Backplane
Overview:
Jus a big ole ethernet switch
History
Backplane:
Rev 1.1.0→ RE @Billy S 22-23. Ethernet did not work because no magnetics. Rev 1.1.1 → @Ben Hyman 23-24. Before this revision we needed to determine if the ethernet issue was the lack of magnetics so with @James May we tested magnetics with Kludge, and it worked. Made rev 1.1.1 which added magnetics, USBc power, and new 30 pin keying. Note on USBc power; if another rev is needed implement a proper power mux current (1.1.1 is power after power mux, rationale was we are only going to use USBc power for testing). Main take away from this revission was look at reference schematics.
Radio Module:
Rev 1.1.0 → RE @Dante Sivo 22-23. Miraculously works. One module has a broken GPS
Power Module:
Rev 1.1.0 → RE @James May 22-23. Basically worked but had a slight volatge sag on the 3.3V rail. Rev 1.1.1 → RE @James May 23-24. Works really well.
Autopilot Module:
Rev 1.1.0 → RE @Ben Hyman 22-23. Mostly functional but there are a few issues. Buzzer not working (same as Grim Reefer need to swap out transistor). Ethernet works! Haven't fully tested to give good feedback
Sensor Module:
Rev 1.1.0 → RE @Mary Dertinger (RIT Student) @Louis Fleisher (RIT Student) and @Alex Moczarski (RIT Student) 22-23. Ethernet didn't work. MS5601/BMP address conflict. 2 layer..... caused some design decisions that could have been avoided. Bodged around Effuse wrong calcs for rising edge trigger - voltage from power mod was 3.26 instead of 3.3.
Rev 1.2.0 → RE @Mary Dertinger (RIT Student) @Louis Fleisher (RIT Student) and @Alex Moczarski (RIT Student) 23-24. Ethernet still doesn't work - we changed 30 pin to same but with shroud so part in Altium changed, footprint looked same but was different, so it was oriented incorrectly. MS5611 wired up wrong - connected main I2c clock to wrong pin was a copy paste error. Had DNP resistors but got populated (happened on 1.1.0 too) 3 bootstrapping for Wiznet and another for a sensor (R1). The issue was they were 0 ohm resistor so you had a short from power to gnd. No Effuse problems. I2c problems: too many transactions; bus will lock up. Same with 400 kHz will also lock up. thought it was because of LSM16
Rev 1.2.1 → RE @Mary Dertinger (RIT Student) @Louis Fleisher (RIT Student) and @Alex Moczarski (RIT Student) 23-24 (Mostly @Louis Fleisher (RIT Student)). Ethernet works and overall functional but still testing.
Requirements
Requirement | Rationale | Satisfaction |
|---|---|---|
The backplane shall provide a method of communication between all modules | Adding simple boards to the system should be relatively easy | SwitchBlox Nano switch |
Each module shall be able to function independently to the best of its ability | Allows computer to be in modular configurations depending on flight needs | Sensor, Radio, Auto Pilot and Power - questionable because you need power mod |
The backplane shall route power to every board (from power mod) | Boards need power | Power mod but backplane routes rails to each module, 3.3V and 5V |
The backplane architecture shall define a standard width and height for each module | Easier to mechanically integrate | See backplane Standards |
The backplane shall survive launch operations | A flight computer should be able to survive a flight | Tested in 2 IREC flights |
Modules shall be individually mechanically accessible | Speeds up prototyping/troubleshooting/repair | NA |
The modules shall be drop-in interchangeable | Speeds up prototyping/troubleshooting/repair | See backplane Standards |
Each module should limit the amount of external I/O | Prevents a rats nest in the avionics bay/payload | See backplane Standards |
Flight History
IREC 22-23 KONG → Flew with no power
IREC 23-24 OMEN → Flew fully assembled but not fully functional, received one packet on from the pad but was useless because we plugged in wrong connector for radio.
Highlights
IREC 22-23 KONG podium session!! @Yevgeniy Gorbachev (RIT Student) presented → see attached presentation