.I could be wrong about that,though.

the LED bike light I have for my bicycle is pretty heavy for a model rocket(16 oz GW or less),it uses two AA cells,not all that bright,and RED LEDs,not white,less visible at a distance. white LEDs will take more voltage to drive.

those disposable film cameras have a nice electronic flash that runs on a single AA cell,and quite bright. The trick would be to make it a repetitive flasher. you can often get the flash circuit free from a sympathetic store clerk.

--
Jim Yanik
jyanik
at
localnet
dot com

A typo I'll never live down

The guys I know use red and green paint on the wing tips and work out direction from that.

I thought pulsed LEDs, green and

Jim Yanik Inscribed thus:

There is a circuit floating about on the web using a kodak camera flash unit to produce a flash every few seconds.

--
Best Regards:
                     Baron.

An A23 battery is the little one that does 12volt car alarm and similar remote controls. They have about 3-5maH capacity. They are only a dollar or two from HK. So, with a cap, a 555 and a high powered LED or three (and a short duty cycle) he should be able to meet most of his needs (I think - maybe an extra transistor for switching the current pulse through the led's )

3-5 mAH??? Call it 4. That is, say, 170 Joules.

That's without figuring in the quiescent current of the 555. A 555 might have a quiescent current as much as 15mA, near

12V. For 30 minutes, at 12V, we are talking about 324 Joules. Just running the 555 alone wouldn't last. The typical is 10mA at 15V. Even if we figure 7.5mA at 12V as a typical value, we are talking 162 Joules.

In other words, "a problem."

The 7555 is max 1/3mA at 15V, which would be less than 10 Joules over 30 minutes. Might need to go there, instead. Even so, we are now at say 160 Joules.

The OP wrote, "5-10 high power (14,000mCd)* white LEDs." They are 4V@20mA or 80mJ/s. And if 10, 800mJ/s. 160J/(800mJ/s) is about 200 seconds of pulse duration assuming no losses on the output. (The output is rated at 0.4V at 15mA sink for a 7555 and at 20mA, worst case, we could be talking .5V there or perhaps another 2 Joules lost -- so no real problem there.) So 200 seconds out of 1800, or call it a 10% duty cycle on the LED stack.

Okay. With a 7555, I think it may make sense except that a series stack of 10 would be 40V, not 12V. And 5 would still be 20V. A stack of 2 or 3 would probably work and leave time to spare (fewer Joules per pulse.)

Oh, well. Simple isn't always perfect.

Jon