Hardware Flashlight 2.0




Introduction: Hardware Flashlight 2.0

A few months ago I posted my first Hardware Flashlight here on Instructables. Though it was a mediocre success, I loved the concept and wanted to have one that worked. So I thought of all the ways that I could improve it. Here's the list of possible improvements I made for the first Instructable:

" Silver solder the nuts together: This will allow electricity to pass through the wall (nuts) of the flashlight, and will make the whole thing much more durable. That way you won't need to worry about the positive wire (the red one) getting entangled in the spring, and will create a much cleaner finish.
Use heat-shrink tubing instead of electrical tape: My best guess is that the insulation of the LED is what causes the loose connection. Heat-shrink tubing would make it much cleaner and would probably eliminate any short circuits.
Use bigger Nuts: Bigger nuts would make everything much easier to work with, and you wouldn't have to drill them out for the button cells to fit inside. "

With this Flashlight, I used bigger nuts and heat-shrink tubing, and it really did turn out much better. I didn't want to buy and learn how to use silver solder, but I bought some glue better than Gorilla epoxy, which turned out to be perfect.

Step 1: Supplies

Usually I take a picture of all the supplies and tools, but this time I lost track of everything I used, so a list will have to do.
Some of the supplies may vary, as the flashlight could be made in various different ways, with various tools. Parts like LEDs may also have different forward voltages, so resistors my vary.
  • 3 white 5mm LEDs
  • 3  680 Ohm resistors
  • 5-7 stainless steel nuts and 1 stainless steel bolt: The size of the nuts and the bolt depends on the width of your battery. The battery should fit through the nuts very loosely.
  • 1 stainless steel nut a size or two bigger than the rest
  • 1 nut of softer metal (optional): I had to use a softer nut, because I found it too difficult to drill a hole into a stainless steel nut manually. A drill press could easily do the job.
  • 12V A23 or A27 battery
  • Small battery holder spring
  • 2  5x5 squares of perfboard
  • Mini tactile switch
  • Stainless steel switch cap or other part that can be used as a cap (I used a small screw)
  • Insulated wire
  • J-B Weld
  • Soldering iron and solder
  • Third hand
  • Drill: a drill press would be ideal, but I used a regular cordless drill by hand
  • Dremel rotary tool: With a metal cutting wheel and sanding bits
  • Superglue/ Krazy glue
  • Needle-nosed pliers
  • Scissors
  • Bare Paint
  • Metal hand saw
  • Clamps
  • Heat-shrink tubing

Step 2: Modifying the Parts

To make the bottom of the flashlight, cut off the end of the bolt until only about 5mm of the thread is left. 

Drill a 5-7mm hole into the side of the nut that will have the switch. Most drills will do the job, but a drill press would be ideal. To attain a clean hole, punch a marking into the center of the nut, then drill the hole using that center point. Use some oil as a coolant, to stop the drill and metal from overheating.

Saw a small 2-3mm groove into the inside of five of the nuts (including the switch nut). I used a regular metal hand saw to do this. The purpose of the groove is to allow a wire to pass through the battery compartment without blocking it, so the grooves should all line up with each other. 

Step 3: The Switch

Sand one of the 5x5 perfboards to a disc that matches the size of the nut. Make the edges of the disc slightly slanted, so that it fits neatly inside and doesn't stand out. To sand the perfboard I used a Dremel 3000 with a sanding bit.
Once the disc is the perfect size, bend the end of the battery spring outwards and solder it to the perfboard. The spring should not be wider than the disc, otherwise the battery could get short circuited. 
Once the spring is securely fastened onto the disc, superglue the perfboard disc to the nut. When you do this, check that the two grooves are aligned to allow a wire to pass through.

For the switch, break off two of the leads, so that only two remain diagonally. This eliminates any chances of any short circuits. Solder one of the leads to the spring from inside the nut, and superglue the switch to the perfboard to secure it. Make sure that the switch is easily accessible from the hole, so that a cap can easily be glued on afterwards.

Step 4: LEDs

Sand another perfboard disc as before, but to the size of the larger nut. 

Poke the three LEDs through the disc as shown above, and bend the leads outward to hold them in place. Check which leads are the anodes and which ones are the cathodes. Solder the leads to the perfboard, and twist the three cathodes (negative, shorter leads) together. Solder them to a wire and wrap it in heat-shrink tubing. Trim the remaining three positive leads to about 5mm, and solder them to a 680Ω resistor each. Twist the ends of the three resistors together and solder them to another wire. Wrap it in some heat shrink tubing for insulation.

Step 5: Putting It All Together

Pass the two wires through a nut, and solder the positive wire to the switch. 

