College Project

How about an optical encoder wheel attached to the screw driver.

‎Gurley Precision Instruments

Not sure this is exactly what you're looking for?

Or, maybe

Sensors for displacement, distance & position | Micro-Epsilon ...

Rotary position sensors - High speed operation to 60.000 rpm - rls.si‎

Or, low tech you could use a dial indicator linked to the screw driver.

The sensor is for finding the disaplacement of the screw not the screwdriver so a optical encoder on the srew driver wont measure what i need.

But i went through you links: In the micro-epsilion products I feel like the three best options that are the eddy current sensors and the capacitive displacement sensor which could be used a the a standard screw is obviously is conductive also the laser triangulation sensor I was thinking the way that it could be done is embed the end of screw in a metallic plate as have the sensor record that what do you think?

For the Rotary position sensor how could i set it up (measure the rotation/turns by the screw) but allow for the linear displacement I have a couple of ideas but not concrete I would like to know what you think?

Im doing some designs on the sensors you given for the expirment with the information so thanks again

It's time for clarification for my simplistic mind.

Are you measuring the turns of the screw (rotations), or the downward travel, (displacement/linear movement), or both?

Displacement/linear movement is governed by the thread pitch, as Rixter explained so I would think one would indicate the other.

So, just for clarity, are you trying to measure the amount of rotation provided by your new ratchet driver? If so, no screw is required and the rotary encoder does not have to move downward at all.

Ultrasonic distance measurer mounted on the screwdriver handle?

..." robots use MEMs gyroscopes, which are sensors specifically designed to sense rotation.

When a gyroscope sensor is rotated about its axis, a very small micromachined mass is moved by the Coriolis Effect to the outside of the sensor.

Figure 2 : Internal operation a MEMs gyroscope sensor (source: Analog Devices)

The ADI iSensor MEMS Gyroscope Subsystems are designed to reliably detect and accurately measure the angular rate of rotation of an object. iSensor gyroscopes are rugged enough to detect rotation under the harsh environment of a robot under severe stress and in complex industrial conditions. Gyroscopes are not one-size-fits-all sensors, and it's important to select the correct gyro for a given rotation. Two important specifications are range and sensitivity. The range is the fastest rotational speed that the gyro can accurately measure and is measured in degrees per second (°/sec). The sensitivity is the output change in millivolts for a given speed - so the faster the gyro is spinning, the higher the voltage. It is measured in millivolts per degree per second (mV/°/sec).

A rapidly rotating tool would require a high range, like the Analog Devices ADIS16266BCCZ which can measure up to ±14,000°/sec. A slowly rotating arm could be served by an ADI ADIS16060BCCZ which has a range of only ±80°/sec. Analog gyroscopes operating at these low voltages on an industrial robot require low losses for the interconnect. The Mizu-P25 Miniature Waterproof Connectors from Molex is a miniature IP67 sealed connector system which is dustproof and waterproof. It is also suitable for high vibration environments and with a contact resistance of only 10mΩ is suitable for harsh environments low voltage systems."...

https://www.mouser.com/applications/robotics-position-sensors/

https://www5.epsondevice.com/en/information/technical_info/gyro/

I like this idea and I look it up and got the basics of how it works. It seems a effective sensor for such a small size which is useful for this experiment but how would i set this up correctly to detect the angluar velocity of the screw? As I have a rough idea but it feels messy on wires, any thoughts?

Thanks

How much time do you have to finish this project?

There is a whole lot to be said for keeping things simple. This is true in general but especially when you're on a tight schedule, and you are on a tight schedule.

Do you really want to still be debugging this project come Finals Week?

