summaryrefslogtreecommitdiff
path: root/ir
AgeCommit message (Collapse)AuthorFilesLines
2014-08-27ir/ir-ilim-z.pdf: add manually generated PDFWerner Almesberger1-0/+0
2014-08-27ir/ilim-z.sch: correct part number of R3 (from 3.3 Ohm to 1.5 Ohm)Werner Almesberger1-3/+2
2014-08-27ir/ir.tex: use ilim-z instead of ilim-ccWerner Almesberger1-2/+2
2014-08-27ir/ilim-z.sch: replaced voltage divider with Zener diodeWerner Almesberger1-0/+102
2014-08-27ir/ir.tex: define what "active" means in terms of lightWerner Almesberger1-4/+6
2014-08-27ir/ilim-cc.sch, ir-ilim-cc.pdf: tune simulated circuit to reduce power ↵Werner Almesberger2-12/+12
burned in R3 Before After R3 3.3 Ohm 1.8 Ohm R5 1 kOhm 470 Ohm R6 10 kOhm 1.5 kOhm C1 33 uF 22 uF
2014-08-27ir/ilim-cc.sch: swapped R4 and S1 to better reflect reality (suggested by Joerg)Werner Almesberger1-21/+21
2014-08-26ir/ir.tex: more correctionsWerner Almesberger2-52/+71
Some style improvements and typos. Larger changes: Section "Combined characteristics" - correct the lowest signal frequency: instead of 20 kHz, it can be as low as 0.9 kHz (IR-UART, 9.6 kbps, 8 bits with parity) Section "IR-UART mode" - correct description of TX/RX use - leave decision open whether LED current is always limited to <= continuous forward current or only constrained by the current-limiting circuit (i.e., one may get variable intensity during a transmission) Section "Mode summary" - new
2014-08-26ir/: prepare final version (still needs proof-reading)Werner Almesberger7-71/+198
Introduction: - resolved to call the device "GTA04b7", never "Neo900" (i.e., remove to do item, no text change) Section "IR modes" - indicate that SIR and IR-UART RX are not terribly important can could be dropped if they add too many difficulty Section "IR-UART" - change lowest "common" data rate from 19.2 kbps to 9.6 kbps Section "Received signal processing" - branch IR receiver to audio codec before high-pass filters - add high-pass filter at tentatively 10 Hz in IR-UART mode. The suggested range was 1-100 Hz, so I picked the middle. Incandescent lightbulbs flicker at 100-120 Hz, so even 100 Hz would be marginal for anything but daylight filtering yet could already interfere with low bit rates. - added drawing of filter selection and with output amplifier Section "CPU powered down" - specify that IR shell be off Section "Configuration" - assume that hackerbus doesn't need the ability to override the configuration of the IR subsystem (i.e., remove the "to do" item, no text change) Section "Reset" - in CPU reset state, bq27200 POR state selects IR-UART and non-POR state (i.e., GPIO explicitly set to "0") disables IR - leave use of bq.GPIO past CPU reset state to the implementor: "When the CPU has left its reset state and can actively control the IR subsystem, it is also able to change the state of bq.GPIO. bq.GPIO can therefore be used (or ignored) as the implementation sees fit." Added bq.GPIO with dashed lines to all IR configurations under CPU control. - confirmed: do not mention hypothetical debug messages from the ROM boot loader Section "IR deactivated" - add "and use as little power as possible for any filters and/or amplifiers." Section "Undriven signals" - specify that the IR circuit has to be tolerant against undriven UART signals but leave the rest to the implementation. Section "Overload protection" - rewrote, now with circuit simulation example
2014-07-25ir/ilim-cc.sch: improved current limiter that is tolerant to beta variationsWerner Almesberger1-0/+84
2014-07-25ir/ilim.sch, ilim-t.sch: add Qucs simulations of IR LED current limiterWerner Almesberger2-0/+125
2014-07-23ir/: draft version, for reviewWerner Almesberger5-55/+116
2014-07-23ir/: complete first draft (still needs proofreading and review)Werner Almesberger6-48/+468
2014-07-22ir/: draft of specification/description of the IR subsystemWerner Almesberger4-0/+529