Step 6: Glue

To glue the nuts together I chose J-B Weld. It was recommended to me by several people and websites as an adhesive to glue metal to metal, and I was not very happy with the epoxy I used last time. Glue together the front part of the nut, up to the switch (don't do the Front nut yet). I held them together using wooden stirring rods and clamps. Let the glue dry for at least a night (about 15 hours).
Shorten the positive wire so it will fit inside the battery, then strip about 1-2cm of the end.
Then start the battery section. Glue on the remaining nuts one by one, superglueing the positive wire into the groove of each nut. Attach the stripped part to the last nut with something conductive, such as Bare Paint or silver solder (or regular solder if that works).
Leave everything to dry for another 15-25 hours.
Now attach the front nut, and clamp it down to dry. I used a vice to keep it in place, because the small clamps were too weak to keep it still.
Finally, glue in the switch cap. When you do this, make sure that you only apply enough for the area of the switch, or it may get stuck to the nut, making it impossible to use the switch.
As you can see above, I ended up using an A27 battery instead of an A23, because the wire was irregularly glued in, blocking the way. 

Step 7: Improvements

Even though it's much better compared to my previous attempt, there are still a couple of things that could be improved about this flashlight.
  • Silver solder (some of) the nuts: This would eliminate any problems with the wire and the battery compartment, as the wire that ran along the battery could just be soldered to the nut of the switch. It might also give the flashlight a cleaner finish than the J-B Weld, making it look more professional.
  • Use a drill press with a stainless steel nut for the switch nut: It was quite a disappointment when I realized that my cordless drill wouldn't be able to drill into the stainless steel nut. A stainless steel nut would have made the whole thing look much sleeker and more professional, as the whole flashlight would have been the same material.
  • Improve the switch: The switch ended up being pretty flimsy and irregular. This would be pretty easy to improve, if I did it again. The switch would simply have to be glued into the nut so that the top is perfectly in line with the hole above it. It could also be done by making the plastic top come through the hole, and then glueing on a shorter cap. The cap was also a bit too big, which was because I needed something that would be big enough for the hole. I should have used a much smaller drill for it.
  • Keychain/ strap: Even though it would be too heavy to carry around on a keychain, a keyring or a strap would make it much easier to use. With a drill press I could drill a hole into the bolt and add a keychain, but other than that I can't think of any possible way of doing it.
Do you have any suggestions, questions, or ideas? I'd love to get any feedback, or even to see your own if you decide to try to make your own.

Be the First to Share


    • Halloween Contest

      Halloween Contest
    • Crayons Challenge

      Crayons Challenge
    • Fandom Contest

      Fandom Contest



    6 years ago

    Wow! I'm going to try a mini one... Maybe even a laser pointer! Very great 'ible :)


    Reply 7 years ago on Introduction

    Let me know how it goes! And post pictures!


    8 years ago on Introduction

    Cool looking flashlight man. I love how more and more people are starting to to handmade flashlights.


    8 years ago on Introduction

    Very cool piece of work. The bolt-framework itself is a good idea, with the potential to be used on lots of other builds.


    9 years ago on Step 4

    You must connect leds as serial. Not parallel. Because parallel connecting consumes very high energy and this leads battery drains quickly. You can use 3 LED serial and only one resistor (I think approximate 330 ohms is suitable)


    Reply 9 years ago on Step 4

    I was thinking about this too as I was making it, and eventually decided on parallel rather than series. I knew that series wouldn't drain the battery as quickly and would only need little resistance, but I was afraid that one of the LEDs might eventually burn out or somehow stop working, breaking the circuit. The LEDs I used were pretty cheap, so I don't know how long they'll last.
    To be safe, I wired it in parallel. I'd rather buy a new battery every once in a while (they're only about $1.50, after all), than have another flashlight project that stopped working.


    Reply 9 years ago on Introduction

    Senoleker is right - well right about the series connection at least. LEDs are an active element and one can "steal" the available current leaving the others "wanting" more. Kirchoff's law will explain why you should wire them in series. However, this also means you have to allow for the voltage drop across each LED.

    And don't guess the value for R! It's simply the supply voltage minus forward voltage drop across the LED (typically around 2v, but check the data sheet) divided by the current the LED requires. For three LEDs in series, you just divide this by 3 and select the nearest preferred value.

    For a single LED - let's assume a 9V supply, 2v drop and 20mA forward current. So we have to drop 7v across the resistor - allowing 20mA through it which, by Ohm's law is:

    7/0.02 = 350 ohms - and the nearest preferred value is 330R.

    The ONLY reason you might want to use parallel LEDs is when you have a very low voltage source and you need a lot of LEDs - this is used in those 9-LED flashlights which are powered from 4.5V.

    As for an LED burning out - used properly, they'll last for 100,000 hours or so - so that's not very likely.


    Reply 9 years ago on Introduction

    Yes, I do now realize that serial would make much more sense, and save power. I know that LEDs have thousands of hours of life, but the ones I used were salvaged from an old device, so I had no way of knowing if they were good quality and how long they would last.


    Reply 9 years ago on Introduction

    Well, if they came out out an existing piece of equipment, it's a good bet they still had 10s of thousands of hours left. The question is really how hard can you drive them? The only way to know that for sure is to get the data sheet - and that's not going to be available.