What of the varying angle of the ratchet with respect to vertical as the user operates the tool? These things are also measuring gravity and other sources of acceleration, so do we track the X, Y and Z components, compute the net vector and subtract out its influence in real time, when all we really need to know is how far a screw moved? Truly.

i see your what your saying about your previous point about time this got three months come janurary its for one of these engineering competions work in your own time kind of thing so can be dropped but would like to work on it. I also get your point about this too but just because these sensors can measure all things doesnt mean we have to use them in the calculations but when space is a concern the how small the MEM gyroscope are is a big plus

"I also get your point about this too but just because these sensors can measure all things doesnt mean we have to use them in the calculations"

That is not my point at all. What I am saying that your gyro is already measuring them and you have to sort them all out and distil all that down into single number representing screw distance. Thing is, other sensors can do this directly. Ones that have been mentioned here. How can you be sure none of them are suitable? How do you know they won't fit? Which ones did you look at and reject on the basis that they were too large? Remember, your objective is to measure screw distance, not build an inertial navigation system.

I mean, you can't have looked at your alternatives in any depth given your post is scarcely a day old. Selecting the right sensor for your application will greatly influence how much time and effort you will need to invest downstream to make it work. Right now is when to take the time to make the right choice.

If you're given eight hours to chop down a tree, spend six of them sharpening your axe.

If this is all for your individual effort in a competition, then how much are you going to credit (the CR4 group) for it's contributions?...

Here are some thoughts: You state that you want to measure linear movement of a screw turned by your screwdriver. It might be easier to measure rotation, and rotation and linear motion of the screw are related by the screw's pitch. For example a 1/4-20 screw moves 1/20 inch with 360 degrees rotation.

Rotation can be converted to an electrical signal in a number of ways. You might find a surplus 20 turn pot and couple that to the screw. Probably the best method is to mount an optical encoder to the screw.

https://www.solo-labs.com/rotary-encoders-understanding-practical-implementation/

http://www.allelectronics.com/item/enc-128/rotary-optical-encoder/1.html

I hope this is helpful.

Mount a multi-turn potentiometer on top? Take it from there....

are you attempting to measure bolt stretch?

A depth micrometer might be what you want. A linear potentiometer can convert that to an electrical signal.

Thanks for all your input Im gotting to sift through all the information provided and Ill reply to each one as i go through them

"electrical sensor that can record each linear movement of the screw as it is turn by the screwdriver."

You're needing to measure the linear motion directly, without requiring knowledge of the thread pitch?

A linear potentiometer would be good for this. Put a known voltage across the end terminals (one end grounded is easiest), the wiper's voltage will be proportional to the wiper's distance from the grounded end.

Now, if you go this route, please be aware that there are two main types of (mechanically) linear potentiometer: those whose wiper resistance (as measured from either end terminal) is directly proportional to wiper position, and those whose wiper resistance has a logarithmic relationship (also called an 'audio taper' potentiometer, whether rotary or linear). This latter type is typically used in audio applications and is quite common, so be sure you're using the direct type (also called 'linear,' linear in the electrical sense as well as mechanical). Linear linear pot vs logarithmic (or 'audio taper') linear pot.

Is this clear as mud?

From a couple of responses I feel that I have giving you's too vague of a concept to go on for the best solution for that I apologise, I will try to be as honest on the expierment as possible:

Im working with a fellow student on a machine for a project in college we are doing several experiments for it one which we lack full knowledge of is the said sensor experiment what the we need is a sensor that accurately measure a parameter of the screw relative to the turning of a ratchet srewdriver ( "clicks" that it makes) which ever you sensor you feel is best dont have a preference to angular movement of the screw turing type of sensor or the linear displacement of it being moved type of sensor which ever one you feel works best the only perference that we really need is it not to large so it can incase in housing inside the machine.

Now I'm completely confused. "Clicks" are only heard (and only happen) when the ratchet is NOT turning the screw, but returning from the full cycle stop position to the next tightening (loosening) cycle.

Again, the screw is not a part of the experiment, but rather how far the driving shaft turns in relation to the handle seems to be the needed answer.

And "clicks" are a result of the coarseness or fineness of the ratchet wheels cogs.

I'm lost.

