Can sound waves turn a screw by vibration?

[930913]

Would it be possible to make a sonic screwdriver by using sound waves, to turn a screw?

The sound waves would make the screw vibrate. Is it possible to get the vibrations to turn a screw?

Would it be easier to have the right side of a screw one metal, and the left another? You could then have one virbration pushing one way and another frequency pushing the other way.

 

[Danger]

I can't see any way that this could be made to work. Sonic vibrations can certainly make a screw loosen up, but not in any sort of controlled fashion.

 

[Astronuc]

AFAIK, sonic waves have been used to loose screws by break the interfacial friction and corrosion products, but not rotate (unscrew) them. It might actually be possible with a particular polarization of the UT wave (basically a particular mode), but I am not sure that such a wave could be applied to enough of the thread. The UT wave would scatter of the threads so as to disrupt a favorable mode.

I have seen some rather remarkable applications of UT but those are proprietary.

 

[Colin1]

There'd probably also be commercial reasons why it wouldn't work ie I don't know what it would cost to mass-produce a sonic screwdriver (assuming it ever became possible to build one) but would unit cost ever be cheaper than for a normal screwdriver?

They'd be useful in high-voltage applications where I guess the rating would be considerably higher than existing electrical screwdrivers.

 

[brennanshaw]

It can be done!

technicaly it can be done! All you have to do is:

Get a tthermal insulation casing and fix an ajustment sensor to it. then biuld an anchionic chamber out of a very strong plastic or glass and then make sure you can put wires through the middle of it. now inside the sonic screwdriver you have to put theese items in the correct order starting from the bottom of the screwdriver going to the top:

the cooling cells then the secondary emmiter cluster then charching cells followed by charging cells, now add an accoustic accelorator to amplify the sound waves, followed by a bracing coil to withstand the soundwaves and protect against damage. now you MUST add the funcion drums to keep the device working. ( bacicly it a technical term for batteries ) then followed by the resonator cage to filter the sound waves. now thread the wires from the resonator cage through thr ancheonic chamber ( the core ) and connect it to your wave prism ( surrounded by micro stabiliser fields ) and thread those wires through to the centeral emitter channel, finaly connect that to the PRIMARY emitter cluster. now you have the small sonic disrupter you've always wanted.

so you see the sonic scredriver can be made with the right no how! just be careful as once the device is all connected and fitted within the thermal insulation casing, there are over 34,000 settings for the tiny device! ( NOTE: Not all the settings have been descoverd yet so please be careful! ) so there you go. if this has helped or intriuged you email me at thx. and don't take over the world with your device!

 

[Cult Of Scaro]

I need to talk with brennanshaw but he has not been on for 3 months. If anyone can contact him outside physics forums can they tell him to check his hotmail e-mails.

 

[wolram]

Undoing some fasteners is an art, one needs to know what method to use, Heat, shock, penetrating fluids, even then they may fail and the drill and tap has to be used, if a sonic screw driver could over come all these it would be worth its weight in gold, as long as it does not cost the equivalent.

 

[mgb_phys]

I always thought the amazing thing about the sonic screwdriver is that the Doctor can always find it! I have a small office and yet screwdrivers and Allen keys can hide very successfully.
The TARDIS is basically infinite in size (internally), plus he has the entire history of the universe to lose stuff - and yet can always lay his hands on the screwdriver!

 

[qbit]

I once assembled a small FM radio receiver. One of its parts is a Mini Coil Former and into it goes a threaded ferrite tuning slug. It was not a tight fit.
I was idly tapping the side of the Mini Coil Former (basically a plastic tube around which a copper coil is wound) while reading the instructions for assembly. To my surprise, I noticed that the slug was screwing up and out of the former against gravity! I experimented with tapping the former in different ways and found I could get the slug to screw in or out depending on the place and intensity of my tapping.
My thoughts on this at the time were, "perhaps a Sonic Screwdriver is possible."

 

[Danger]

My thoughts on this at the time were, "perhaps a Sonic Screwdriver is possible."

Cool. So, did you follow up on it? Maybe some experiments?

 

[qbit]

Cool. So, did you follow up on it? Maybe some experiments?

No I didn't follow it up, Danger. The mini coil former is now securely soldered into the circuit board.

Now that I've got some time off work, I would be interested to see if I can get say, a loosely fitting nut to move in a preferred direction along a bolt by vibrating the bolt with different devices. I think I should try two types of experiment. One where the device is in contact with the bolt at various points. And two, more relevant to this thread, where the device is not in contact with the bolt.

If anyone's interested I'm happy to post my method and results in this thread.

 

[Danger]

Please do. I don't expect that it will lead to anything marketable, or even tremendously practical, but it will be interesting and educational. Others might even have suggestions to improve your methods as you progress.

