Building a sonic screwdriver

[scienceroboticspunk]

I have searched the forum and could not find a thread similar. I feel like this thread belongs in hardware because it is building of something technology related, if not mods please move it. For the people that do not watch Doctor Who the sonic screwdriver is a tool The Doctor sometimes has with him. It has multiple functionality.

I want to build my own model that might be similar to the one in Doctor Who but with some of my own touches. Some ideas for features I want to build into it would be for it to be rechargable, bright LED flashlight, break a wine glass(frequency of 800-1500 hz, adjustable frequency), a laser pointer, a normal screwdriver out the tip, and to be handheld and light. If possible I'd like to build this for less than $100 USD(just the components). I have as much metal as I need.
I have limited ecectrical knowlege(soldering, wiring, resitors, LEDs,...simple stuff). I have access to a lathe, a mill, a drill press, anthing I could ever need for soldering, heat shrink/ heat shrink gun, and basic tools. I might be downplaying my electrical knowlege a bit, I am the caption for electricals on my high schools robotics team but I am no where near the skill of an electrical engineer so I'll keep with saying I have only limited electrical knowlege. I also have mechanical knowlege and full access to my schools metal shop.

I'll post my results when I'm finished but I would like some tips. I am unsure on if I should really have a brain for this or just control it with switches and knobs. I could use an arduino as a brain but I have little expierience with them and it looks like the mini is what I would need to use. If anyone else as some ideas for this or knows were to get started help is appreciated.

[Meteorswarm]

The LED requires at least a battery (rechargeable if you like, you should probably just go with an AA or something similar and use a rechargeable one if you care about that) and probably some resistors to make the voltage acceptable. The sound driver might be a little trickier.

[scienceroboticspunk]

The circuitry for the LED shouldnt be a problem. My big issues in the planning phase right now would be sound, and the battery. I am unsure what kind of sound I can even use to resonate a wine glass to make it break, if that could even fit in the screw driver idea. The powering of it with the rechargable battery might be an issue too though. I'd want to find a way to just build in the charger so I just need to plug it into the wall and so I dont have to take out the batterys.

Any other tools/gadgets that should belong besides the speaker, the light, the laser pointer, and a screw driver?

edit: I'd also like to get atleast 2 colours out of the lights, red and blue, normal flashlight would be a nice 3rd addition. Only one colour would really be needed for the laser pointer.
I'm gonna make a few drawings of this and talk to my digital electronics teacher tomorrow

[Link]

First off, let me tell you that if you want that kind of circuitry, you're probably going to be limited to the Eleventh Doctor's model, or the one the Doctor will end up giving to River Song. None of the older models are big enough to hold all that, lest you go with extremely expensive, tiny parts and multi-layer PCBs (or ASICs).

For both the glass-shattering sound and the laser, you're almost certainly going to need a higher voltage than any common rechargeable AA battery would give you. Unless you want to go with a special type of battery, that means you'll need on on-board boost converter. That means you can get a higher voltage, at the cost of more current. Supposing you want 8 volts at 350 mA to drive either the speaker or the laser, you're going to be pulling some 2.5 A from the battery, even with a high-efficiency switched-mode boost converter - around an hour of continuous use with a fully charged modern NiMH battery. Not much, but maybe just enough.

Now, the sound generation is a difficult one. I'm going to assume you want the classic whirring sound as well. If that is the case, you pretty much only have one option, and that is to use a microcontroller. You can store the whirring sound digitally; preferably in an uncompressed, directly usable format (i.e. PCM samples). If your sound sample is encoded as 16 bits, 22050 Hz PCM samples, and is at most a few seconds long, you can easily store it on a simple (read: cheap) SPI or I²C EEPROM or flash chip (1 Mbit is good for up to 2.8 seconds at the aforementioned sampling rate and resolution). I'd recommend a separate PCM IC, because it's nigh impossible to get any appreciable sound quality without one.

The glass-shattering sound is ideally a custom-generated sine wave. This is easy enough to generate with a microcontroller; use a lookup table with samples of a sine wave, and pick the next value from the table when a timer triggers; then pass that value to the external PCM. A potentiometer attached to the microcontroller's analogue-to-digital converter determines the frequency. Based on the value of that potentiometer, you can calculate the time you need to delay before picking the next value from the lookup table.

