The components for the earplugs in the transparent plastic version. QuietOn Sleep’s components are still so tiny and the soldering wire is thin as well. To assemble an earplug, most of the tasks have to be done with the help of a microscope. Components and wires are connected to the PCB by soldering, which happens at a lower temperature than welding. Here, some of the components have just been manually changed using a soldering iron. Sometimes we use a hot air soldering gun for the same purpose.
Note that mass production doesn’t use soldering wire and manual soldering tools. Instead of solder wire, solder paste is printed on the PCB and a pick-and-place machine puts components on top of it. Finally solder paste is molten in a reflow oven.
As many of you have asked us on the project’s details, this update is to provide you with more in-depth information on the current R&D progress, what have been done and what need to be improved. However, the schedule is still unchanged from the last update and we’ll keep you posted in the next updates. For now, we target to have the mass production in January week 4 and the first batches can be delivered in February. In addition, we would love to share with you some interesting facts and knowledge on the designs. We also make a short recap to summarize the project’s progress over the last few months for the latest backers.
In the following sections of the update, the design idea and principles are presented first to provide you with the foundation and rationale, followed by the progress update.
The earplug has a Hall effect sensor which switches the device off when there is a magnet nearby it. There is a magnet under each earplug’s holder in the charging case. The earplugs stay in inactive mode when they are in the case and only drain a miniscule amount of the battery. When the earplugs are removed from the case, they are on automatically.
With the real plastics, the magnet switch is confirmed to work well after the usability test.
The charging case of QuietOn Original has metal clips that hold the cover shut. While they work, they are easy to bend so they don’t hold strongly enough or at all. The locking mechanism is changed into magnetic lock on Sleep’s charger. There is a neodymium magnet in the top and bottom part of the case. They attract each other and keep the case shut. The magnets need to get close to each other in order to pull strongly enough. This means the plastic layer can only be 0.5 mm on both parts.
With the real plastics, we assembled the earplug for testing (as shown in the previous update). The plastic layers between the magnets was 0.6mm, which is already good. However, as explained above, according to our design, 0.5mm will make it work even stronger, which is doable but requires some special attention in tool making and plastic molding parameters so that the area fills properly and looks clean.
Feedback has been sent to the tool maker. Fixing process is within 1 week.
Top cover of the earplug is made of two materials with a two-shot molding process. Most of the cover is made of hard plastic. Some parts are made of soft elastomer, such as gaskets, and parts that make the button move easier.
The elastomer forms a groove that has soft walls. This is the acoustic channel that leads from the mic holes to the mic component located behind a hole on the opposite side of the Printed Circuit Board. The soft walls of the acoustic channel function as gaskets that are pressed against the PCB and make it airtight.
In the latest PCB, there was a track routed across the area where the gasket touches it. The surface of PCB was not smooth where there are tracks because a track was made of copper and is thicker than its surrounding area. This unevenness caused an air leak. Accordingly, we restricted the gasket area from routed tracks and made it a smooth, solid copper fill.
The New PCB with correction is coming in the next build, estimated time: 2-3 weeks.
The microphone used by active noise cancelling and hearing mode circuits has two holes in the top cover. The given reason is that if one gets blocked with dirt, the device can still continue to work. Two is the maximum number of holes we can fit in the space constraints. The holes are small and placed next to each other. This has caused some difficulties for the plastic manufacturer. The narrow area between the holes has a risk not to fill completely, which requires an increasing pressure to unusual high values. The design of the small holes correspondingly leads to small metal pins in the mold. Thus, it involves a risk to damage the pins or reduce their lifetime.
The tool maker proposed drilling the holes, but this suggestion was rejected. Drilling was difficult to do because you have to drill through hard and soft plastic. Further, the entrance of the hole was supposed to have a rounded edge. However, the drilled holes leave a sharp edge which could collect dirt and grease from a swiping fingertip and eventually block the mic hole. A rounded edge would reduce the risk. The microphone acoustic channel makes two 90-degree bends so there is no direct path for dust, hairs and light to reach the sensitive microphone component. This meant that we couldn’t make more mic holes because they would get too directly above the hole of the microphone component itself.
After some suggestions and correspondence, we have found dimensions that should be doable with molding process and still follow the initial design idea and principle. Tool maker gets things fixed: 2-3 weeks.
MODE SWITCHING BUTTON
We conducted the usability test after we had the first assembled product with the real plastics. The mode-switching button turned out to be very stiff. The button is a hard plastic part in the top cover which is suspended with two narrow strips of the same plastic material. Remaining gaps are filled with elastomer in order to keep the device air and sound proof. There is also a thin layer of elastomer on the inside surface around the whole area to increase adhesion surface area. There are two reasons for stiffness:
- There is an unnecessarily big gap between the button and the actual switch component on PCB. That’s why when pressing the button, it takes some time to bend the plastic strips before the switch actually takes effect. This increases the pushing force.
- The gap between some components and the elastomer is too small. Once the elastomer touches a component, it becomes more difficult to press further.
We did quick prototyping and FEM simulation (Finite element method) to test if the hard plastic and elastomer need to be changed and how. Gap between button and switch component was reduced, and elastomer was reduced around components to make them more room. This lets the cover bend more freely and actuate the switch sooner. Feedback was sent to the tool maker to adjust. Estimated time: 2 weeks.
Check out this solution in the below short GIF:
ESTIMATED TIMELINE OF MILESTONES
Packaging materials and the user guide are ready.
CE testing process: Test cases to fulfil the standards of EN 62368 has been agreed with the authorized test house. This process will take place in parallel with the other R&D steps. Targeted to be finished in W2.
Audio fine tuning in the final assembled product: W2
Ramp up for mass production: W4
Estimated first delivery: February.
RECAP FOR THE LATEST CUSTOMERS:
In mid-September, on a flight back from the factory in Poland, our audio and electronics engineer discovered that the performance of the earplugs on airplanes was less effective compared to the design and calculation in stimulation. The team figured out the problem and solution, yet it took more than one month to measure the results and adjust the design. More detailed information can be found at these two updates: https://sleep.quieton.com/2018/09/17/pre-order-no-2/
The new design was applied into the new plastics and other tooling molds. We managed to have the real plastics for the assembled earplugs at the end of November for audio and other testings at the lab. The progress was updated right away in this update: https://www.indiegogo.com/projects/quieton-sleep-world-s-most-powerful-snoring-killer/x/16536470#/updates/16