    Cree (and some others) make LEDs with internal current regulators so you can chuck a fairly large voltage range at them and not worry about blowing them up. Some manufacturers are making Cree-like SMT LEDs that don't have the regulator with predictable results!

    If you do decide to remake this (and it's actually a pretty cool toy) you might want to source your LEDs directly - and give yourself the benefit of knowing what you're getting. High-brightness, low power... etc. The range is mind boggling.

    Ironically, I just stripped two of my older LED flashlights - and made them into high-grade microphones - but your 'ible would work quite well for that too.

    Check out the "business end" which is a Panasonic WM61a - but be quick if you want one as they're not longer manufactured and stocks are starting to run low.


    Reply 9 years ago on Introduction

    Thanks for the ideas, I'll look into those. I haven't done anything to do with Audio so far, and don't have any experience with it, but the microphone's a cool idea.

    I was also thinking about making a supercap project, maybe also with hardware, but so far I haven't figured out yet how I would do the USB cable to charge it.


    Reply 9 years ago on Introduction

    A mic is technically simpler than what you've already done - and with this strength and construction weight could, potentially rival commercial mics up to $100 and probably a lot more. Many are "phantom" powered which is a challenge, but for smaller projects where a high-grade mic is required - such as on a DSLR - a self-powered, or even unpowered, electrect is the way to go.

    My Canon 600D/Rebel supplies the power to the microphone module (they typically only need 2-10V to drive the internal FET, but you could easily mount a power source (and the required decoupling capacitor) direct into the shell.

    If you want a slightly more complex challenge, you can make it balanced and/or phantom powered. (Circuits are widely available.)

    What people don't seem to realise when buying lower end (and even some professional mics) is they're paying for the name and the housing. The active element is probably nowhere near as expensive as they imagine and much of the sound quality is down that alone.

    The only thing you need to be careful of is to make sure that the microphone element is unencumbered by hard, sound-reflective surfaces.

    A mic built like this has a very clean, cardioid or hypercardoid response which is ideal for vocals and music. It's even possible to do stereo with an XY configuration at the head... and so it goes on.

    I mention all this because you seem to have a keen interest in building strong, reliable kit - and that's what music and video people demand - so if you've got any friends with those interests you're likely to find yourself is some demand!


    Reply 9 years ago on Introduction

    Thanks for the info, I had no idea of all this. I only have a little tech experience, such as working with mics, some instruments, and speakers, but I never got into much detail.
    The microphone would be a little too heavy to hold for longer periods of time, though. I would probably have to drill out the nut as much as possible to give the components more space, and make the whole thing much lighter and smaller.
    Would I need to add some sort of mesh over the mic as well, or is that unnecessary? I could just buy one of those mic heads, so it would look nice and professional (and I could match the steel nuts with a steel head).


    Reply 9 years ago on Introduction

    When was the last time you saw anyone stood around with a microphone for very long? '-)

    A decent mic is generally pretty heavy *cheap* ones usually include some sort of internal ballast (i.e. a bit of concrete - yes really!) to give the impression that they're more substantial than the really are. The Singstar mics for the PS2 (which use an unpowered dynamic insert) have something similar. But the body is actually just a bit of plastic resin!

    Most professional mics (and certainly ALL of the old ribbon models) had to be held in a stand simply because the magnets (in the case of ribbon) are so ridiculously heavy. "Never mind the quality, feel the weight!"

    There's something to be said for a heavy-duty case since it will take rough handling with ease and the moment of inertia is such that it's less likely to pick up every last vibration in the room.

    The mesh is technically unnecessary - back in the day, even some speaker purists would swear you should take the front off your speakers to make them sound better. There is a modicum of truth in that, but it's more applicable to mics where the pressure waves are smaller.

    The trick is to get some acoustically transparent foam - it doesn't have to be expensive and it looks the bee's knees. I got three 20mm "heads" from eBay for under $2! And yes, they do add a certain je n'cest quoi? to the look.

    The foam head stops wind causing very low frequency noise chuntering all over your recording - in some cases you need to add a "dead cat" or a "zeppelin"; particularly if you're outside. Indoors, a foam head is fine.

    Most performers, because they're performers not studio techs, need a pop screen too. This is a simple mesh placed a few inches away from the mic to stop the inevitable plosive pops we all make.

    At the lead end you should (although you don't *need* to) add a plug so the mic can be connected to leads of differing lengths. A pro mic would typically use an XLR for this but you can use whatever suits you. I'm cheap (*cough*) so I don't even bother. ;-)

    Seriously though, if you can get a couple of WM61a (the "b" version has pins and is a LOT easier to solder) you'll be the doyen of your musical and video friends. I've built a couple of these mics based around aluminium torches and they sound fantastic.

    If you can't find the WM61a/b, just about any other electret condensor insert wil do the trick but be aware that the quality will not be as good. It's something of a rare diamond - as cheap as chips but with almost Rolls-Royce quality. So good in fact, that it's believed to have been used in at least one professional grade measurement mic - and those guys don't muck around!