- Work this week
- This week I finished our second 3d printed housing. I cut more lenses housings and 3d printed the part to attach to the PCB, and assembled and tested everything. This housing works as well as the first, so mechanical work on the boards is done.
- I discovered that the “low” power setting on one board is pretty high. This seems to be causing some inconsistency in communications, as the receiver cannot always tell it is supposed to be low.
- I worked with Anju and Roger to integrate the whole communication pipeline together. We both ran tests using a mirror to reflect data back to a board and using both boards. Reflecting data back worked well, but using both boards was difficult for some reason. I am continuing to debug why. We did this sending raw bytes, not data encoded in our actual format.
- Schedule
- We are still working on integration. The comms pipeline has been integrated in its entirety, but it appears buggy, so there is still work to be done. Once that is complete, we will move on to testing.
- Deliverables
- Next week, I will fix the laser on one board to have a lower “low” power setting to make communications more reliable. I will also continue work on integration with Anju, and will make a test setup to hold the boards stable for easy testing. I also plan on disabling the IR laser to prevent accidents, since we are not using it at all.
- V&V/Testing
- Have run:
- I have used an oscilloscope to measure the rise and fall times of the signal powering on the lasers to ensure it is sufficiently fast for 6.25 Mbaud communication, which it is.
- We have run a communication test at 6.25 Mbaud through the lasers, with USB and FTDI interface running quite a bit faster, to demonstrate that the hardware can support communication at our required speed. Because we have one laser, not two, this requirement has doubled to 5.42 Mbaud, so we made sure to test above that speed.
- I have done some hand testing with optical angular error. It appears that this error must be significantly smaller than we were hoping for and set as our requirement. I plan to do further work testing this once I have a test stand setup that makes it easier to hold the boards stable, and document more precisely what is required.
- I have measured the current draw of our device and made sure that it draws below the USB-C spec of 3A when powered on 9V USB PD.
- I have tested the board in a variety of lighting conditions (light, dark, different rooms, natural vs artificial) with ambient light filtering on and it consistently performs the same. This verifies that it will work regardless of ambient light conditions.
- Plan to run:
- Measure the received data on digital line to FPGA when receiving 6.25 Mbaud data to ensure it is clean, and rise and fall times meet our spec. Just because it appears to work does not mean this signal is clean.
- Run a test of the full system with the full software stack running and verify speed meets our requirements.
- Run a system stress test transmitting a large file while intermittently blocking the laser and verifying it still gets transmitted.
- Test the boards at a distance of 1m to ensure data can be transmitted at that distance.
- Have run:
Anju Ito, Roger Lacson, and KJ Newman