 

[qbit]

Here's my experiment that should be easy to reproduce. It did get one result I wasn't expecting. The reader might not be so surprised, however.

Materials used:

right handed wood screw (made of metal)
mass = 3.9 g
length = 54 mm
head diam = 8mm
thread diam = 4 mm
thread lead = 2 mm

metal washer
mass = 2.3 g
thickness = 1 mm
outer diam = 19 mm
inner diam = 7 mm

metal nut 1
mass 2.2 g
thickness = 3 mm
outer diam = 11 mm
inner diam = 4 mm

metal nut 2
mass 3.2 g
thickness = 5 mm
outer diam = 11 mm
inner diam = 4 mm

flanged metal nut 3
mass 6.9 g
thickness = 8 mm
outer diam = 18 mm
inner diam = 6 mm

flanged metal nut 4
mass 11.3 g
thickness = 10 mm
outer diam = 21 mm
inner diam = 6 mm

hexagonally splined drill bit
variable speed drill
vice
2 x small pieces of wood
spirit level

Notes:
In all cases the inside diameter of the nuts and washer were much greater than the diameter of the thread on the screw. This set up meant that the nuts and washer appear to 'hang' from the screw, rather than be threaded onto screw.
The female thread lead of the nuts wasn't measured but appeared to be smaller and the thread angle to the axis, greater.

Method and discussion:
The head of wood screw was clamped between the two small pieces of wood which in turn, was clamped by the vice. The screw protruded horizontally from the vice and this was checked as best as possible with the spirit level.
The washer was used for the first trial and was placed about half way along the wood screw. The drill was set to its highest speed and the drill bit applied parallel to the length of the screw and close to its end such that splines on the drill bit impacted the thread.
Interestingly, the washer began to slowly rotate (counterclockwise) in the opposite direction to the drill (clockwise). It did not move up or down the screw. Note that the thickness of the washer is less than the thread lead and the washer remained in the groove of the thread where the washer and screw were in contact. The drill was then set in the opposite direction (counterclockwise) and the washer began to slowly rotate in the opposite direction (clockwise). It did not move up or down the screw.
Different 'sides' of the screw were tried: left, right, top and bottom. All had the same effect.
The drill bit was then applied perpendicular to the screw such that the splines impacted the point of the screw. The washer rapidly bounced toward the head of the screw and it was impossible to tell if it was rotating.
The nut numbered 1 was then tried with the bit parallel. All sides were tried and the results were the same as those for the washer except that the nut traveled toward the head, following the thread, when it rotated clockwise and toward the end where the bit was being applied when it rotated counterclockwise. If it traveled too close to the bit, the nut would 'jump' thread back toward the head.
The drill bit was then applied perpendicular to the screw and nut 1 rapidly bounced toward the head of the screw and it was impossible to tell if it was rotating.
The same results were obtained for nuts 2, 3 and 4 except that some rotation was observed when the drill bit was applied perpendicular; although it appeared erratic and no overall trend in one direction was observed.

Comments:
As for the nuts to moving up and down the thread, it wasn't a slow process. I suppose each nut would have done a full turn in about a second but I didn't measure this and the rate of turning appeared to vary.

 

[qbit]

Can I or should I post a picture with this?

 

[Danger]

Can I or should I post a picture with this?

Sure; that would be great. And I want to compliment you upon the professionalism of your report.

 

[qbit]

Compliments are always welcome, Danger.

In the image you can see nut number 1 with the drill bit aligned on the left side of the screw in the parallel position. Not exactly aligned: it's hard to take a photo with one hand and hold the drill in the other. The point of view is from above the rig.
Also visible is the vice, wood, washer and the other nuts. The level can be just made out in the background.
It would have been great to make a short movie clip of the nuts moving along the screw, but as far as I know they can not be uploaded to PF and I do not a have a rig to secure the camera or an assistant. Yes, it's lonely work.
The results of this experiment imply that with the right 'vibrations' it might be possible to unscrew a screw that say, has it head sheared off or is a one-way slotted screw. The perpendicular drill bit trials suggest that a screw would 'wriggle' its way out if not for the fact that a tight thread prevents 'jumping'. The parallel drill bit trials suggest a screw could turn its way out assuming the torque overcame frictional forces. The interesting bit is that, assuming there is enough of the screw protruding from the surface to apply a drill bit parallel, one would set the drill motor rotation anticlockwise, as if to screw a right-handed screw out. Intuitively, without doing the experiment, I would have thought to apply a parallel drill bit clockwise and attempt to engage the screw like a cog: forcing it to turn anticlockwise and therefore out.

 

[qbit]

I will try to find the time and construct a rig to test this 'unintuitive' hypothesis. I expect I'll have to use a well lubricated nut and bolt.