As for the laser, laser diode modules aren't too expensive (I'd recommend DealExtreme for them). You'll probably need a constant-current driver; they aren't difficult to build, and usually require only a single transistor, a high-power resistor, and a few diodes.

Finally, it's important to note that you will have to learn how to solder extremely tiny surface mount parts. There is simply no way you're going to get everything into such a small device with through-hole components, and having the board made by a professional shop would cost a fortune.

P.S. Arduinos aren't nearly small enough to do this nicely, unless you use a custom-built one, which is overkill in terms of its general-purpose capabilities, but still lacks the specific parts needed for this, such as the PCM and higher-voltage power supply.

[Velifer]

To break a wine glass, you're going to need something that creates around 120-150 db at the glass and sustains it until you can dial in the right resonant frequency. That's quite a bit of juice.

Some of the other functions from the list in the OP's first link:
uploading software - USB flash drives are down to just slightly larger than the size of the connector. or tuck a microSD in somewhere. If you're ambitious, use Xbees.

Hacking into computers -bootable distro on that usb!

scanning for heat signatures: IR photodiodes. Already have a speaker. Beep beep beep...

forcing a Star Whale to regurgitate by overloading its chemo-receptors. - ampule of skunk musk. A little goes a long way. Collecting it is the problem.

And while an Arduino is too large, the chip and crystal are quite small and could fit nicely. Get it working then get it off the board. Power is the real problem. Supercaps? Ultracaps? SUPERULTRAMEGACAPS!

[alexriehl]

Built 2 and I'll buy one. I've been looking for something similar for a long time. Or, better yet, post the plans when you're done.

[scienceroboticspunk]

I will give plans and a full construction list of everything needed along with pictures once I am finished. It might me a bit longer than a few weeks though because it is no longer necessary in a few weeks. I now plan on getting the girl a store bought one and will just build one for my own complicated and personal use at myleisure and as I can get the material and expenses for something complex but handheld.

[Moose Hole]

Put it on instructables.com. Lots of people will build it and improve upon your design if you do.

[GenericAnimeBoy]

Your idea is much more ambitious than Thinkgeek's Version, which is actually just a gussied up normal screwdriver with some LEDs.

If you want more info about breaking wine glasses, there's a Mythbusters episode that has a ton of useful information. That's going to be extraordinarily difficult with a device that small, as you're going to need a very small, very potent driver, and I can't think of anything on the scale you'd need other than maybe Cell Phone speakers (which are still not particularly small and woefully underpowered). It will probably need to be coupled with a waveguide of some kind for the same reason they had to put a board with a hole in front of their speaker in the episode--near field directionality is important for efficiently inducing sympathetic vibrations. Finding or (more likely) fabricating a speaker and waveguide optimized to the appropriate frequencies would probably be crucial. It'll be pretty cool if you manage it, though.

[Cue]

Okay, well, in regards to an Arduino, after checking with my highly accurate reference data (I have the official replica plastic screwdriver) I can confirm that an ArduinoNano would just about fit inside one, so there's your starting point. Do note that the screwdriver in question is Smith's, which is the only one you can really feasibly do this with.

ThinkGeek's screwdriver is poor, I have it too and it's not even an electric screwdriver (Seriously, making it electric), and it weighs a metric tonne.

I'd recommend getting an induction charger if you want to be clever, then just grab the appropriate battery to go along with it.

Now, if you wanted to be extra clever, add a couple of extra buttons and turn this into a universal remote as well, with the main button doing the sending and the other two doing the selection (have it do different sound and light patterns for different binds or something, I don't know).

[Moose Hole]

Now, if you wanted to be extra clever, add a couple of extra buttons and turn this into a universal remote as well

Just don't confuse the "Change Channel" and "Break Glass" buttons.

[hintss]

for the laser, you'll need a driver circuit (I think). and you'd think you could generate the sound using a 555 somehow. and for a speaker, I'm thinking a sounder from the alarm things on the demo units at electronics stores might work (I think its a really loud piezo, maybe.

