MT9D111, OV2640, OV7670 > POA030, CP8108. Weird Title.


I have been working with imaging sensors for some time now but was unable to "speak" to them, you know how the opposite sex doesn't speak to you even though you ask her "what her problem is."

Very sticky situation as you can see.

I had been largely working with korean image sensors namely Pixel Plus' POA030R and Clairpixel's CP8108 but was being largely unsuccessful even though I had help with the coding from some friends. After realizing that it was first important to work with sensors that people have already worked with, I went back to basics and started working on two but now three sensors; MT9D111, OV2640 and OV7670, the latter being quite the popular, not-so-muscular choice. 

I had realized that XCLK has to be provided if the module has no oscillator of its own but  didn't realize that the clock i was providing through the MCU was faulty, somehow when I calculate the clock output to be 8Mhz, it was only spitting out a mere 4. After playing around, got a clock of 6MHz, just enough to drive the modules to work. And it did.

I am betting getting the data won't be such a hassle as I2C or SCCB was probably the most strenuous portion. With that cleared out, I should be able to get some image data out from the camera for the very first time. 

Very exciting!

Test setup:
Custom Built UART to RS232 to USB (computer).
This allowed me to "see" what's happening with the cam module.
It's actually our Science Unit EGSE (Electrical Ground Support Equipment)



SNUSAT-I Camera Dev, board features multiple camera inputs
for Korean, OVxxxx and MT9Dxxx sensors
Simple I2C test setup for camera. Was trying to give XCLK from Arduino
but didn't work well. The clock signal is too unstable.
I thought it was the pull up resistors that was causing problems but nope. 
Results:

I tend to document everything up in ppt style much like how J did it when he was here at the lab. The documentation has been shared below:

Comments

Popular Posts