 

[qbit]

Well, there's a lot more going on than I can account for. And I really shouldn't be surprised.

I tested a bolt:
mass = 11.1 g
length = 30 mm
head diam = 12 mm
thread diam = 8 mm
thread lead = 1 mm

This bolt was designed to fit:
flanged metal nut 3 (numbered for previous experiment)
mass 6.9 g
thickness = 8 mm
outer diam = 18 mm
inner diam = 6 mm

As the reader will have observed, there are notable differences between the screw in the first experiment and the bolt in the second. These differences may account for the opposing observed differences in results. Or they may not. In my opinion, only further experiments will decide.
Apparently there in no strict definition that distinguishes a bolt from a screw. For myself, I've always thought of a screw as 'the mechanical pin that can be turned and tightened into place as a result of its thread with a screwdriver' and a bolt as 'the mechanical pin that can be turned and tightened into place as a result of its thread with a spanner or its analogue'. It's a personal, tacit definition, and I'll stick to it for the purposes of this exercise.

The bolt and nut are not a 'tight' fit which I would consider to be something like a screw in wood; but rather an 'close' fit. That is, the bolt and nut were machined for each other.
In this experiment, the nut was clamped between the two pieces of wood which in turn, was clamped in the vice. Also, the nut was clamped so that bolt was vertical. That is, clockwise rotation resulted in the bolt going down with gravity. Note that the head of the bolt is not the end of the bolt upon which the drill bit is acting on. It is hexagonal and so it is not possible to apply the drill bit and have the bolt rotate freely. No lubricant was used.
Perpendicular drill bit orientation resulted in bolt rotation but in a random way, with no overall discernible pattern.
Parallel drill bit orientation predominantly resulted in bolt rotation in the opposite direction. This is as one would expect: the drill bit was engaging the bolt as though it were a cog but not matching rotational speed like a true cog. However, the bolt would occasionally stop, and occasionally momentarily turn in the opposite direction.
Simply stopping I could explain by simply assuming that the bolt and nut frictional force at that time was greater than the resultant torque from the friction between the drill bit and bolt. But to turn the other way, even momentarily, means there is more than just a simple balance between frictional forces with a net result in torque.
I should add that I could not control these same direction of rotation events (SDORE). They were fleeting when the drill was rotating clockwise (so the bolt was going down with gravity) and were fleeting anticlockwise (so the bolt was going against gravity).
I tried different orientations and points of contact on the bolt. Some resulted in SDORE more frequently than others. Some resulted in no SDORE that I could see.

I could try a few more follow up experiments, but I strongly suspect I don't have the equipment, skill or knowledge to figure out what's really going on here.

 

[Danger]

Simply stopping I could explain by simply assuming that the bolt and nut frictional force at that time was greater than the resultant torque from the friction between the drill bit and bolt. But to turn the other way, even momentarily, means there is more than just a simple balance between frictional forces with a net result in torque.

I really don't know anything about this stuff, but I'm wondering if you might be getting a 'reflection' of the vibrations coming back through the bolt from the clamped end. Sort of like an interference pattern?

 

[qbit]

I really don't know anything about this stuff, but I'm wondering if you might be getting a 'reflection' of the vibrations coming back through the bolt from the clamped end. Sort of like an interference pattern?

In the second experiment, there is no clamped end. Only the nut is clamped. It's not a great photo but you might be able to see this set up in the image in the previous post.
I think I follow what you're saying though. The only way I can think of explaining the results from the first experiment are in terms of interfering waves.
You've made a good point. Perhaps I should retry the experiment with the bolt head abutting the nut. In experiment 2, I forgot to state that the nut was mostly half way on the bolt.
I think I should also try to get an oversized nut to move along the bolt used in experiment 2 in the same way nuts moved along the the screw in experiment 1. There are a large number of physical differences between the screw and the bolt used.
Do you have any ideas on any other experiments I could do?

How did the link on the word 'friction' appear in my previous post? I didn't put it there.

 

[Danger]

How did the link on the word 'friction' appear in my previous post? I didn't put it there.

The site software automatically links certain words to the dictionary.
I've got to go to my pool match now, but I'll get back to you later.

 

[qbit]

This just keeps getting weird.

True, my rig would not like pass NATA accreditation, but I can't understand why I can't replicate the 'same direction of rotation events' I described in a previous post with the bolt in what I consider to be a very close replica of experiment 2.

