Laser diode self-mixing: Range-finding and sub-micron vibration measurement--make money online

today on Applied Science I'm going to

show you how to take a laser diode and

without any other op tool components

make submicron measurements from a few

centimeters away but first nerd thunder

is a collection of youtubers that are

doing a little cross promotion in

December including Jerry Ellsworth who

actually inspired me to start making

youtube videos in 2011 so be sure and go

check out those channels in the

description okay we're gonna start out

with a simple setup we've got a laser

diode here emitting light this is the

pattern here without a lens on it and

we're drawing about close to 24

milliamps and I'm measuring the current

out of the monitor photo diode here

which we'll talk about a minute and so

if I just take a white card and pass it

under the laser beam check out the graph

here showing the current history over

time did you expect that that's actually

kind of a surprise right I mean what's

going on here there's no sensor other

than the laser diode itself and if we

add a lens here the effect works a

longer range even because the lens is

collecting more light coming back from

this spot and if I use a card with a

retro reflector on it now the signal is

pretty massive it's about 10x bigger

because the retroreflector is sending

all that light back in so what's going

on here I mentioned that the laser diode

has a monitor photo diode in it so let's

cut one of these laser diode packages

open and you can see what I'm talking

about these laser diode packages are

typically three lead devices and so if

you're thinking there's something more

going on than just a light emitter you'd

be right so I cut some of these open by

mounting them in a lathe in a collet and

then using a dremel to very carefully

grind away the outer shell until it fell

off and here you can see a whole bunch

of diodes that I've cut up from

different sources now you'll notice that

in some of these cases there's actually

two separate components within the

package there's actually the laser diode

itself and then there's also a monitor

photo diode sort of on the backs top of

the package and what happens here is

when the laser diode is on light is

emitted forwards and back

words from the laser diode junction and

the backwards emitted light hits this

monitor photo diode that's inside the

package and so the point of this is that

if you want to really carefully control

how much light is coming out of the

laser junction due to things like

manufacturing tolerances and temperature

changes you really want to know how much

light it's emitting not just how much

current it's consuming so the

manufacturers put these monitor photo

diodes in so that your circuit can

actually do closed loop control of how

much light is really coming out for a

better or worse modern photo diodes

typically don't have this monitor photo

diode in there any longer if you buy

like these cheap laser pointers off of

ebay which are actually a really great

deal for like 12 bucks you can get all

three of these if you cut open the laser

packages from inside these laser pens

typically the photodiode is not there

even though the packages still have

three leads probably because they're

using the old tooling it's just that

they they either ignore the third lead

or it's just connected together with the

other one in these cases the laser diode

can handle pretty high currents and so

as long as you drive it in a constant

current configuration and you aren't too

concerned with the exact amount of light

output you can ignore the manufacturing

tolerances and the temperature

variations in at least one case I also

found instead of a monitor photo diode

actually an entire integrated circuit

with photodiode in the back of the

package and so this one is actually

doing the light regulating inside the

package which is not going to work for

this experiment that I'm going to show

you today and here's a really grody one

this one is totally cost optimized

they've gotten rid of the metal package

altogether and just have the laser die

like right on a piece of PCB material so

in the old days like the 1990s when

these laser diodes were new it was

common to have a setup like this so this

is how the three leads are wired up and

in fact I found this batch of old laser

diodes at Jame Co that were on closeout

and actually bought the last of them

it's actually very difficult to find

laser diodes that have a monitor photo

diode and are cheap you know kind of

like they were in the old days like this

so anyway so in this first setup this is

how we have it configured so 24 million

amps coming in this is the emitter side

of it and this

the photodiodes side of it and the photo

diode is reverse-biased and so there's

about two volts at the at the third lead

of the package and then we're going to

put our current meter here so that the

reverse leakage current through this

photodiode is what we're measuring and

it's about 50 micro amps or 60 micro

amps when the thing is running so this

is already pretty neat I mean we've got

basically a on-off sensor but it also

functions as a distance sensor and so if

I slowly move the card in and out like

this we're actually getting an analog

interpretation of how far away the card

is and with a little bit of tweaking you

could actually make this into quite a

useable circuit so what's happening is

the light is coming out of the wood

laser diode reflecting off the card the

lens is focusing it back into the

package and we're picking up the signal

on this monitor photo diode now what's

cool here is that the laser diode light

is coherent meaning it's all in one

phase like one way of coming out of the

package so when we put a reflector down

