School project, laser diodes

Hey there

In school i am working on a project where i need to measure a box. I am currently not sure how to do it but have come up with this ida.

To put a row of laser diodes pointing into a row of phototransistors, and then measure the lenght of the box determined on how many of the lasers that have been "cut of"

Though i have had some trouble finding some laser diodes with a low divergense, only down to 7.5 degree, (to be honest i don't even know if this is high or low).

Is this possible, and do you guys have any recomendations on what laserdiodes to use.

Hi,
how "big" is this box & do you measure with reflected or through light?
K.

I was thinking of measure with trough light

The box length would vary from 10 - 30 cm

and the distance between the photo transistor and the laser diode would be aprox 18 cm

You can select special lasers with a so called "line optic"; means your laser dont has a spot, but more a line=exactly forobject lighting & detection with only one element & the otherwise I would use a position sensitive Photo resistor or diode...
Are these boxes moving or all stationary?
K.

Added after 34 minutes:

These company is one of the best (+ cheap)European address for VCSEL/Laser & detector elements, accessories: www.roithner-laser.com/Optics-LD.htm
K.

Added after 10 minutes:

These company is one of the best (+ cheap)European address for VCSEL/Laser & detector elements, accessories: www.roithner-laser.com/Optics-LD.htm
Another good formation: Your reliable partner for opto-electronic components: www.frlaserco.com/
Component deliverers:
Thorlabs - Your Source for Fiber Optics, Laser Diodes, Optica: www.thorlabs.de/
Laser Diode Modules - Edmund Optics, www.edmundoptics.com/onlinecatalog/Browse.cfm?categoryid=239
Laser line generators,www.photonic-products.com/products/optics/ppl_line_generators.html
Fiber Optic Laser Diode Manufacturers, www.medibix.com/CompanySearch.jsp?cs_choice=c&clt_choice=t&treepath=16174&stype=i
New Line of Micro Laser Diode Collimators, by BFi OPTiLAS .. ://optics.org/cws/product/P000013433
Line-cross generating laser diode module Laser Components: www.lasercomponents.com/de/
The laser line object detection method in an anti-collision system for powered wheelchairs : ://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F10041%2F32216%2F01501078.pdf%3Farnumber%3D1501078&authDecision=-203
To end, in your country too: www.laser2000.com/
K.

Horsevad said:

Hey there


Though i have had some trouble finding some laser diodes with a low divergense, only down to 7.5 degree, (to be honest i don't even know if this is high or low).

Click to expand...

Laser diodes have wide beam angles. They need a lens to create a spot.

Keith.

Laser diodes (and lenses) are expensive and you need lots of them to use the 'cut off' method. You need one laser, lens and photo transistor for each unit of length you want to measure. For example to measure 30cm with 1mm resolution you would need 300 of each device.

A more practical way of doing the job would be to use one laser (or an ordinary LED with lens or slit filter) and one photo transistor and move them so the box passes between them. It would be easier to measure how far you moved while one light beam was interrupted. I have in mind one of two methods, first to pull the LED/sensor across the box area at a known speed and time how long the beam is interrupted. The second method is similar but keep the sensors stationary and move the box at a fixed speed. This method is less restrictive on box length and lends itself to measuring more than one box on a conveyor belt. You also don't have to move the sensors back to the start position for every measurement.

Brian.

Hi,
Im thinking, goal of lens usage is some other, namely: collimation...
A laser has in both axes very different light width, as elipsoid, & as horsevad wrote too; it has some degre divergence (not a wide beam angles!) these both non circle + divergent specs are per optic to a round/circle spot & practically zero divergence to correct.
It makes collimation optic.
Otherwise if you can live wit 2-10mW; select a VCSEL, it has original practically no divergence & a round spot; means you dont need an optic to correct and collimate it, + their are more cheap too!

Why I linked Roithner`s(& others) optics is only because their are (for my idea) needed only as "line generator"...
But of course, line-laser isnt the only one solution to these detecting.
K.

Thanks for the responses, I will look into the line laser thing.

The box could be moving while being measured.

I actually also need a convinient way of measuring the with, and as the box is on a conveyor it is not really possible to do the "cut of" method there :S

Our teacher babbled about somthing called a linecamera, but it looks really fancy.