Thats true but im refering to when you are using a rachet screwdriver and say turning it clockwise to tighten a screw into something (wood, metal etc) if you turn anticlockwise till you heard say ten "clicks" then turn it back clockwise that would drive the screw further into the material than say you when turned it anticlockwise for five "clicks"

We want to use the rachet clicks as the unit in the use of the machine so need a sensor to see how much the screw moves for each click the rachet is turned so the rachet clicks are a turn into SSI unit of measurement that can progress be a peice of software so the user use the clicks as a frame of reference for there input but the software of the machine can use a calculable SSI unit.

I'm getting worried here.

The clicks are dependent on the diameter of the wheel and the number of cogs,or the pitch.

You refer to a machine. Is it a machine or a hand tool?

The clicks are indicative of nothing if this is a hand tool.

Repeatability of a hand tool is a human factor, not a quantifiable by clicks of backward motion.

Just take off the cover of the ratchet wrench and count the number of tits, they will vary from wrench to wrench....but that can then be converted to degrees..

360° / 24 = 15° per click....You could mount 2 micro switches to create a signal with every click in either direction...not a very dependable design I think...

An ultrasonic sensor would measure the target surface where the head of the screw will stop, or a laser displacement sensor suitable for polished metal, the ultrasonic will be more widely applicable but may have problem with power and size for any such instrument installed as a piece of a screwdriver, extensive miniaturization probably needed, and power consumption optimization, lots of engineering to be enlisted.

The accuracy will depend on how true to perpendicular the operator holds the screwdriver, but you will not be limited to a certain screw pitch as you would for a rotary solution which could be easier to miniaturize... Maybe standardize on a Torx bit to improve accuracy...

Machine and clicks? My impression of a screw inserting machine is that it keeps turning until the screw is tight--thus no clicks.

I believe he is trying to design a complex, combination lock with no visible numbers that gives electrical input to a lock release mechanism, by counting the "clicks" of some mechanical ratcheting device and sending that to the bolt release mechanism.

Lyn, I'm with you on being confused. Is he using the type of screwdriver that you push axially and a spiral inside it makes the screwdriver turn (I don't remember that they make clicks)? Or is he using a screwdriver attachment on a socket set ratchet handle (which does make clicks)? Also, he has spelling, capitalization, and English usage problems which don't seem typical of a college student. But I don't think we know where he is from, so English may be his second language.

How was the objective described to you? Exactly, in your professor's own words? Did he/she describe it on paper? If so, can you post an exact copy of that description here, word-for-word? Perhaps if we knew that it would be much clearer as to what you're actually trying to accomplish.

Maybe you could just tell us what the purpose of this experiment is...? I still don't have a clear picture of what you are trying to do....If we knew what the purpose is, we might be able to guess what you are trying to do.... For instance: is the purpose to measure how many turns it takes to tighten a screw? Is it to be able to program a certain amount of turns of the screwdriver? Is it to measure force required or used, to tighten the screw? Is it just to count the number of turns the screwdriver has made? Is it to measure the speed of the screwdriver turning? In other words what are the exact requirements in simple language....

Im sorry its clear I went the wrong route at the start of this thread thinking it be alot simpler if just gave a vauge concept of what was need to be done my bad

Ok I wont try bog you's down on details that you's dont need to know but long story short got a machine that for the controls of one of the parts it will behind a door panel that has to be locked rather than put a padlock on it or combonation lock on the door which are fine we wanted to try something different as idea to possibly make it more secure hence the experiment.

So instead of a traditional lock there would be a set of screws in the panel to open the door you loosen the screws using the "clicks" as the code so say for this example sake the screw moved 0.5mm out of the panel with every "click" of the rachet screwdriver then say the code for this screw style lock is 453578 the process would be the following steps:

1. The user would turn the screwdriver to the right until you heard four "clicks" each click representing a whole number to them starting from one going to a indeterminate number just now.

