• Project Log 2: Making a Design Language

    WalkerDev • 09/08/2024 at 13:29 • 0 comments

    To start off, let's create a basis for how we will design these! 

    First off, we know this is directly meant for VR related things (gloves for example). With this in mind, i'll be making a flex and rotational sensor (will make more sense later). The rotation sensor can use any standard material for now (we also will save working on this for later). However, for the flex sensor, we will use PCBWay as a reference.

    From their array of materials, we will select the Nylon HP-PA-12. I used this in the past to make a headstrap for the headset! It also works because at our price point it should be extremely cheap.

    Next, we have to think about how everything will actually work for the flex sensor. We will use a linear array of 3x2mm Neodymium Magnets at the base, sitting on a flex PCB ribbon (to still stay connected during extreme bends, while Nylon is flexible, I don't trust it to hold a magnet without popping it). We have a secondary flex PCB with the hall effect sensors above, aimed at the magnets below it. 

    Finally, to protect the hall sensor from reading outside forces, we use a flexible ferrite sheet. Ferrite is used within medical equipment and other high magnetic sensitivity equipment where stray magnetic force results in failure. In our case, we are just using it to ensure you can still use your phone while you have the glove on!

    To read more on this, you can check out the link below:

    https://www.hilltech.com/products/emc_components/nanocrystalline-shielding-amorphous-shielding.html

    To put everything together, i'm imagining there is a singular hard point at the back of the flex sensor with a piece sticking out so you can put some sort of mounting hole through there, then close it with a snap on piece. The flex sensor moves forward and back within a mini rail it slides front and back on! Behind the circular piece should likely be a tiny JST connector, allowing you to send the data to a board of your choice (Which we will design later). 


    We will start out with a 2.2 inch strip, separating each sensor/magnet by 0.3 inches but have the first magnet a little further than the rest so we can later do some magic code for when the magnets overlap.

  • Project Log 1: Project Start

    WalkerDev • 03/02/2024 at 15:56 • 0 comments


    I got this idea based off of a research paper a friend showed me and decided to go full prototyping mode with it!

    So let's say we have two tapes right, E is our tape with magnets on it, which is slightly longer than D (which has hall sensors). They are always B distance from each other and are held together at the start only (The yellow line). We set the components through a pattern (C being a larger value than A) and we ensure the starting end of the tapes are connected while the rest moves.

    When straight, the sensors overlap a certain way (Not directly on top of each other). The first and third magnet will be closer to the center while the second directly overlaps with the hall sensor. 


    When raised, the tape at the end will be closer to the middle.
    When down, the one at the end will be further away.
    Put this on a flexible PCB and you never would have to worry about flex degradation EVER again.

    However, a problem comes from this. If the hall sensor overlaps between two magnets, the value might not be understandable if you're thinking "Gee, the filter to get the data is going to be very hard to extrapolate". In that case, you lack creativity. If we were to put this on a finger, the first magnet would be closer to the hall sensor, the second would be slightly away and the third would be the furthest. The same is vice versa.

    In short, as you raise the system, the first value will always increase while the third always decreases. Then, you can estimate the way the tapes are overlapping (which is much easier than you are imagining) to estimate if the hall sensor in the middle went left or right (and from there, the arc).

    I am working on a flex PCB, 3D model and code for testing