Line Scan Camera Board

[#1]

The Freescale linescan camera is based upon the TSL1401CL sensor from TAOS Inc.

Design Files

• Schematic & 3D Render (Courtesy of eli_hughes)

Images (Lens removed)

Freescale Linescan Camera Specifications

• 128-pixel linear image sensor (TSL1401CL)
• Focusable imaging lens
• 5-pin physical interface on PCB on .100" grid
• Simple three-pin MCU interface with analog pixel output
• Lens: 7.9mm focal length, f2.4 fixed aperture, manual focus, 12mm x 0.5mm thread
• Exposure Time: 267µS to 68mS
• Resolution: 128 pixels
• Built-In amplifier stage to improve white/black differentiation.

The lense used on the board:

Alaud Optical

    8.0mm f.l. Lens w/IR filter = Part Num: AB0825C   

    M12x.05 Lens Holder = #9 or #10

Useful links

AMS TSL1401 Product Page

Line Scan Camera Use

Freescale App Note: Line Scan Camera

[danielhadad]

Trying to get a cheap oscilloscope to help tweak this camera.  Seems that this should be enough for $50:

http://www.gabotronics.com/development-boards/xmega-xprotolab.htm?showall=1

Can't handle more than 2MSPS or 200kHz (analog) but that should be sufficient for focusing this camera right?

Daniel

[#3]

You don't need a super fast o-scope for the camera.  For students I recommend the myDAQ from NI.  For my personal (not a student anymore) toolbox I have been wanting this Oscium iPad Scope.

You could also very well use the FRDM-KL25Z boards analog input to take samples for you. Sampling the data is the easy part.  The challenge is to graphically represent that data as we would expect to see on an o-scope...might already be an app out there that would work like the P&E Toolkit, we turned a DEMOQE board into a scope.

[danielhadad]

Well, I'm not a student anymore but I'm definitely having to go back to school learning a bit for this project.  Definitely fun! 

I'm more of an Android guy.  The board I mentioned above will go with this app letting me use my Nexus7 to handle the display: https://play.google.com/store/apps/details?id=com.nfx.noscpro

A fancier option for android would be OsciPrime: http://www.osciprime.com/    and https://play.google.com/store/apps/details?id=ch.nexuscomputing.android.osciprimeics

The MyDAQ seems good but seems I would have to get labview or something to handle the display?  It's slightly more expensive ($180) but that appears to be a student price and you'd still need to get LabView.  Specs say analog bandwidth is 400kHz which is only slightly better than that of XMEGA (200kHz). But sounds like we don't necessarily need to ramp clock up to full 8MHz capability-- or wont' want to (noise issues?).

Maybe I'll the the XMEGA and see how it goes -- will post here about it.

[eli_hughes]

Hello Daniel:

Sorry for the late reply!      So there are a couple options.  

1.)  You don't need a fast o-scope as you can control how fast you want to clock out the data.   For example, the TFC example code starts off at a 50 Hz refresh.    Multiply that by about 256 and that is the rate you will need (into the several KHz). 

2.)  If you are willing to spend a few hours hacking,   you could wire the camera into a freedom board and use the TFC API (see the FRDM-TFC page).    IF you get things attached to the correct ports,  there is a program that will beam data to the PC for you.   Right now it is a Labview app ($20 student version)  but I am planning on writing a version that uses the "processing" GUI (Free and open source).  

3.)   a bit more than you wanted to spend but this is a great little box:  http://www.rigolna.com/products/digital-oscilloscopes/ds1000e/ds1052e/

4.)  this is a cool one.....   http://www.seeedstudio.com/depot/dso-quad-aluminium-alloy-black-p-1034.html    They have some lower cost ones as well

[danielhadad]

Thanks!  Got the Xmega Xprotolab which should be sufficient sounds like.  Want a scope that can work independently from board I'm 'testing' just so I can focus my code on the application.  Still, might be good to add an 'focusme' algo to maybe have it easier to focus the lens.  Need to put a servo on the lens...

Will keep y'all posted here (or maybe via my 'status').

Daniel

[danielhadad]

2nd image above link broken-- appears likely due to TFC wikidot site going away. 

[danielhadad]

Xmega Xprotolab self test (AWG hooked to channel 1)

[danielhadad]

I already updated that I thought.

[#10]

fixed

[#11]

it was in two spots.  Fixed.