here it's actually coherent light being

reflected back into the package and we

have the possibility of getting the

light to interfere with itself so in the

next setup I'm going to show you how to

measure this interference pattern which

will let us measure really small

movements far away from the laser diode

in this setup we're going to use the

oscilloscope to look at that micro amp

current signal coming from the monitor

photo diode so here you can see the

power supply that we're using for the

laser diode and this is about the 24

milliamps going in I'm actually using

this in constant voltage mode because I

very carefully characterized this

specific diode and I have a current

limit that currently it is in current

for a constant voltage mode we're not

using the multimeter anymore I actually

have a circuit over there that's

converting that micro amp current signal

into a voltage which we're displaying on

the oscilloscope so let me show you what

that circuit is okay here's the circuit

pretty basic this is known as a trans

impedance amplifier because it's going

to convert the current signal from the

photodiode into a voltage signal that we

can look at on the oscilloscope and so

the way this works is if there's current

flowing in or out of the op-amp leg here

the op-amp will servo the output

to try to balance this out since we have

the positive or the non-inverting input

grounded so if there's current flowing

out then the op-amp will servo current

in to keep this node at the same zero

voltage right and the feedback resistor

is a hundred K so this means that if we

have one micro amp flowing in or out

here there should be a hundred

millivolts here going through this

resistor to compensate for that and the

bandwidth here is going to be a few

hundred K maybe even a couple of

megahertz but it's not super critical

and there's a little feedback

capacitance here to clean the signal up

ideally we want to reduce capacitance in

these cases because we want high

bandwidth but for what we're doing today

just to make it clearer to see on the

oscilloscope I threw in a little

capacitance there and when doing

circuits like this it's very convenient

to use two 9-volt batteries like this

back-to-back so that we get a very clean

voltage that's also floating relative to

everything else and I've built it up on

this board here with a couple of sockets

for the resistor and capacitor so that I

can quickly swap in and out values this

whole thing is DC coupled but then I

have the scope set to AC coupling so

that when we look at the signal we're

really looking at only the AC component

of current coming out of here and the

plus nine volts gives us a lot of

Headroom so that 50 or 60 micro amp of

DC that we don't really care about it's

going to be ignored and then if there's

a little bit of ripple current on here

which is the signal we're interested in

that will end up making its way all the

way through to the oscilloscope here's

the signal on the oscilloscope and the

full scale from top to bottom here is

about 200 millivolts so it's kind of

like plus/minus 1 micro amp of current

and the drive current for the laser

diode is holding steady at about 23 and

a half millions it's warmed up and

pretty stable and then here we've got

the card with the retroreflector tag on

it

and if I move this around a little bit

you can see the signal kind of goes nuts

and it's still moving when I'm not

touching the card however as you can

imagine this is pretty sensitive if I

basically just tap the card a little bit

we can see there's a ton of signal in

there but it's a little bit hard to pick

out exactly what's going on so to make

this easier to see what I'm going to do

is put a tiny little speaker into the

path and the speaker has a

retro-reflective sticker on it zoom in

so you can see this easier in a second

and then we're gonna drive the speaker

with this function generator here so

currently we've got 60 Hertz at 300

millivolts peak peak so I'll turn the

speaker on and the yellow trace on the

oscilloscope is going to show us the

drive signal going into the speaker and

then the current signal from the monitor

photodiodes being displayed up here

let's zoom in on the signal a little bit

you can see there's something

interesting happening here if the

speaker is responding perfectly to the

sine wave which I think it is at 60

Hertz this is why we're going very slow

you can see that when the speaker is in

motion we're getting a signal from the

monitor photo diode and then when the

speaker comes to a stop because it's

gotten all the way to the top or the

bottom the signal changes and then it's

back to high speed when it's moving

again so let's zoom in on the speaker so

you can see a closer look at it and try

to figure out what's going on here

here's the setup

and we've got the same laser diode with

the lens below it and the lens is

focusing the spot on to that retro

reflective sticker that's on the speaker

and this speaker you can't hear anything

because it's only 60 Hertz and the

frequency response of this speaker at 60

Hertz is not very high and also it's

we're just driving it at very low

amplitude just 300 millivolts peak peak

and we saw that that interesting looking

signal on the oscilloscope indicates

that the diode is able to detect the

frequency of the speaker moving even

from this far away and we can't hear

anything so it's a very sensitive

detector so here's what's going on here

the laser is emitting coherent light

right and it comes out of the laser and

strikes something and comes back

maintaining some amount of that

coherence so that if we have a distant

object the exact position of that object

relative to the waves of light coming

out is important if it happens to be

that the object crosses the laser beam