Hali,
If you must measure both axes, as one possibility, check for a cross line generator (lens) & measure the reflected light with position sensitive detector. Otherwise; two line cameras & moving object(or camera + light) is possible a better way...
Otherwise these system will need two lenses, (one is build onto your VSEL/Laser diode), the other before the x-y sensitive photo resistor or diode (UDT`s diodes i.e.), but I afraid, their need more distance as ca. 30cm! w*w.udt.com(newly is= osioptoelectronics.com), Position Sensitive Detectors, HAMAMATSU PHOTONICS Deutschland GmbH: ://web2.cylex.de/firma-home/hamamatsu-photonics-deutschland-gmbh-1676746.html
K.

With my lack of knoweledge on this area i must ask
Do you have a source on these position sensors.

Furthermore id like to ask if i can credit you guys in my repport
(and if you have an education in this area it would look rly good there too)

And actually i need to measure all 3 axis :S but i think i got the height figured out too, the box size goes up to L 50 cm W 15 cm H 20 cm

Also i need a quiet cheap solution, the school have given a 200(273 $) budget.

I have settled with a VCSEL and a foto diode, and then let the box move, and then measure the time it is cut of, and then i will measure the height with a bunch of diodes and foto diodes. But the width is still being very though ...

Hallo Horsevad,
Im in the past 7 years no more active in optical project areas, so dont know exact what real actuell is, but whats your time scedule?
The budget is practically nothing :-(...
Another idea, what about over two cheap (web?)cameras and a crossline lightsource of direction (more or less)top= for the X & Y axes and the second line light for the hight measurement, the values you will calculate from shadow forms?
K.

Well I have to be done in about 2 months for this project and it is 7 hrs in school every week and about 100 hrs of homework.

And the weebcam thing sounds a little to fancy for my skillset, but it is ofc. a possibility.
And the budget sucks, and this is the final exam project in this class

but thanks again for the help

With normal cameras is the environmental light surly a problem, you need practically some pulsed source for blending/switch sun/lamps disturbances/effects out...
If your place is to less for hight measurements, what about reflected light?
Do you have triangulations sensors in the school?
By moving object their are usable too..
K.

Nope, this school lack equipment, money and good teachers, unfortuantly, so mostly our projects are just "ideas" described in a report without anything really being build, so i think ill have to stick to the very basic "cut of" method

I write only snap ideas...
If you has 3 line lights (must not be a Laser or VCSEL, but I would take NIR VCSELs, i.e. at 850nm), the most semiconductor cameras has enough sensitivity at these IR wavelength (you can check it) & 3 cheap cameras...
A switched line-ight from at 45 degree to objects surfae (on all axes one) and the calculated shadow on objects surface (a "square"-line deformed from straight line, problem is parallelity to mooving/lighting axes, you need a trapezoid form to calculate instead square...
Of course light can be visible too.
These is in my opinion the cheapest solution.

Added after 3 minutes:

Check pls these posiibility: http://help.adobe.com/de_DE/Acrobat/9.0/Standard/WS58a04a822e3e50102bd615109794195ff-7bf2.w.html
**broken link removed**
Sorry, I found only in German_but seems more detailed as in english version....
**broken link removed**

Well to keep it inside the "topic" of our project we schould have something that can be combined with the microcontroller we have used the whole year, and i think setting a webcam to the controller is a little above our skills, but the idea with a VCSEL seems possible (We will just use a limited amount, so that we can only detect the few lenghts that we need) but do you need any special drivers or anything to make it work, or can it be connected like a normal LED diode ?

Also; VCSEL is a diode!
You have relative exact to handle it_give no over current!!
Otherwise their are very cheap, see pricelist from Roithner-i.e., most are Taiwan products, but Osram is nowadays one the best producer.
I have had for a DESY project lot of highest speed VCSEL-DIEs(chips) 10 years ago-from Taiwan, their are super & we qualified at CERN for (particle)ray doses too...
You can calculate with a forward voltage of 2.2-2.4V and current of 5-20mA, a simple serial resistor (or current source for stabil light energie), as for LEDs too.
K.
I didnt know theses system, but my idea was similar :
Messprinzip CAM3D ist ein optischer Sensor basierend auf kodierter Beleuchtung. **broken link removed**
3-D volume measurement www.vitronic.de/en/vipac-camera-based-auto-id-barcode2d-code-reading-and-ocr-3-d-volume-measurement-weighing/
A very interesting script_Monokularen Rekonstruktion zur Vermessungund Rekonstruktion in 3D:
**broken link removed**

Thanks for the help most apreciated