What would a bleeding-edge Shakelight be like?

I keep seeing these threads on shakelites...

As a guy who hangs out at CPF and has read about them over and over I probably wouldn't touch one with a ten foot pole.

But having said that...

As a guy who's lived without electricity for long stretches on numorous occasions I'll admit to being fascinated by the concept of shakelights. I also know how just a little light is so much, much, much better than no light at all when the chips are down.

I would spend real bucks for a top of the line product that was designed for flashaholics and with top build quality. Today that might mean a Nichia CS U bin with a very high quality reflector all behind a UCL. I don't know enough about the rest of the components to specify what they should be...but some of us would.

I know that with the market flooded with cheap crud there would probably be no chance of making money on the one I've discribed, but I'd definately buy that one even though it would cost a few times more than the others.

The cheap ones have proliferated to such an extent among the Joe Sixpacks that I'd want the Super Shakelight just for the wow factor when I could show it to someone who had no idea what such a light would be like if it were designed and well built on the bleeding edge of technology. I'd love to have one in my kit. Remember, I'm also one of those guys who loves dim lights anyway.

So, what would a shakelite be like if it were made with the best parts available and designed and assembled with quality and efficiency in mind right from the get go?

Has anyone ever done any shakelight mods?

Any thoughts on a bleeding-edge Shakelight?

IMO, a hand-powered battery charger would be better. I dunno how viable that is, but it would be perfect for outages because most of our lights are battery powered anyway

I've seen solar powered battery chargers for AA & AAA at Advancedmart. Now if there was one for RCR123's I'd be all set. The ultimate in guilt free lumens.
I'd still go for a crank type charger if I could hook it up to my bicycle trainer.

Sub_Umbra said:

I would spend real bucks for a top of the line product that was designed for flashaholics and with top build quality. Today that might mean a Nichia CS U bin with a very high quality reflector all behind a UCL. I don't know enough about the rest of the components to specify what they should be...but some of us would.

Click to expand...

Sub_Umbra,

I like what you are thinking here... Can someone here design a top-notch shakelight?

I imagine something small, like an arc AAA. Something that could be carried around the neck.

Why not? If a shakelight can be done in a large format, why not a small format?

Having said that, I have no experience whatsoever with complex electronics. So, if anyone here cares to take up the challenge, I commend you

Dan

bruner said:

....If a shakelight can be done in a large format, why not a small format?....

Click to expand...

I think the main problem is one of efficiency. You are generating the power to be stored in the capacitor by moving a magnetic field past a fixed coil. The smaller the magnetic field and the smaler the coil, the more the light will have to be shaken. I think some of the better shakelights are a well designed balance between minimal shake time, usable light output, and acceptable run time in a reasonable size.

How about 4 or more of the latest high efficiency LEDs? Then underdrive them so they are even more efficient (10mA perhaps). Use a regulated or semi-regulated boost circuit so the brightness is more constant. Provide a moon mode so that the light doesn't just cut off. Lastly use rare earth magnets to get the most power from a small sized light.

And how about an indicator LED that flashes (just while shaking) so you know when the light is fully charged. And perhaps it could serve double duty by flashing to indicate low charge when the flashlight is in use (30 seconds before moon mode say).

This is starting to sound cool. I want one.

Another idea: have 7 LEDs like the famed Proton: 6 white and one red. Then you'd have the choice of moderate brightness white or night-vision preserving (and long running) red light.

Greg

Greg -- I want one, too.

What keeps tugging at me on this is the fact that some (perhaps most) of the shakelights are just hype and even the ones that were designed well are using old components. On some of the shakelight threads some have actually commented favorably on the individual lights that they own. If some of them are actually good enough to be useful I think that they could be better yet if top flight components were used.

Everything about shakelights have always tended to fall on the marginal side.

• Not much available power
• Not very bright LEDs
• Relatively low transmission lenses

And yet some find them useful.

What seems to be needed is an LED that will put out more light from less juice, focused by a high tech reflector that pushes the light through a high transmission lens that will let more light out where it can be used.

Since some of them are useful to some already it seems that with better components they could

• run brighter and
• run longer with
• less shaking

I should probably add that if you need something brighter than a non-Ultra Infinity to find your own bathroom at night you'll probably never have a use for any shakelight no matter what improvements may be made to it.

Greg,

I keep thinking about what you said about underdriving multiple high-efficiency LEDs... I think most shakelights in the past have relied on a focued beam. This is really interesting. I don't see any compelling reason why the reflector couldn't be dumped entirely and just go floody with an array of LEDs. It would be different -- but cool.

Yup. We need an Arc or HDS version of the shakelight. Everything engineered for the best quality and performance in the smallest sized package (AAA and CR123 respectively).

The ones I've seen are too big. Something the size of a 2AA maglite should be feasible (except maybe for limiting the size of the capacitor too much).

Bare LEDs shining out the front will loose the fewest lumens, since they are measured that way. So adding an AR coated lens would not drop the lumens much. Luxeons aren't as efficient and then you have all the optics or reflector losses that reduce the net lumens.

Greg

Sub_Umbra said:

Greg,

I keep thinking about what you said about underdriving multiple high-efficiency LEDs... I think most shakelights in the past have relied on a focued beam. This is really interesting. I don't see any compelling reason why the reflector couldn't be dumped entirely and just go floody with an array of LEDs. It would be different -- but cool.

Click to expand...

A shakelight will never have the power to be a "thrower". They would most likely be used for closeup work, so a multi-LED flood beam is fine I think (even though normally I like Luxeons).

