Summer 2011 Scoutfly Update¶

This page documents the problems identified and changes necessary during the summer 2011 update of the Scoutfly board. The board design referenced is in the files scoutfly.sch and scoutfly.brd in the scoutfly directory on the ECE repository (roboclub8.frc.ri.cmu.edu/home/svn/scouthw, not currently associated with this wiki). The material on this page was mostly compiled and organized from Kevin Woo's posts to the mailing list.

Design Problems¶

• +1.8V floats at 2.24 V when unloaded, but it returns to nominal under load
  • This is probably just the way the chips are supposed to work
• U$3 is generally faulty and should be replaced with another model
  • kwoo has the beginning of a schematic, which needs to be built and tested
  • U$9 (MOSFET attached to G2 pin of U$3) should be removed
    • The part isn't working correctly
    • We need a source of reliable parts or reevaluation of need
• Voltage and current cutoff resistor values (R6 and R26?) may be wrong
  • Probably why the battery charging chip is acting strangely
  • Do math for resistance values again
  • Switch to a simpler charging chip if possible
• 1.8 V shutdown not working
  • It should default to shutdown, but it doesn't

Changes¶

• Connect VREG_EN (EN pin on U$3, the pushbutton power controller) to VCC with 10 kOhm pull-up resistor
  • Done
• Remove R4 and connect pin 4 (~PB) on U$3 directly to VCC
  • Done
• Change R26 to 100 kOhms
  • Done
• Change wire and silkscreen label on user button 0 (S3) to be user button 2
  • Done
• Flip J3 (FFC connector) so that mating side faces out
  • Done
• Change U$28 (gyroscope) to part LPY5150AL
  • Done
• Update design date on silkscreen before next board spin
• Enlarge drill holes for mini-USB connector
• Make sure that all traces bends are 45 degrees
  • Was not mindful of this while making previous changes
• Add resistor to BOM_SIG0 or verify that it is not necessary

Power supply board¶

To save money while updating the power supply circuit (fuel gauge, battery charger, pushbutton controller, and related smaller components), a smaller board is being spun containing only the power supply components and an AVR. This board is not done yet, and a list of changes to be made follows.

• Fix charging circuit
  • Choose Schottky diode and rectifying diode for charging circuit and add them to schematic
  • Find a low-ESR inductor (< 100mOhm) for charging circuit
    • Calculated 4.7uH from equation in charger data sheet, assuming 31% ripple current and .45 V voltage drop over Schottky diode (reasonable assumptions?)
    • Such as RL622-4R7K-RC
    • MPI4040R3-4R7-R might also work, but it has 15% tolerance, which would give ripple current over 35% in the worst case
    • Make sure that the inductor is shielded to prevent spamming surroundings with magnetic field
      • MPI4040R3-4R7-R seems to be the only one on Digikey that satisfies the shielding requirement in addition to all the others
  • Check that value of capacitor on charger's RNG/SS port is appropriate
    • Kevin says it is. The equation in the data sheet indicates that this would give a soft startup time of 2 ms.
  • Make sure that sense resistor can handle power going through it
    • Max current should be when charging at 2A. P = I^2 * R = 2 A * 2 A * .01 Ohm = .04 W
    • 1/4 W resistor should do it
  • Tie CLP to Vin
  • Add decoupling capacitors to allow charger to power board by itself
• Pushbutton controller
  • Ensure that PWR_KILL is pulled up to VCC when not controlled by AVR
• Misc
  • Check AVR supply voltage/decide on LDO output voltage
  • Set up ISP header
  • Add pull-up resistors for reset line, decoupling capacitors for AVR
  • Add serial driver and USB port
  • Enable second LDO if it is needed
  • Tie battery ground to system ground
    • Make sure all current goes through gas gauge sense resistor
      • I don't understand how this works
    • Find a low-ESR, non-ceramic capacitor
    • Enlarge drill holes on USB connector