The Week in Review
This past week saw promising progress being made via range testing, firmware bring-up, and PCB layout. We were able to pair and get data sending between two of our Wi-Fi transceivers, and had acceptable results during range tests at ~300 meters, with the help of two 9dbi antennae. This range result was still under our target for Wi-Fi range, so more testing will be necessary. As a fallback, we’re looking at inlining a Wi-Fi amplifier, putting a directional antenna on the base station side, and, in the worst case, sticking with one-way communication for reading car data
Risks
- Complexity has been added to the design (from revision 0) by virtue of adding a second radio and splitting operation across multiple PCB’s and multiple radio transceivers. This complexity is not entirely trivial, and could cause system failure if details are overlooked.
- To try and avoid these potential pitfalls, we will remain on the 2-revision timeline described in our proposal presentation.
- Moving to a 2-PCB design will itself mitigate this complexity, by directly encapsulating system failure
- The radio transceivers may have insufficient range. The cost of this is the inability to adequately communicate between the car and ground control application on race day. The following are proposed mitigations:
- Transition from a omni-directional antenna to a directional antenna at the ground control
- Use an on-board RF amplifier. This would require added hardware complexity (adding the amplifier to the RF path on the transceiver PCB), and will only be implemented if we can’t otherwise meet our range requirement.
- Risks that have been resolved as of this week’s entry:
- Basic schematic integration (Does the MCU/radio turn on and execute blinking lights firmware template?)
- SPI communication on STM with the XBee radios
Design changes
- Moved from a single board to a 2-PCB layout with 32-pin ribbon cable interconnect
- Splitting the RF modules onto their own boards was necessary to fit into physical footprint constraints while keeping the board manufacturable and under cost
- 32 pins provide plenty of flexibility, and the ribbon form factor gives spacing
- Known unknown: EMI risks, with SPI at 3+ Mbps on a 5”+ cable
- Sources indicate this should be tolerable, but we will keep an eye on it
Updated schedule
No updates: We’re on schedule in terms of PCB schematic and layout, as well as firmware bring-up and implementation. (See our Gantt chart for more details.)