Greg

It looks like max efficiency is at about 5 mA (see this thread http://candlepowerforums.com/vb/showpost.php?p=1052877&postcount=1).

Look at the BestHongKong 40000: 100 lumens per watt at about 5 mA.

So 5 mA x 3.1v x 6 LEDs = 0.093 watts, so that's almost 10 lumens.

A 1 Farad cap can supply 1 amp for 1 second with the voltage on the cap dropping 1 volt. Or 30mA for 33 seconds for a 1 volt drop. So if the boost circuit works over a 3 volt range (say 4V down to 1V) then you would get nearly 10 lumens for about 100 seconds. Not great. A 10 Farad cap would give you 16 minutes before moon mode. Still not great. That's the price of a regulated output I guess.

If you can tolerate 10 lu dimming to 5 lu over the run time then the average current is 3/4 of the above so the run time improves by 33%: 2.2 minutes for 1F, 22 minutes for 10F.

If you can tolerate 10 lu dimming to 33. lu over the run time then the average current is 2/3 of the above so the run time improves by 50%: 3 minutes for 1F, 30 minutes for 10F.

What was the size of 10F again?

Greg

My 4.7F 2.5v caps are each 0.500" x 1.366" so fit where an AA would go. I recall the 5.5v rated ones were far larger, but I don't have any to measure.

I can see it now, CPF members modding electric toothbrushes as shakelight hosts and their waterproof chargers. And why not? There's already a membrane switch on the side.

Put Flubber in the ends to bounce the magnet back and forth .

or

use some other means to amplify the input energy like a chamber for an explosive propellent like hairspray (like a potato gun) or use a spring (like pinball) or rubber band or cone (like a slingshot) so that the time spent adding energy is reduced. Or a zip strip like they used to used to propel the little toy cars. And add an adapter for CO2 cartridge like a BB gun.

Use available tech to reduce energy robbing friction within the tube. Use the most conductive/least resistive coil material that's reasonable for the price.

xochi said:

...Use available tech to reduce energy robbing friction within the tube...

Click to expand...

Could the coil be held away from the sides of the tube with magnetism? I wish I knew more about some of these subjects.

I would think that whatever method is used to reduce friction may be inhanced by pulling a vacuum inside the tube before sealing it up.

Hmm, I've been thinking about this after reading this thread about a week ago.

I believe that if you actually need this kind of light size doesn't matter as much as runtime. So with that in mind, I have thought about what type of light to mod;
1: It needs to be relative small and not to clumsy to shake
2: It needs to be big enough to accommadate a big capacitator.
3: It need to be long enough so that you can maximise the effect of the magnets travel(I'm no electrical engineer, so I might be wrong on this part)
4: It needs to be quite sturdy, as this is the light that'll shine when all others have stopped.

So, what easily accessable light do we have that covers these factors?
My choice is the M@G 2C (or 3C) light. It's got space, but isn't so beefy that it'll be hard to grip around while shaking.

I see two ways to do this, use the "C" as a host, and handbuild everything to minimize loss of that precious juice, or you could make drop in units with the cap/magnet/other stuff mounted on in a tube that is secured by screwing it in the tailcap threads, and have a pr-base led drop in in the front.

The new "tailcap drop-in" could include electronics and maybe some small leds that could act as both as a "batteryindicator" for the cap and as a "find me"-feature.

What do you think?

I like your thinking except for the Mag 2C idea. I think an aluminum body won't work: the moving magnets will induce eddy currents in the aluminum and waste energy there instead of inducing currents in the coils. Also, I think you need a light flashlight so you aren't tiring yourself out shaking it up and down.

Greg

greg_in_canada said:

the moving magnets will induce eddy currents in the aluminum and waste energy there instead of inducing currents in the coils.

Click to expand...

Hmm, didn't think about that, but I'm not really sure how a shakelight works either... :thinking:
I thought maybe they worked like a motor, with the magnet inside the coils, so that you could have the magnet travel inside a tube wrapped with coils, but I'm no EE :shrug:

greg_in_canada said:

Also, I think you need a light flashlight so you aren't tiring yourself out shaking it up and down.

Click to expand...

Good point! :sweat:

I think the best way to improve the shake light is to do away with the shaking. I see it as a verry in efficient way to generate energy. The hand crank or squeeze generators yield much more energy for the effort put in to doing the work. The only advantage I can see to the shaking method is the case of the light can be made compleatly sealed.

Our University has an engineering show every few years and one of the things the electrical engineering students do is demonstrate this effect by dropping a magnet through a copper plumbing pipe. They also do it with a ring magnet around the copper pipe. In both cases the magnet falls very slowly.

The coils will intercept some of the magnetic field but the rest will go through the walls of the flashlight and if they are conductive will induce eddy currents. So some of your shaking energy will go into heating up the aluminum (slightly).

Greg

Brlux said:

The hand crank or squeeze generators yield much more energy for the effort put in to doing the work.

Click to expand...

I got the squeezetorch from Qualitychinagoods, but I broke it when I tried to reconnect the batteries (yes they don't have a cap but batteries) so that I could run it continuously.(they weren't connected from the factory) Now it runs on batteries, but not from the squeeze action.

greg_in_canada said:

The coils will intercept some of the magnetic field but the rest will go through the walls of the flashlight and if they are conductive will induce eddy currents. So some of your shaking energy will go into heating up the aluminum (slightly)

Click to expand...

Ok