at a high point in the waveform then the

wave comes back and has constructive

interference here and we end up with a

can extra light signal hitting that

monitor photo diode in the back of the

package but if we move the object

one-half wave one

toward or away from us now the thing is

impacting at a lower part of the wave

form and when it reflects back we get

destructive interference back here and

the light experienced by that monitor

photo diode will be less so that if we

have a thing moving back and forth here

it's constantly crossing the waves of

light and every time it moves a half

wavelength we get a half cycle of output

on that monitor photo diode this is

basically an interferometer it's just

built with no extra components basically

the diode is doing the mixing for us

which is why it's called a self mixing

interferometer so if we look at the

waveform on the oscilloscope we can

actually say quite a lot about what the

speaker is doing I'll stop this just so

we can get a close look at it as the

speaker moves up and down in this

sinusoidal fashion we can actually count

the number of waveforms the number of

constructive and destructive

interference events that that photodiode

is witnessing and since we know the

wavelength of the laser light is about

650 nanometers every time we see one of

these Peaks that means another

wavelength has has gone by right so if

we count this up and let's just say it's

about 10 then we know that it's 10 times

650 nanometers or about six and a half

microns of movement that the speaker is

undergoing at this Drive strength so to

test our assertion let's try changing

the amplitude of the speaker so I'm

going to change the amplitude here and

I'm gonna go down so you can see that

the yellow trace now has less amplitude

and sure enough now we're only getting

about four peaks of constructive and

destructive interference between these

spots where it turns around so now I'll

head back up in amplitude now we're at

500 millivolts peak peak and I'll keep

going now we have a whole ton of them in

here and so if we stopped and actually

counted this carefully we could say at

this drive strength which is 800

millivolts peak peak there's way more

Peaks to count and hence the speaker is

moving further and that makes sense

because we're driving it at a higher

strength let's try a few other things

I'll go back down in amplitude and now

I'm going to give it a ramp waveform

let's take a look at this ramp waveform

so I've got the symmetry not quite 50

percent so it's kind of slower here and

then faster here so the speaker is gonna

have to move more quickly here to cover

that same amount of distance and you can

see if you squint at it yeah it looks

like there's sort of less waveform going

on here and a little bit of a denser

waveform here let me make the symmetry

even more lopsided so you can see now

definitely the wave Peaks are further

apart in here and they're closer

together in here but something

interesting happens if we keep going so

if I make this waveform really lopsided

it actually starts oscillating if the

speaker itself is now being rung like a

bell right so this this slope here is

sharp enough where it causes the speaker

to start wiggling on its own accord and

we're actually seeing that in the

interferogram here of whatever the

information from this interferometer we

can even see it in here too right

so as the thing is ringing we can

actually read that ringing signal with

this laser interferer interferometer if

I bring the signal back down closer to

symmetry we can see even before there's

ringing in the electrical signal we can

actually detect it with the

interferometer back to the sine wave

everything is nice and controlled and if

I change the frequency here the distance

that it covers so the number of Peaks

that we'll get is actually the same

because the amplitude is the same it's

just they get further apart right so the

actual number of interference peaks

between the top and bottom of the drive

signal should be the same they'll just

be spread out more in time so if we go

higher and higher it's still the same

Peaks now that they're getting

compressed together so the amplitude of

this signal actually doesn't tell us

much of anything it's really the number

of peaks and there's actually something

else that's interesting here that we can

check out if I change the speed back

down again and I'll stop the scope

notice that the peaks are actually not

symmetric so when the speaker is moving

toward the laser diode the shape of this

waveform is different than when it's

moving away from the laser diode this is

not an ax

and actually there's something very

interesting going on with the physics

here that allow us to infer the

directionality of the movement in

addition to the the movement itself

pretty cool I'll link to the paper that

covers all this because the physics is

slightly over my head I should point out

though that this is a very very

sensitive system so the exact drive

current going into the diode matters the

exact temperature matters the the type

of movements that it can measure are

also limit it has to be right in the

frequency band and we are using a

retroreflector to sort of increase the

amount of signal going back end

nonetheless there's quite a lot of cool

stuff going on also some of you might be

wondering why the pattern looks all

stair-steppy like this we'll talk about

that a little bit later at the end of

the video but partially it's because the

laser itself is responding to the

interferometer so instead of using a

monitor photo diode I've heard that it's

possible to actually use the laser diode

itself as the sensor so if you very very

carefully monitor the current going into

the laser diode and you're shooting

