Emergency light project

Someone posted about a light like this. It looks pretty good doesn't it? BUT what a bummer! You have to keep squeezing it, at high speed, for as long as you need the light! The second you let up, the glow dims and dies. Not good.

So... add this and this and 2 LEDs, some solder, a little creativity and voila!

Ok... this post lacks a little because I still can't post pictures of the finished product. I'm working on it. Item 2 was a IC bridge rectifier diode (a small square about 5mm on each side), necessary because the output from the generator part is AC, which is fine for filament globes but terrible for LEDs, and the second item which is that 1.0F 5.5V supersized electrolytic capacitor mentioned in another thread.

I set up the LEDs parallel with the capacitor, separated by a simple switch so that I can switch the LEDs on or off. I also added a Schottky diode (1N5819) on the negative terminal of the bridge rectifier between the capacitor and the rectifier because I didn't want the capacitor to discharge in to the generator coils. Was this necessary?

I also used 2 Nichia white 5600mcd LEDs in parallel. Soldered them in to an old PR bulb base. Should I have gone with only 1 LED? Would that run brighter or can the capacitor supply enough voltage and current to run both LEDs at the nominal brightness?

I will strive to put photos, test parameters and results up as soon as possible, but I would like your input on whether the Schottky diode was really necessary and thoughts if I should have stuck with only 1 LED instead of using 2.

I've charged up the capacitor. Takes about 1 minute of moderate squeezing to get it up to 5.5V (open circuit). Will post runtime results soon.

hmm...do you need the rectifier?...LEDs are diodes, it's not that they don't like AC, they won't let AC pass, as long as the voltage not over their rating it should be fine...

i'm not real sure what happens after the generator passes through the cap, but i would imagine that this process smooths the AC into DC, or do caps pass both ways?...

I have a squeeze light similar to that one and just put in the 2 led bulb from my turtle light 2 no other modifications. Before the conversion, one squeeze would run the bulb for about 1.5 seconds. Now with the led it runs about 5 seconds on a squeeze.

Actually, they don't like reverse voltage. I had a couple of LEDs break down because of that. I wanted to protect the LEDs from that. As another plus point, the bridge rectifier helps convert what would normally have been lost as reverse current in to forward current. I'm not quite sure how much I'm losing because of this component though, since it's kind of hard to run the generator with the cover open.

Electrolytic caps are polar, ie they have a positive and negative side and you have to watch how you put them in. I believe that for ceramic caps, it doesn't matter which way you insert them in to the circuit, but in the same way, the side which is connected to the positive supply terminal will become positive.

I think that in the case of the electrolytic caps, a reverse current will cause some internal damage if the voltage and currents are too high. For ceramic caps, you wouldn't get anything useful from it if the input is full reverse AC because the charge on the capacitor keeps changing and you lose any charge you may have built up from the previous cycle. So for all intents and purposes, I think the bridge rectifier is unavoidable in this circuit design.

KenBar, if you read this, I believe you took apart a predecessor of the NightStar. Did they use a rectifier circuit? What were the other components in there?

I've been playing with this light for a while. The output is about that of a CMG Infinity if you run it straight off the capacitor. But pumping it the way you would have had to before putting in the modifications, you can get a very bright light that is a little better than the Arc Flashlight.

BTW, I'm running from the capacitor to the LEDs without a resistor. I believe that is the best way to do it? But between the generator and the capacitor, there is a 10 ohm resistor from the original circuit. I believe it was put there so that a current will be introduced in to the circuit. Since the velocity of the magnets through the windings will determine the output voltage, it is hard to think of a way to determine the optimal resistance to insert, such that a reasonably moderate pumping will produce the required voltage and the maximum amount of current. If anyone has an idea, please let me know. (Voltage has to be within 30% of the capacitor voltage of 5.5V for charging to take place.)

Thanks for any further input. Papasan, thanks for enquiring.