However, I tried the bolt from experiment 2 in a similar set up as experiment 1. That is, the bolt is horizontal and the drill bit applied parallel to the bolt.
Let's call this experiment 3:
I used:

Metal bolt
mass = 11.1 g
length = 30 mm
head diam = 12 mm
thread diam = 8 mm
thread lead = 1 mm

flanged metal nut 4
mass 11.3 g
thickness = 10 mm
outer diam = 21 mm
inner diam = 8 mm

Previously, I had mistakenly listed this nut as having an inner diameter of 6 mm. I am relieved that PF is not a journal :)
Anyway, parallel application of the bit resulted in the nut behaving just as I had described in experiment 1: rotating in the opposite direction to the drill bit and following the thread. Lovely to watch. Completely consistent with the first experiment. So, it was nice to know that despite the fact that two very different bolts (aka screws) were used, the same results were obtained. I probably should add that the bolt, (from experiment 2) appeared to me to be more sensitive to the drill bit. Backing off from the drill revs resulted in less erratic behaviour, less jumping and greater consistency.

Now for the weird bit. Let's call this experiment 4.
I put the close fitting nut used notably in experiment 2:

flanged metal nut 3
mass 6.9 g
thickness = 8 mm
outer diam = 18 mm
inner diam = 6 mm

on the same horizontal bolt as in experiment 3 (above).
It turns out that the direction and magnitude of rotation of this nut is highly sensitive to where I apply the drill bit to the bolt and despite the direction of rotation of the drill bit! In other words, I could get the nut to rotate in either direction depending on where I placed the drill bit on the bolt (at various points I've tried to indicate in the image. Note also, I've included the screw from experiment 1 for comparison). I could also get it to 'jiggle around' in a given position. I got the feeling I was playing a piano. It was a little frustrating that I could no longer predict how the system should work, but otherwise fun to play with.
Any ideas how to map the vibrations or at the very least determine what frequency range they're in? I fear that without that kind of data, it would not be possible to put together a model of what's going on. This is a whole new level, yeah?

I suppose it would make sense to do another experiment where the bolt is turned around in the vice so that it is the bolt head that the drill bit makes contact with. If a sonic screwdriver were ever possible, this is the end, I assume, it would have to deal with. Or would it preferentially deal with the material the bolt is embedded in? Perhaps waves traveling through both media need to be considered?

 

[smulvey]

This question is fairly un-related to the doctor Who TV show since he doesn't use the screw driver to un-screw things, in season 3 of the new series "The Evolution Of The Daleks" we see he uses it to losen bolts and screws not to turn them. P.S Am i a Dr. Who Finatic?

 

[craig55]

Hey guys been following this topic for a while now, some amazing results, yet been thinking I know we are looking at sonic waves here, but what about a powerful focused rotating electro magnet, wouldn't that theoretically be able to undo a screw ? and I would have thought something like that would be quite easy to build ?

Craig

 

[steven house]

Hi I am new so please tell me if I am out of line,
But has anyone considered if the effects change with alternating levels of heat, From say three lasers of deferent intensity's?,
Plus in a real Sonic Screwdriver I think a number twist control may be smarter like the old combination bike chain lock,
001 turn screw clockwise, 327 fix wire, and 498 turn to next TV channel,

 

[shockfish08]

Hey Brennshaw, I'm really interested in this whole Sonic Screwdriver concept and really want to make one of my own. You wouldn't happen to have some pictures of the components that you described in your post about building a Sonic Screwdriver?

 

[skeptic2]

Please do. I don't expect that it will lead to anything marketable, or even tremendously practical, but it will be interesting and educational. Others might even have suggestions to improve your methods as you progress.

Don't be so sure. I used to work in a factory that used electronic component feeders. The components were dumped in a bin that vibrated. Around the inside of the bin was a helical ramp and the parts vibrated themselves up the ramp to the device that inserted them in the boards. The ramp went upwards in a clockwise direction. My guess is that the bin was vibrated with a sharp hit in the counterclockwise direction and a much softer return in the clockwise direction.

A screwdriver could be made using a similar principle. To tighten a screw, a rotary impeller could be hit hard in the clockwise direction with a spring return. The impeller would ratchet the screwdriver blade clockwise and ratchet itself in the opposite direction on the rebound. With the correct ratio of masses between the impeller and blade the screwdriver may exhibit no net torque. Such a screwdriver might have a use in space.

 

[shockfish08]

So does anyone have any other ideas about how to construct a portable sonic device that could jiggle out screws or perhaps coax door locks open?

One idea I have is that the sonic wave vibrations may be more effective if the emitter of the sonic device is in contact with the bolt/door lock, as it might be able to transmit the impulse of the waves through the mechanical apparatus more directly.

 

[julieisawuwu]

Has anyone heard anything more from BrennanShaw? in any shape or form...and is what what stated possible?

one last question...are all the items he stated easy to get hold of?

Thanks!

 

[Doctorwhofan]

Is he bonkers? He can't even spell correctly! And what in the world are charching cells? Ok, can you give me an ingredient list to what brennanshaw just posted? English preferred please.
Thanks in advance!