laser light back in there it will

actually affect the laser diode itself

so you in fact don't even necessarily

need the monitor photo diode you can

actually use the laser diode itself I've

tried this myself and didn't have a huge

amount of success with it but in the

academic literature apparently it is pop

possible if you have a high enough gain

amplifier let's look at a couple other

waveforms quickly with a square wave I

can actually hear the speaker now so up

until now all the waveforms that we've

been looking at to my ears have not been

audible and I'll lean in so you can

maybe hear this and the reason that we

can hear it is because the sharp

snapping voltage is actually oscillating

the speaker membrane itself and we can

actually see that with the trace just

like we did with the sawtooth waveform

the reason that it's audible is because

of this actually this response wiggle

it's not necessarily the snap itself and

we can see that it rings for a long time

and just about here it's finally almost

done ringing when it gets snapped again

so I can also switch to this kind of a

waveform that has a much

more gentle slope so I can't hear it

anymore because there's less ringing

going on and it's pretty much down to

almost no movement remember that one

interference peak is 650 nanometers so

the chance that this set up I've got

with the alligator clip holding a laser

diode and everything I it's reasonable

to assume that just air currents are

causing it to shift by 650 nano meter

this setup can also measure constant

velocity not just oscillating things

like speaker cones and so in this setup

I've got a motor here with the shaft

wrapped in retro reflective tape and

what I'm gonna do is use the laser beam

to just gray as the side of the motor

shaft so that as it's spinning that

piece of retro reflective tape will sort

of constantly be moving toward the laser

diode or away from it when we switch

directions and this should give us a

constant signal so let's check the

oscilloscope again and I'm going to use

a variable power supply so that I can

change the speed of the motor very

finely so let's check out the scope

again okay so now the signal looks

pretty hectic because the speed is

relatively high before we had that

speaker cone that was moving just you

know six microns in the span at sixty

Hertz or something which is a pretty low

linear velocity but now at the edge of

that motor shaft things are much faster

so the frequency that we would expect to

see here is much higher so we could zoom

in and look at this kind of wave by a

wave but there's an easier way to do it

we'll use an FFT on the scope to take a

look at where the where the most energy

is for this signal and it's clipping

here because we've got so much signal we

actually have to turn things down so I

will turn this down a little bit okay

a little bit more okay so now the FFT is

showing us where there's the most

frequency and don't worry about the

roll-off this is actually mostly due to

the amplifier that I'm using and just

noise in general so if I move this the

motor out of the path of the laser beam

we saw that peak just went away in here

and try to ignore all these little other

noise Peaks they aren't part of the

experiment

so if I put the motor back into place we

can see that this peak here corresponds

to the speed of the motor shaft and I'll

prove it to you by changing the speed of

the motor you can actually read off the

voltage here that we're feeding the

motor and so if I slowly turn this down

we can see that the peak here starts

sliding down the spectrum 10 kilohertz

now and we can almost actually see the

10 kilohertz here let's see if I can go

any slower so we're down to 500

millivolts at the motor and there is

kind of a peak here and you can actually

see the signal the raw signal waves here

let's speed up again you can see the

voltage is coming up the peak is here

I'm gonna keep going up in voltage three

and a half volts now the peak is up here

100 kilohertz if we keep going you can

still see it even though we're falling

off into here you can see death

definitely still there so now I'm going

to move the motor out of the way and you

can see that the peak is gone and if we

put it back in here there it is again

so it's actually a very effective

tackler you can aim this at anything

that's moving and not only determine its

vibration to within 600 nanometer

accuracy you can also determine its

speed very easily just by looking at the

spectrum of a signal coming back there

pretty cool ok so I think this is pretty

cool but there's actually another trick

that we can play wouldn't it be nice if

we could use this system to measure the

distance from here to here seems almost

impossible because this interferometry

approach requires some sort of movement

to actually get a signal right like no

matter where you are in the waveform if

this thing isn't moving there's no

signal we get so there's no way you can

measure an absolute distance with it but

there is a way can you think of it what

if I told you that changing the current

into the photodiode actually shifts the

color of the light very slightly it's

true and if we modulate the current

going into the photodiode it will

actually produce light that is very

slightly red or very

lightly blue depending if we're

injecting more or less current so what

we can do is feed the laser diode a

ramped current waveform so we're

constantly increasing the current for a

period of time and then resetting and

changing the current again and what will

happen is if we emit let's say a red

photon and it comes out here and bounces

off the object by the time the photon is

reflected and comes back in we're now

feeding the laser diode a different