This is one of the neatest projects that I've seen on the board so far. I've got one of the Russian flashlights ordered that I want to do the same sort of mod on and I'm trying to glean as much useful information as I can before I start in on it (so that I don't ruin anything, hopefully). Concerning the memory capacitors, I saw what I believe to be the same capacitor for only $4.00 at Jameco. http://www.jameco.com/cgi-bin/ncommerce3/ExecMacro/Jameco/home.d2w/report?cgrfnbr=657&ctgys=503;516;657; At that price, you could add almost 3 for the price of one of the others that was listed earlier. How big are these memory capacitors? Has anyone got theirs in yet? How many do you guys think could reasonable be fit in one of these hand powered lights. How many do you guys think would be prudent to try to put in one? I know if you go overboard and put in too many, it would take a long time to ever fully charge them. How much power do one of those generators put out when cranked at full speed? Has anyone measured the output on one yet? Also, if the voltage is only 5.5 volts, shouldn't you use a step-up if you are going to use a 2 LED bulb, or perhaps a better solution, why not just use a resistor and a single LED bulb? Which would be the better way to go? Wouldn't the resister idea lead to more useful light with less hand pumping? I can't wait till my Russian flashlight comes in so that I can join in on the fun. Just imagine, with this sort of setup, we should be able to build the ultimate emergency flashlight that will never need batteries and that you could store for a lifetime! I'm very excited about this project because this sort of thing has been a life long dream of mine. Am I a true flashaholic or what? Hehe.

Wouldn't you have to do the same amount of cranking whether you have a capacitor or just do what I did?

Each time I crank I get instant bright light that lasts for 3-5 seconds and then I crank again. It seems to me that if you are using a capacitor that allows the light to run, say, a minute... you would have to crank the thing 12 times to charge up for that minute run.

I wonder how much useable light you would get during the initial cranking time. You would be trading instant light for longer light when not cranking but would still end up cranking the same amount?

Am I missing something here??

Steelwolf,
How many times do you crank it in a minute to get it fully charged? Also, how many cranks or how long before you begin to get useable light? Is the 3 minute run time useable light or until the bulb quits?

I am curious if 1 Led makes s significant difference in run time, it didn't in mine.

Steelwolf: Well, looks like you were right about the NightStar being just about the ultimate emergency flashlight. It does seem to me that it could have (and should have) been built much more compact though. For those that would like to check it out, the URL is: http://www.innovativetech.org/ I just spent the last hour reading everything on their web site and from the looks of it, it would be very hard to build a better emergency flashlight, but on the bright side, our flashlights that we are building will cost only about half as much to build. Still, just as soon as I can afford to get one, I would be very interested in purchasing a NightStar to play around with.

Where is everyone getting their Russian Dynamo Flashlights from in the States?
I've conducted a quick Google search and came up with these two sites which have the light for approx $11 USD. Is this a good deal?
http://www.preparedness.com/dynpumflas.html
and http://adidam.com/gateway/emporium/health/7086_26299_26345.htm

Regards, Ed

Edwood: I got mine from http://www.lighthings.com/ Just do a search for Russian and you will find the page. I paid $9.95 + $4.00 shipping for mine. Also I noticed that All Electronics has them for $9.95. http://www.allelectronics.com/index.html IMO, this seems to be the toughest, best constructed, dynamo flashlight for the money. I only hope that I (along with everyone's help) can do a good capacitor/LED mod on it. Did I happen to mention how excited I am about this project? Hehe.

Quick question. If you were going to use the 5.5-volt memory capacitor and a single LED LEDCORP bulb that is designed to run in a 4 cell light (6 volts), would you need to use a resistor at all? I guess a better question is this, when you are cranking the light, does the voltage ever go high enough over 6 volts to damage the LED, or is the voltage limited by the 5.5 volt capacitor and will never go over 5.5volts? Will the 5.5volt capacitor act as sort of a voltage limiter in the circuit or not?

You should be set with a 4 cell LED. Since the load is so small if you run the 4 cell LED the voltage is around 6.4v. So 5.5 should be fine.

Brock

Free: The reason for the capacitors is so that if you have to run the light for an extended time, like for example an hour, you won't have to crank every few seconds for the whole time. This would have many advantages besides just a lack of aggravation. Any situation that you need to do something with both hands, such as changing a tire, reading a book, or putting up a tent, you will need a way of storing a charge so that you won't have to constantly crank the light every few seconds.