[bates.and]

I am currently in the process of designing and building my own also. Going for the 10th Doctor's version though, I think the 11th's is too weird. Either way I know how to machine and have access to milling and turning machines so I have been building it up in solidworks and basing my measurements off of the toy version. The main issue I have encountered is building in the ability to produce the "sonic" sound. I tried to extra the circuitry from the toy but after I got it out it no longer worked and the board doesn't have any markings or info on it to help me determine what's going on. If anyone has any thoughts on how to build in the sound I would love to hear. I figured going with a small buzzer would be my best bet and have done some experiments with getting that sound out of one using an arduino. With that I've gotten mildly close but there is no way I could ever fit an arduino in. So anyone have any ideas?

[Neilo]

or you can go with a real type bu you would have to mail me one

[chris661]

Hmmmm...

Unless used at very short range, I don't think you're gonna shatter a wine glass.

I saw a band (think it was Status Quo) star on a TV program where they tried using some of their guitar amps to break the glass.

A 12" guitar speaker (often ~ 100dB @1w) with a Marshall head (think it was 50w, may have been more) failed. Calculated SPL would be (taking 50w and 100dB@1w@1m) 117dB.

Now, IIRC, 50% efficiency is 108dB. This is acheived by compression drivers feeding into horns (acoustic transformer). Taking an optimistic value of 100dB @1w for something smaller, you'd need 50w again to get to the same volume as the above guitar amplifier. 50w out of a single AA battery won't happen for any amount of time, and I assume you know what happens when you ask such high currents of a battery. This power still won't break the glass, and very few tweeters of sufficiently small size can stand 50w.

Sorry to rain on your bonfire, but I don't actually think its gonna be possible.

Chris

[GenericAnimeBoy]

Hmmmm...

Unless used at very short range, I don't think you're gonna shatter a wine glass.

I saw a band (think it was Status Quo) star on a TV program where they tried using some of their guitar amps to break the glass.

A 12" guitar speaker (often ~ 100dB @1w) with a Marshall head (think it was 50w, may have been more) failed. Calculated SPL would be (taking 50w and 100dB@1w@1m) 117dB.

Now, IIRC, 50% efficiency is 108dB. This is acheived by compression drivers feeding into horns (acoustic transformer). Taking an optimistic value of 100dB @1w for something smaller, you'd need 50w again to get to the same volume as the above guitar amplifier. 50w out of a single AA battery won't happen for any amount of time, and I assume you know what happens when you ask such high currents of a battery. This power still won't break the glass, and very few tweeters of sufficiently small size can stand 50w.

Sorry to rain on your bonfire, but I don't actually think its gonna be possible.

Chris

To be fair to the idea, just setting the glass in front of a 12" cone doesn't work very well because the near field directionality of the speaker cone doesn't work well for transferring energy into the glass's resonant mode, or some such. (Start watching at 1:03) Don't get me wrong...I still don't give it much hope of success, but power isn't the only variable in the equation.

[chris661]

The board with the hole helped because E=1/2 mv^2. Make the radiating area smaller (but keep the volume displacement of the speaker the same), and velocity increases, and so does the energy transferred.

The uncovered speaker would (assuming reasonably pistonic behaviour at ~500Hz, no bell-mode breakup etc) be fairly omnidirectional, so much of the energy may well be wasted. Wait, that's not entirely true: the polar plot of a 12" speaker would show diminished output toward the extremes of angles as you end up with a distance between each edge of the cone (near and far) significant compared to the half wavelength of the sound, so you'd get a little cancellation. That assumes, again, pistonic behaviour of the speaker cone - something that's rare in guitar speakers (referring back to Status Quo in my previous reply).

So...

They used a 12" at close range with a 2" actual radiating area. That would increase the velocity of the air at the glass by 36x (ratio of 6:1 in radius, then square it for area), and as E is proportional to v^2, that's a massive energy gain.

They also put the glass far closer to the speaker: in free-field, double distance = 6dB loss (quarter power). If that glass was about 10cm from the exit, that should be a 20dB increase (100x power) over the same speaker 1m away.


In short, getting very close to the glass will make it easier, but for the whole process of dialing in the right frequency (with a resolution <1Hz in 500) and then playing it back is something that's just not possible in something portable powered by an AA battery.

Chris