2. The user would turn the screwdriver to the left back to a start notch on the panel and lossening the screw.

3. The sensor would measure that it has now moved 2mm so it send this message to a computer

4. The coumputer has a software program on it that records this movement of 2mm as a input of 4 into the security code.

5. A button on the panel is pressed to register to the computer that you want to input a new number.

6. Repeat steps 1-5 for the rest of the numbers ie turn it for five clicks, input, three clicks, input......

7. Once the code has been enter the computer will send a signal to release the bolt; unlocking the panel.

What my query is what would be the best sensor to use given its task and that it housed inside the machine so space is a factor.

Is this any better or am I still loosing you's?

OK, but this seems overly complex.

Look at rotary switches:Rotary Switches, 6-position, 4-pole from StewMac. Golden Age Pickups + Electronics

Rotary Switches - Main Product Image

Maybe something like this...?

http://www.instructables.com/id/The-electronic-digital-combination-lock-SAFE-kit-w/

http://www.ebay.com/sch/electronix2384/m.html?item=170725534347&hash=item27c008ea8b&pt=LH_DefaultDomain_0&rt=nc&_trksid=p2047675.l2562

What we have, is failure to communicate.

We are back to small optical encoders for this, if all you need is a signal output to indicate degrees of rotation.

There are also digital output encoders, in the form of a potentiometer -

Yes, though many of these meant as front-panel controls have detents and are fairly low resolution (24 pulses per revolution, for example, is quite common), but there are hi-rez ones of comparable size. This one for instance, and it's cheap (40 bucks):

http://www.robotshop.com/en/6mm-rotary-encoder-1024-p-r.html?gclid=Cj0KEQiAhNnCBRCqkP6bvOjz_IwBEiQAMn_TMem8ayrs3FaT6pguQHVIESFwQHpjbK1ZI2v9SvDCKt8aAuf08P8HAQ

We can have some fun with this. Since it's a screwdriver, it is not likely to be driving a machine screw of coarser than 0.05 inch pitch (1/20). Per each step of a 1024-step encoder, that would advance the screw by a bit less than 0.00005 inch. Surely the OP does not need such precision, though maybe a few lessons in basic mechanical devices and arithmetic.

Well, if he needs the rez, there it is. If not, he can discard however many counter LSBs as needed to make it coarser, if that's what he wants. The thing's cheap enough. And besides, this is just an example of what's out there. Other resolutions are available; coarser and finer. Whatever he chooses is his call. What do you think the rez of this one is, the one pictured here?

Looks pretty high-end to me, but I'm not all that familiar with that hardware.

• up to 5,000 cycles per rev.

And those are their 'coarse' ones, and when you decode the quadrature it's 4X that.

Look at the $5.00 digital caliper to see how they have detected movement.

http://www.ebay.com/sch/i.html?_from=R40&_sacat=0&_nkw=digital+caliper&rt=nc&LH_BIN=1

Have you looked at LVDT (Linear Variable Differential Transformers) sensors?

Not cheap but will do what you ask - measure linear displacement.

Explanation 1

Explanation 2

ixthus.co.uk

No i havent, went through you links they seem a bit expensive but its a option thanks

I'm looking at this for a prototype:

Any thoughts?

Seems a bit too simple...have you thought of using a screw as a lock in the middle line far right so that the blue ball doesnt fall through the box?

Maybe. I'll try that. I'm only running it at 10% of the design speed so far.

So you are reverse-engineering a Cobalt screwdriver, but you want to know if you can make it more "click" economical, and also decode some panel door combination. By turning a set number of clicks on each screw. I say just get out the Black & Decker cordless screw driver, and have at it.

Why not put a small microphone near the screwdriver, and send the signal to a counter, after you "massage" the signal into a pulse train of constant amplitude? You should use a micrometer with a digital readout, and USB interface to record the linear motion of the screw into/out of the work piece.

Now, that's really Rube Goldberg!

Lol!

Easy one, use sound from the click of the ratchet as cited by Solar eagle. Ineterphase it with Adruino

Though, an intensive signal processing might be involved. But it's doable.

no. of click is proportional to angular displacement.