I understand the concept. I am just not sure how much cranking it will take to charge the capacitor and have useable light. I am very interested to see if this is going to work though.

Since I seem to be the only one who has completed the project and have a running model, let's see if I can answer all the questions and hopefully someone can come up with a improved "mod2".

Firstly, to X-CalBR8:
The capacitor I have is probably similar to the one you referred to. This one is an electrolytic type made by Elna. It is a cylindrical shaped object with a diameter of 21.5mm and a height of 8mm. The legs are about 6mm long and 5mm apart. It is quite small, but I still could not find sufficient clear space to put the capacitor inside the body. It sits outside with 2 holes drilled in to the flashlight body to allow the connections through. BTW, your website quotes it in US dollars, the one I referenced is quoted in Aussie dollars. (Even then, your's is a little cheaper before GST and shipping.)

You would probably be better off using only one capacitor, but 2 in parallel sounds interesting. I'm using only one and it already takes a minute of hard squeezing to get it properly charged up to full capacity. But why I say the 2 in parallel set up might be better is because of the lower resistance of the capacitors when arranged this way. Also, it is easy to charge the capacitor up halfway, but seems to require more effort as the charge reaches capacity, and then seems to lose it faster when the charge is near full capacity. So by having 2 in parallel, you might find it easier to charge and achieve longer runtimes for less effort.

I put in 2 LEDs because I figured that the capacitor should be able to supply the needed current and all that would suffer would be the run time. The intensity should be about 1.5x of using only 1 LED. Will tell you more about it when I build the 1 LED bulb. The only advantage I can find with using 2 LEDs is only when you are actively pumping the charging handle while the LEDs are running. It gets very bright. Not like the Arc as I stated earlier but closer to the PLW3. With a single LED, the runtime should approach that of the Nightstar (5min), but you lose out because it wouldn't be as bright when you pump the light while it is in operation. In both cases, I really don't think a resistor is needed, but I could be wrong. Hope Craige or Gransee jumps in soon.

As for the generator out put, I can't measure the raw output because that is in high frequency AC and my multimeter can't seem to get a good reading. Probably need a osciloscope for this. Besides, it is really hard to run the generator with the body open. Will try again, plus the post-rectifier voltage. Current is out of the question because everything has been soldered together.

As to the ultimate emergency flashlight, I think the NightStar is as good as it gets. The handcranked flashlight is not waterproof in the slightest, feels cheap in the hand and might break quite easily, and requires quite hard pumping which will increase the chances of something breaking or cracking. But for me, I think it's cheaper. About AU$35 for all the components (flashlight, Schottky diode, bridge rectifier, capacitor, LEDs, solder, wires and glue) compared to AU$120 for the NightStar.

KenBar, I would like to find out how the emergency light performs with the 9V step up. Do post and let us know. About the NightStar, I understand why you would not want to talk too much about it. But please just nod if I'm on the right track with the bridge rectifier. I sort of figure that it is needed because the output from the NightStar's generator would also be AC. And I really wish I knew if that Schottky diode was really needed.

BTW KenBar, did you get the pictures I sent? Were they clear enough? I got to get me one of them close-up lenses real soon.

The run time of the light I built is about 3min before it gets a little too dim to be useful. But it actually glows for a very long time. The advantage of the squeeze type generator over the shaking generator (a la NightStar) is that you can pump up the power even as you use the light. And it does get very bright. Almost like the PLW3 that I have. But having the 2 LEDs has probably reduced the run time quite a bit. I'll probably buy a few more cheap PR globes and replace it with a single LED and see how long that runs. And I keep wondering about that Schottky diode (I'm starting to repeat myself, so you know it is really bugging me.)

Happy experimenting guys, and don't hold back those thoughts and questions. I will try to answer if I can.

But will the 5.5-volt capacitor act as a voltage limiter? Or what about when the capacitor reaches it's full charge, will it ever allow more than 5.5 volts to travel through it when you are actively cranking the light?

