Who: Tim, Tyler and Andrew
- sensor output 10.4mV @ 650mG maximum Earth field means gain = 240
- 0.1 degree => 100 uG resolution
- f3dB = 100Hz to track airframe vibrations / roll
- Fsample = 1KHz
- 14 - 16 bit converter
- requires servoing (closing the loop) on the offset "strap"
- Switched capacitor filter?
- Screw it all: can we use the PIC's 10bit ADC converter? Answer: yes
- 9 bits = 0.7 degrees
- Still need instr. amp, but now use 1 pole RC filter
- Need Set/Reset strap components
- Need offset strap components: PIC's can output 20mA, so they can run straps directly
Start with a v3 Misc. CAN node:
- New, more reasonable size (same as recovery node?)
- linear 5V micropower power supply... LOW NOISE! Ferrites, filter caps, etc.
- PIC18F458 1/10MHz crystal (PLL'd to 40MHz = 10MIPS) running PicCore
- JED connectors for CAN and ICD2
- Don't forget taps for the CAN and UART.
Then for the magnetometer:
- HMC1001, HMC1002: tight holes to align the HMC1001 properly... HMC1002 alignment? (Not a huge deal, as calibration required. HMC1001 shouldn't vibrate.)
- Instrumentation op amps with appropriate gain and a single pole RC filter (f3dB = 100 Hz) Maybe get duals and use 4th as ~2.5V ref.
- From inst. amps right into the PIC's 10bit 55 ksps ADC (fsample >= 1 KHz... digital filtering?)
- Resistor divider 2.5V supply - requires a single op amp. 4th instr. op amp?
- Set/Reset circuit for each HMC part - or multiplex? -- Multiplex might be difficult, need ~4A currents.
- Make everything ratiometric - HMC, 2.5V reference, ADC. Might need absolute reference for offset strap current.
- Offset strap run off PIC 20mA pins. Use resistor to read current of offset strap.
- Thermistor on board to measure temp and trigger another calibration if delta T too high.
- Non magnetic components! Probably need to De-Gauss.
- Tin on ground planes... does it matter? No.