Get yourself ready for the May update, it’s been a busy and exciting month at Cocoon HQ.
Three is the magic number
Each Cocoon contains three circuit boards and this month, the three came together for the first time.
The dimensions of one capacitor had been entered into our CAD software in mils (thousandths of an inch) rather than millimetres, meaning we didn’t identify the conflict between it and a component on another board at the design stage.
Fortunately the problem can be easily remedied making the error significantly less costly than NASA’s $125m metric mishap!
Putting it all together
The circuit boards have been placed inside the Cocoon chassis and tested. The motion sensor is now fully functional and night vision performance is looking good.
For aesthetic reasons, we wanted to use a single piece of material for the front of Cocoon, avoiding any material joins. Testing has shown that infrared light from the night vision LEDs refracts through the material affecting camera image clarity in some circumstances. An opaque divider separating the LEDs from the camera will have to be inserted but rest assured that aesthetic appearance will still be a priority.
Our March update explained the problems we had with WiFi signal performance which forced us to redesign Cocoon’s processor board. The new design was reviewed by the WiFi module vendor, the antenna vendor and an independent wireless consultancy.
The characteristics of the antenna are altered by metals in the circuit board and the components it is placed beside. The characteristics caused by the manufactured processor board are now being measured and the effect of them will be compensated for before we progress to WiFi performance tests in the first weeks of June.
Camera driver surprise
Early development and testing of Cocoon’s camera circuit board took place in conjunction with a development module provided by the image sensor vendor. This allowed us to develop camera software before our other circuit boards were complete.
However the chipset on the development module was, unbeknown to us, compensating for an unusual ordering of colour channels in the signal from the camera. With the development module now out of the equation, we have some unexpected work on the camera driver to do in order to ensure a good quality video stream.
We’re working with the image sensor vendor to do this and expect to resolve the issue this month.
We welcomed James Salt as our test lead. James has spent the month testing the Cocoon Android app and is busy recruiting staff to join his team. A dedicated test team will ensure we have the information we need to deliver a high quality product.
Paul and Nick have been preparing tests for the first stage of compliance testing which will take place in June. A Cocoon will be placed in an electromagnetic compatibility (EMC) chamber and put through its paces whilst being blasted by a range of radio frequencies.
We’ll be checking both that the functionality of Cocoon is not impaired and that Cocoon itself does not generate any radiation or power line fluctuations that could interfere with other devices. The test is one aspect of the regulatory certifications required for electronic products.