current and hence it's operating at a

different frequency so when the photon

of a different color comes in it will

interfere with the laser diode which is

now operating at a different frequency

and we'll get a signal out of that right

because the whole reason that this

interferometry works is because the

wavelengths aren't lined up so if the

photon is old basically it went out and

came back the longer it's spent outside

the laser diode if our ramp keeps

changing the color of the light varies

slightly that means we'll have a bigger

shift and so when all this shapes out if

we give the laser diode a constant

current ramp the further away objects

are the more shifted they will be and

we'll get more interferometric signal

pretty cool now this is not an easy

thing to get going so I'm going to set

up the test setup here we're gonna use

the we're gonna use channel 2 on the

function generator to send a 520 Hertz

sine wave into the photodiode and it's

going to be set up kind of like this so

we've got most of the current coming

from the source measurement unit into

the laser diode and then I've got a

capacitor here coupling in the function

generator and it's like a 500 Hertz

signal at 30 millivolts so this little

bit of current ripple will go in and

modulate the laser diode and then we'll

use the same setup to look at it on the

scope for the physical setup what I'll

do is put the card with the retro

reflector on the table here and we'll

look at it on the scope and see what

kind of signal it's producing and then

to shift into a closer object I'm going

to slide this block in that also has a

retro reflective sticker so we can kind

of slide this in and out to get a

distance of about you know 2 centimeter

distance change so we'll see what that

looks like

okay this setup is a little bit finicky

but I think it proves the point shows

that this is possible the turquoise

trace shows the current injection from

the signal and generator so we're

running at about five hundred and twenty

Hertz and at the photodiode we're

getting a plus minus five millivolt bust

here and since the main amount of power

is coming from the source measurement

unit and this is trying to supply a

constant voltage and this is

superimposed on top we can actually

check the the current coming out of the

measurement unit and see that that's

actually sinusoidal which makes sense

because we're altering it here and this

thing is just trying to supply a

constant voltage so it notices the

current keeps changing which is correct

and then the turk or the pink trace is

coming from the monitor photo diode and

it's not surprising that that also has a

sinusoidal shape right because we're

making the photodiode

or making a laser diode brighter and

dimmer at this rate so you know

necessarily we're gonna see that change

but there's actually another signal

superimposed on top here notice that the

waveform is nice and jagged like this

and it's true that the signal is

sometimes hard to catch it's a little

bit hard to pick out but you can see

that it looks stair-steppy so instead of

being nice and smooth it has this really

chunky pattern to it and those chunks

are actually the interference waves that

I was talking about that only occur on

the slope so if we turn off the current

modulation

although stair-step enus goes away

there's still a little bit going on here

from just vibrations in the table but

when we turn on this additional waveform

we see this very regular pattern of

stair-stepping and so what I'll do is

make a copy of it will count the

Stairsteps there's about four steps on

each one of these up-and-down parts of

the wave and this shows a little bit of

a cleaner waveform that I captured just

a minute ago but it's about four

Stairsteps so we'll get that running and

now I'm going to slide in a much closer

pattern so we moved up about half the

distance toward the photodiode and look

what happened now there's only one stair

step between here and here for this

given slew rate and the way that the

whole thing is set up and how much the

frequency changes which is

depending on current which is a

characteristic at the laser diode now

we're only getting one wave of

interference as opposed to four waves of

interference for this whole setup so the

the technique is sound it requires quite

a bit of calibration and setup to use it

and the other downside is that if this

thing is running and I tap on the table

the table noise I mean just just tapping

the table like this produces a whole lot

more signal than this very sensitive

single stair-step kind of a deal so the

technique requires a fair bit of signal

massaging and sort of compensation to be

useful which is why typically this is

not used in fact in industry very much

but it is a pretty cool technique

nonetheless being able to extract you

know submicron vibration information

distance information and speed

information

all from one laser diode without really

any external components pretty cool okay

well I hope you enjoyed that and I will

see you next time byeTop Paid Keyword :  earn cash online, google make money from home, earn money online without investment by clicking ads, free earn money website, online money making jobs, earn money online without investment by typing, online work for money, best online earning sites, make money online with google, online earning websites, money making websites, online earning websites for students, invest online and earn money, best online money making, online money income, view ads and earn money without investment, earn money online by clicking, online money income site, money earning sites, online earning sites, best website to earn money, free money earning sites, money earning websites, get money online, online earning tips, online earning without investment,