We have been quite busy since our last update, but are excited to share our progress with you! In the past month we've been evaluating prototypes, touring factories, and hiring new talent. Let's dive right in!
Printed Circuit Board Layout
As discussed in the April update, we identified some issues in our PCB that needed to be resolved. Since then, we’ve implemented the changes and begun evaluating them on a fully assembled unit.
1. PCB/Battery relocation: Recall from our previous update that we added a second battery and shifted the PCB to the arms of the glasses. We’re happy to say that we made samples with this new arrangement and it all fits.
2. Flex PCB: We also mentioned using a flexible printed circuit (FPC) to ensure that the components fit safely within the curvature of the glasses. We have redesigned the PCB architecture such that the batteries, bone conduction transducers, and rigid PCB are all connected by FPC sections. See the image below to understand how it all fits together.
3. Component height: We also rearranged components to reduce interference with the walls of the glasses. On such a small PCB, this was a tedious task. We’ve successfully assembled the PCB with components and can confirm that there is sufficient clearance within the glasses.
Our original LED location on the inner side of the arm of the glasses was meant to blink subtly in your peripheral vision. While testing, we’ve found that the LED is too far outside most people’s peripheral vision. Because of the hinge location, we have been unable to move the LED far enough toward the front of the glasses to make it more visible. The only way to make it visible was to boost the brightness of the LED—but that basically meant the LED was bright enough to shine noticeably across your entire face.
We have been exploring alternative LED placement, and decided to shift it to the front of the frames. The benefit here is that it falls well within your peripheral vision, making it hard to miss. By setting the LED brightness to a soft setting and adding a slow fade in/out, the result is actually quite nice. See the GIF below.
The new position of the LED ensures that notifications are visible yet soft enough to keep Vue discreet.
Preliminary CNC Testing
We have made new prototypes using a method called computer-numerical-control (CNC). CNC machines are able to automatically cut parts from solid blocks of material, resulting in a more precise and higher fidelity prototype than can be achieved with 3D printing. The next stage is to evaluate this design, fix flaws, and re-evaluate the fixes until we are satisfied with the quality. For example, we may find that a dimension is off by a fraction of a millimeter. Or we may find that the thickness of the plastic over the electronics is interfering with the antenna. Or we may find that a section of the FPC is too loose or too taut. This is an essential, iterative part of pre-production work, as any changes after tooling will be expensive and time consuming.
Once we are satisfied with the product’s performance, we can can begin cutting tools for injection molds. Check out some images of the CNC version below.
Progress is great, but we also want to make sure to give backers insight into the true complexity of manufacturing a product, and that includes talking about headaches. We detailed a few in our previous update, but we've also encountered some unexpected ones with our new CNC version.
One such example is the touch panel. The touch panel is mounted on the underside of the temple piece, and senses your finger's movement over the surface. Our previous prototypes were 3D printed and painted. When we CNCed our prototypes, we used a different kind of paint to finish them. It turns out that the new paint interfered with the touch panel's ability to sense gestures. When we were testing gestures on the CNC version, sometimes the glasses didn't detect anything at all. After mounting and re-mounting the panel, as well as conducting numerous tests, we finally determined that the paint contained small amounts of metal which affected the dielectric of the plastic, resulting in poor performance.
This doesn't impact timelines (since we fixed the issue), but as you can see, there are always unexpected outcomes during product development. This find just means we'll have to double-check the pigment used during manufacturing to make sure it is compatible with our touch panel.
When you manufacture a product, a lot of things have to happen in parallel. While we continue preparing our design for manufacturing, we must simultaneously evaluate contract manufacturers so that we can seamlessly transition from development to production. Below are the facilities we've been evaluating.
1. Injection molding factory: Injection molding is a process for producing parts by injecting material into a mold. For Vue, we will create molds for the plastic frames, and then assemble the electronics inside those frames. Check out this brief video (credit: engineerguy) that provides a generalized visualization of injection molding.
Creating these molds requires careful precision and is very expensive, so it is important that we have finalized the product as much as possible before creating them. In the meantime, we’re evaluating top-notch injection molding facilities to make sure we’re ready to make a high quality product.
We’ve been impressed with the facilities we’ve seen so far, and are excited to select a quality partner to move forward with.
2. Testing & Design Facilities: In addition to touring manufacturing facilities, we've also been meeting with partners in the traditional glasses industry. We've established a partnership with a group that has helped us refine our designs such that our frames look like traditional glasses, as well as offered us the ability to use test equipment to evaluate our products. Check out some images from these trips below.
An example of a test that we'll be conducting is hinge fatigue. Below you can see a video used to test the lifespan of hinges on traditional glasses.
We plan to subject Vue to similar tests, perhaps on these very machines.
New hire—production manager!
We are thrilled to announce we have hired a full time production manager. After interviewing many candidates, we are happy to welcome Jia Yao to the team. Jia has over 10 years' experience in manufacturing and supply chain management. His experience spans a variety of products, including hardware, machinery, toys, and even furniture. Most recently, he spent ~3 years working as a mold design engineer at Foxconn, where he designed molds for computer products and saw them through to production.
We are confident that Jia's extensive experience in manufacturing consumer products will be a huge asset for Vue!
The immediate next steps are to review the new prototypes and electronics. This is an iterative process of identifying problems, creating a fix, re-printing PCBs, tweaking software, and conducting a systems test again. Once the prototypes meets our quality specs, we will then begin to cut molds for our frames, and will work toward mass production.
Surveys and Support
Just dropping another friendly reminder to check your email for your surveys (if you haven't already finished them). Also, remember the following:
- If you chose plano (non-corrective) lenses or no lenses you will not receive a prescription survey.
- If you need but haven't received a prescription survey, please email firstname.lastname@example.org.
- We've extended the prescription deadline to June 30th, 2017.
- If you would like to update your shipping info or check your order (style, color, etc) then you can use BackerKit's survey recovery tool to look up your order.
Until next time!
the Vue team