Project Logs

Collapse

• Exploring the camera

  elementcarbon12 • 06/14/2015 at 06:20 • 0 comments

  I just played with the camera a bit. I connected it to my computer and after a bit, it shows up in the Windows device manager as Amba simple class. Google founts lots of hits for people looking for drivers, often for cameras which sounded promising. I checked the VID and PID which is 4255 and 0001 respectively. That matches a GoPro Hero 2 camera. What the heck is in there?!

  I decided to tear the camera down. It was pretty simple, there are six screws, 5 of which were easily visible, one hidden under a black sticker which left the typical VOID marks when I peeled it off. Well, it's unsupported so there's no warranty anyway :P

  Then there were a pair of slots in the bottom rear placed near the center line. I put a Flathead screwdriver in them, wiggled a bit, and the popped off allowing me access to the innards. It should not take any significant force to pop the clips, but it will take a but of force to pull the back of the housing off. Again, it's not a lot, if you feel you are using a bit too much strength, you probably are. Wiggle things and see what's moving and what's not.

  This is what it looks like with the first piece off (please excuse the messy desk :P ).
  

  Then I removed the from fascia.

  See that rectangle in the front? Here is a closer view of the camera assembly.

  It's a mechanical shutter! Also note the gears on the left hand side. The camera is on a gimbal. I pulled back the shutter and tried connecting it to my TV again, and I got video!

  Here are the boards from the camera starting with the one that runs along the front of the unit.

  That's the main part of the board, and FPGA, an ST SPI flash part and one more tiny IC in the top left corner of this image. I have no idea what P1 and P2 are for. Perhaps programming or debug headers for the FPGA. I don't have any experience with this sort of thing so I don't think it will do me any good.

  Here is a closer shot of the ICs with the marking on the ST part clearly readable.

  Here is the tiny IC in the top left I mentioned before.

  Finally, there are these odd features that run the length of the front of the board. Here is a close up of a few of them.

  These features are really small and it was hard to grab an image. There is something that looks like a segmented border around a big via and 4 pads. They are labeled UXX but I can't spot any ICs on either side. The front fascia has mesh covered slot where these features are. My guess is they are used to pick up audio using capacitance, but that's a wild guess.

  This board also controls the shutter, has an LED, and might have an IR sensor in the same package as the LED (it's either that, or a second LED telling you that you are "on air").

  On to the video board.

  That big one is a Silicon Image Sil9134 which handles the HDMI.

  Here is a better view with the companion buffer chip.
  

  There was a collection of NXP ICs scattered around and a couple other small ICs. I suspect the there are a couple MCUs to handle USB, motor commands, and some motor driver ICs.

  Here they are in no particular order.

  And finally, a possible way in to the camera subsystems, unpopulated headers!

  One is clearly labeled I2C but I have no idea about the other 2.

  At this point, it's been a while since I've poked at any hardware for fun so I'll need to update my toolbox. Maybe P9 is JTAG, but I don't recall any 10 pin pinouts off the top of my head. P2 is also pretty mysterious, maybe it taps into a UART.

  That wraps up tonight's adventure. I think I need to see if the set top box can do anything without the service. I'd like to see if I can sniff the USB traffic with a hardware analyzer and try to get started on the motor commands. I'd like to be able to activate the shutter and manipulate the gimbal.

  I also want to take a look at the HDMI chip and see if I can get video and audio out of the camera in a way that is more SBC friendly.

  Overall, the camera was pretty simple to take apart and the video board looks like it won't be too hard to figure out. The FPGA is a bit daunting though....

  Read more »

View project log