DIY Handpower LED Crank Light

[joecarrow]

Hi All, I'm still quite new to CPF, so I appreciate your patience with my obvious questions.

I've recently decided that I'm interested in designing and building my own flashlight. My original intention was to locate the most state of the art LED I could find, and mate it with one or more 18650 cells. I have prior experience with these cells professionally, so it seemed like a minor jump to put them into a flashlight.

In my search for a bleeding edge LED, I discovered this forum. It quickly became apparent to me that I'm not the first to think of this kind of flashlight, and the market is practically flooded with such lights. I like doing things a little differently, and I really can't devote the resources as an independent Joe to surpass the attention to detail of the companies described in the LED sticky thread.

My original thread is here, and I have since decided that a Cree XR-E is probably the LED I was looking for.

Since I'm not satisfied with doing what everyone else has done, I said- why not make a hand-powered light that's a little bit more? I've been doing some research on the matter, and I found a thread about a bleeding edge shake-light, and a mag-light modified with crank generator.

I guess the main thing I'm asking in this thread is- has anybody seen another project like this? Do you have any suggestions where to look?

What would people be interested in for a documented project?

That's all for now, thanks!

 

[easilyled]

This sounds an excellent idea.

You could use either Cree-XRE's or Seoul SSC-P4-Ubins.

I've seen one or two manufactured with far less powerful Nichia leds and
they also look a bit delicate and as if they could break easily.

What I would be interested in was for one that's built like a tank, virtually
bomb-proof and that can take a lot of force from the motion that charges
it up.

In addition if it could store the charge so it could light up for quite a long
time before having to re-crank it, that would be great.

I wouldn't mind at all putting some reasonable energy into cranking it up
then if some really nice output came out of it.

It may be also worth using some adjustable system for the level of output
like a flupic driver for instance

 

[Daekar]

That's interesting... Those are about the same size as some of the motors we make at my job. I never thought about using them as generators, although that makes sense, considering we test them for Back-EMF... Hrmm...

 

[joecarrow]

Ok, so we have-

1) Durable. Saving on batteries is pointless if the light breaks.
2) Ergonomic power input. Allows forceful use of large skeletal muscles to allow significant power input into the light.
3) Long useful charge life. We don't want to walk around cranking.
4) Switchable levels, allowing power to be used as required.

This is a good start to a set of requirements, I think. Since it can be cranked in the field, I think that a single Li-Ion rechargeable of reasonable size should allow enough storage. I think that a Cree XR-E switching between 350 and 700 ma would be a good bet.

Do you know of an efficient board for this LED that's will will roughly switch between these two currents based on a single rechargeable Li-Ion?

 

[joecarrow]

Daekar- It's neat that you work there, I actually looked at your company's website before while sourcing motors for my last job.

A gear-motor seems like the only way to go for this project. It's the only way to get a few stages of high-quality high-reliability gears without breaking the bank. I'm actually wondering if a cordless drill could be made to work. Cordless drill motors come with really durable gears, and are much smaller and lighter than the crank light I linked to.

I think that a pull-cord is the right form factor to get a lot of useful muscle power. Here's an example. You could have a carabiner clip on the end of the pull cord, and clip it to any solid object, then use the flashlight handle body as a sturdy handle. If you could clip it to your foot, you could do bicep curls or keep your arm locked straight and use your leg power to pull the cord.

The trick with this whole thing is that you need a small motor for low weight, and you need to spin the motor very quickly to get good generation.

 

[joecarrow]

A little googling yielded this. Apparently someone out there has already managed to get 5.7 watts out of a cordless drill. I think that by using a pulley and a stout pull-cable, you could get more RPMs out of it, and thus more watts. Just a guess, but I think you could double that to 10 watts. Now we're in business!

I happen to have a damaged 18 volt cordless drill around here. I'll have to crank it up later with another cordless drill to turn it, and see what I can get out of it at various RPMs.

 

[Daekar]

A little googling yielded this. Apparently someone out there has already managed to get 5.7 watts out of a cordless drill. I think that by using a pulley and a stout pull-cable, you could get more RPMs out of it, and thus more watts. Just a guess, but I think you could double that to 10 watts. Now we're in business!

I happen to have a damaged 18 volt cordless drill around here. I'll have to crank it up later with another cordless drill to turn it, and see what I can get out of it at various RPMs.

Definitely let us know what you find out! That would be a COOL custom light...

I wonder if you could incorporate the 18v battery packs? I know the 18v drill I have is NiCad, so maybe that's not a good idea... hrmm...

 

[joecarrow]

I did my first test tonight, using my Makita 14.4v to drive the old broken drill. I had to take it all the way up to full speed in second gear to reach 12.5 volts.

Here's my setup so far:

The Makita has a no-load speed of 1300 RPMs, and it didn't get slowed down much by the load of the drill.

If I assume the voltage output is linear with speed and doesn't bog down much under load, it'd be 4.2 volts at around 400 RPM. With a handspeed of 3 ft/sec, that works out to a spindle diameter of less than 0.05" for a pull-cord style input. Since I don't have a rod of unobtainium and a spool of carbon nanotube, I guess I'll need to think harder about how to make this work.

I'll be back.

 

[RustyKnee]

I did my first test tonight, using my Makita 14.4v to drive the old broken drill. I had to take it all the way up to full speed in second gear to reach 12.5 volts.

Here's my setup so far:

The Makita has a no-load speed of 1300 RPMs, and it didn't get slowed down much by the load of the drill.

If I assume the voltage output is linear with speed and doesn't bog down much under load, it'd be 4.2 volts at around 400 RPM. With a handspeed of 3 ft/sec, that works out to a spindle diameter of less than 0.05" for a pull-cord style input. Since I don't have a rod of unobtainium and a spool of carbon nanotube, I guess I'll need to think harder about how to make this work.

I'll be back.

The load down will be pretty much proportional to the current output of the "generator" (I say pretty much as there are frictions from brushes etc whicj won't change with load current). This applies to all motors/generators.

The voltage should be proportional to speed if the motor is ok.

You want a really low speed motor as this will make a low speed generator. Or a gear box...but that adds friction again (might end up wasting more in friction than you get out in juice).

Stu