Steelwolf: Thanks for taking the time to give us such a long and informative post. I'm anxious to get every little detail that I can so that I can gain a better understanding of exactly what is going on in the circuit. About the 13 volts that you mentioned, did that occur before the capacitor or after it? 13 volts would blow a single LED, for sure, is why I ask. I believe you are definitely right about 2 capacitors being better than 1 for various reasons. I think that is what I'm going to try on my mod just as soon as I get my Russian flashlight in. It should be in tomorrow, I hope. I wonder if Radio Shack carries the 5.5 volt memory capacitor. If they did, it would save me from having to order them and wait days to get them in. I'm also anxious to hear from someone that has taken voltage readings from the Russian flashlight to see if they are the same as the other hand cranked flashlights that have been tested. Has anyone got theirs in yet?

The 13V is after the capacitor. This was measured across the contacts for the bulb. The LEDs are not any the worse for it. I really doubt if the capacitor dumps 13V across the LEDs. This is more likely to be a reflection on the charge available rather than the actual output voltage. I have tried running the LEDs while handcranking like mad, and plugging in the LEDs immediately after measuring the 13V. In both cases, the LEDs survived and there was no noticeable damage. BTW, my setup is 2 LEDs in parallel. Haven't tried single LED.

Hmm… That is strange that a 5.5 volt capacitor could allow 13 volts to exit it. That is just such a large voltage difference. I wonder if this is the effect of the capacitor or the result of the large peak voltage that is produced by the light when cranking at very fast speeds. I suppose that proves that the capacitor does not act as a voltage limiter if it allows that large of a voltage to exit it. Didn't you say that you had also included a rectifier circuit before the capacitor? It seems as though it should have a cut off point of the amount of voltage it would allow to exit it also. Have you tested the voltage entering the capacitor to see what the peak voltage is there? It would be interesting to know if it goes over 13 volts at that point. Also, does the voltage exiting the capacitor only hit 13 volts after the capacitor is fully charged or before? What troubles me is this: If the high voltage peak is a result of the capacitor, then the effect could be doubled when trying to use 2 capacitors. Also, the reverse might be true. 2 Capacitors could act to smooth the voltage out and keep it from spiking when cranked vigorously. It would probably take some experimentation to find out which theory is the correct one. Wouldn't it be possible to build a voltage clamping (or even a voltage smoothing) circuit at the capacitor exit to keep the voltage from spiking above 6 volts and keep the LEDs from receiving a deadly surge? I'm very surprised that you haven't had a blown LED, even though you are using a 2 LED bulb instead of 1, with that large of a surge.

More answers to your questions. Hasn't anyone else built theirs?

Free: You probably have to do a heck of a lot more cranking. One minute of moderate cranking will give you about 3min of light at the intensity of a CMG Infinity. Or you can really open all stops and crank like a maniac. That gives you a brilliant light that dims in about 30sec to the Infinity level. The light seems dim, but in darkness, is quite usable for a long time. The capacitor discharges very quickly at the start but reduces its output to a more sustainable level and provides a long run time. Also, because the capacitor output reduces as time passes, the LEDs actually don't go out for a long, long time. They just keep glowing.

I think that because of the character of the capacitor, it might actually make sense to use 2 in parallel. The charge is more stable but the available current is higher and you don't have to crank it as hard to achieve a stable charge. What do I mean by stable charge? If I were to crank like a maniac on steriods and speed, I can actually bump the open circuit voltage all the way to around 13V. Of course this is very unstable and the capacitor quickly loses its charge. It seems happy to stay around the 3-4V range forever, not quite at the 5.5V advertised. I believe that 2 capacitors in parallel would give us brighter LEDs for a slightly longer runtime with less maniac cranking required.

Of course, all these are open voltage values. I believe the capacitor acts as a sort of voltage limiter. I managed to hook up my multimeter while I was cranking like a madman, but only got about 6V. When the rotor stopped spinning, the reading jumped to about 13V. Managed to repeat this result several times. I'm going to try and hook up the multimeter so that I can measure the output voltage from the rectifier for different rates of cranking.

BTW, I refer to moderate cranking as about 2 cranks a second. But that only tells half the story because if you pump hard, you can accelerate the generator to a higher speed in between each crank. It has to run quite fast because you need to achieve approximately 5V to charge the capacitor. But this is no worse than trying to run the original bulb at a yellow glow.

Hope I managed to answer all the questions clearly.