• Progress in the last month and a half

    Tyler Stevens • 11/03/2020 at 04:17 • 0 comments

    I haven't been updating this project as much as I would have hoped I would, but let me give you a reader's digest of the work I've been doing the last 6 weeks.

    First on the docket is a partial redesign for the base. While not complete by any stretch of the imagination (I haven't even sourced a Type C connector, let alone begun designing the connector daughterboard or supporting plastic components) but it does show good progress in how the final baseplate look.


    From an electronics side, most of my time has been spent reading electronics spec sheets and establishing power to all the chips. I have selected a CYUSB306x for its integrated high speed CSI-2 bus, integrated uC with serial interface tools, and integrated USB interface hardware. For the camera, I selected a Arducam IMX219 8 MP camera module with integrated auto-focus hardware. The theory is I should be able to load the microcontroller with OpenCV autofocus algorithms, though my experience with OpenCV is limited, so it'll take some time. Also of note is the fact that this camera is designed to work with the Raspberry Pi Camera Module V2. Thanks to the outstanding work done by DrYerzinia on his Raspberry Pi Camera Module V2.1 Reversed project, I had a full pinout for this module, as well as tide and true hardware selections already made. It was a simple decision for a project that is increasingly ballooning in hidden considerations and design.

    As I'm still working on the circuit design, I don't really have a board ready for you. My notebook* and the circuit schematics I currently have are an embarrassing ratsnest, so unfortunately not a lot of fun visuals in this entry. Hopefully for my next post I'll have something a bit more fun to show you :)

    *I do plan on photoscanning and uploading my full journal so that you can see my design process, but due to the logistics of doing it page by page, I'll probably wait until I at least have the beta hardware design up and running.

  • Getting the PCB Dimensions

    Tyler Stevens • 09/14/2020 at 10:48 • 0 comments

    When I started capturing the board dimensions I made a lot of assumptions about the shape. As I progressed, however, I quickly learned that the exact shape of this board is surprisingly nuanced.
    The cutouts were not of uniform depth and needed to be nearly perfect to ensure I could get the screw hole cutouts in the right locations. It took a few redesigns to get it just right.


    Since there was no universal geometry I could base my work off of, I used a pair of calipers and measured around the board, using locations from my previous measurements as a base for my next ones (I'm sure I've broken a fair number of cardinal sins doing this, but since I have no classical training in CAD this is the best I can do for the moment). Regardless, I'm happy with how this turned out. After working all the way around and mirroring my design, I was able to draw a line bisecting the board that was 10 mills off from what my calipers reported. Considering the rough finish on the sides of the PCB and the fact my caliper's batteries were almost dead, I'm quite happy with this. 

    I'll order a blank test board once I've queued up enough PCBs to justify a order from JLC. There's plenty more to do in the meantime, like developing a circuit, sourcing a CMOS sensor, designing a mount for the camera module and USB receptacle, reverse engineering the audio/ LED status board, and checking the feasibility of manufacturing an aluminum USB C cover on the base to replace the Firewire one.

    Stay tuned!

  • The First Steps

    Tyler Stevens • 09/14/2020 at 10:28 • 0 comments

    This is a quick summary of things I noted during the disassembly. I'll keep it brief, since there are much higher quality summaries than I could ever hope to recount on here. I also didn't take many photos of the process. 

    One of the biggest concerns I have with the disassembly process is the difficulty of getting the pan/tilt connector out of the black plastic housing. It was challenging to say the least and a bit heart pounding. I see this as the biggest potential failure point for anyone trying to do a restoration on their own.

    Another challenge was separating the two halves of the interior frame. I felt a bit nervous about breaking the clips, but if you wedge a guitar pic between the two halves and gently nudge each clip as you gingerly ride the pic up along the crack, you should be able to separate the two haves without too much trouble. 

    Finally, there's the microphone daughter board. The LED indicator light is apparently fused to the internal frame either with adhesives or some sort of ultrasonic welding. I tried to remove it but couldn't see any way to remove it without damaging it. Since this restoration is meant to be non destructive, I've elected to use the stock audio board in my design. Ultimately the sound quality should be decent enough for a webcam, and it incorporates a built-in off the shelf ADC, so I shouldn't need to fiddle with optimizing the microphones' (yes there are two of them!) dynamic range. Sadly I might have to destroy mine to reverse engineer it successfully. More on that when the time comes...