| Commit message (Collapse) | Author | Files | Lines |
|---|
|
This replaces "USE_CONFIGURABLE_RGB_VOLTAGE_LEVELS"
with "USE_AUX_THRESHOLD_CONFIG", which controls the brightness of
button LEDs while the main LEDs are on,
and during post-off voltage display.
Same basic concept, but works on single-color LEDs too,
and lets the user finally configure POVD thresholds.
The code for this is a bit messy, but the aux LED code as a whole
is pretty messy since it wasn't designed for the things it does now.
The entire thing needs a refactor or rewrite someday. But not today.
For now, this is just enough to make the pull request
cover more use cases before merging into trunk.
I've tested it on a variety of lights, but am not yet entirely
comfortable with it. However, it worked on at least these:
- 1-color button LED, no RGB
- front RGB, 1-color button LED
- front RGB, hardwired also to RGB button
- RGB button, no other aux
These may need extra changes,
and may have extra config options which do nothing...
- front RGB, no button LED
- 1-color front aux, no button LED
- no aux at all
- attiny85 lights (some could theoretically support the new options,
but none even try)
|
|
(the product name changed upstream)
|
|
- changed from RGB button to RGB front aux
- complete recalibration of voltage sensor
- had to change party strobe timing to make it work again
- still couldn't fix moon, it's flickery AF no matter what I do
|
|
After testing on every device I can, and getting several users to also
test this, it appears to reduce and sometimes completely eliminate
preflash on most devices... and the cases where it wasn't reported to
help, at least it didn't make things worse. Some units apparently just
can't get the flash eliminated completely, despite trying lots of things.
Instead of turning the chips on and then waiting 4ms, it now turns the
preflash absorber on, waits ~0.6ms, sets misc params, then turns the
boost chip on, then waits ~0.6ms, then turns the preflash absorber off.
This seems to work best on li-ion power, where on my devices it completely
eliminates any preflash. There is still a very mild flash on AA though,
which I wasn't able to get rid of. But it's like... 0.003 lm for just
a few milliseconds, really not bad. Even in the worst case reported by
a user, based on the video they took, it looks like just 0.01 lm for
a few milliseconds.
|
|
for some reason, and didn't commit... saving now to change branches,
but should delete this commit if it turns out there was no reason for it
|
|
- calibrated party strobe
- removed duplicate or commented-out code
- added a basic readme
|
|
- changed model number from 0281 to 0171
- cleaned up blink_negative and AUXLED_RGB_DIFFERENT_PORTS a little
(but the latter needs a complete refactor, as soon as the
hardware abstraction code can handle aux LEDs better)
- cleaned up USE_LONG_BLINK_FOR_NEGATIVE_SIGN a little
- removed USE_OTG_IN_MOMENTARY since it's not actually used
- moved hw/loneoceans/lume-x1-avr32dd20/* files into hw/hank/lume-x1/
- superficial cleanup on hank/lume-x1/hwdef.*
- removed some of the extra stuff from hank/lume-x1/anduril.h
- adjusted calibration (especially ramp table) on hank-lume-x1
(ramp shape is pretty close to a D4K-boost now, but with more firefly modes)
(calibration is based on a sample size of 1, further testing needed)
|
|
https://github.com/loneoceans/anduril/commit/d83ebb75dab8c462b7efa841bccc00a136ff15a2
The [PR](https://github.com/ToyKeeper/anduril/pull/37) has a lot of other
stuff in it, so I'm just picking out the parts needed for this particular
light, and leaving the rest for later.
Will need further edits before merging into trunk.
|
|
(they are now two separate build targets)
|
|
(because otherwise the light destroys its own LEDs at full power)
|
|
- battery voltage readings
- RGB button aux
- less moon flicker (but brighter, less efficient moon)
- reduced blink brightness
(old brightness was so high it was throwing off battery readings)
|
|
|
|
Adds two entries to the battery voltage settings menu, the first
isathreshold for switching aux to high, and the second sets a minimum
level for it to be displayed, also effectively allowing the feature to
be entirely disabled if not wanted.
|
|
|
|
|
|
(so 0 to 255 now goes from 0.00V to 5.10V)
|
|
(also, fixed bug where a totally empty li-ion didn't get limited)
|
|
and not letting the magic smoke out of updi adapters any more (probably)
The alkaline detection might be a little too lenient though; it
could potentially fail to activate limits when the cell is completely
full or stronger than an average alkaline. One of my test cells
measured at 72 / 75, so if it was just a little stronger it'd pass...
but most alkalines I tried were in the 40 to 60 range and failed easily.
OTOH, if I make it easier to fail, it's likely to trip on normal li-ion
cells, and I don't want that.
So as a future enhancement idea, maybe it should have a smaller sag
threshold for AA and a larger threshold for li-ion. That would reduce
false negatives for AA, while still preventing false positives for li-ion.
|
|
(saving progress here so I can work on a different branch)
|
|
|
|
|
|
- new ramp
- production style config defaults (simple mode, Hank config)
- candle tuning
- fixed way-too-fast thermal regulation
(might still be a bit fast, but it's a lot better)
|
|
|
|
|
|
For 3V LEDs on a boost/buck dual fuel driver, this allows
routing power around the boost converter in li-ion mode,
to reduce resistance and increase maximum output.
The AA/NiMH mode is unaffected, and boosts as normal.
|
|
Same as PL47G2 but with FET PWM levels safe for 219 emitters.
Same as difference between PL47 and PL47-219.
PL47G2 base allows low mode for aux LEDs, PL47 does not.
|
|
(it's an odd case with a 2 channel driver which only uses 1 set of LEDs)
|
|
|
|
|
|
This helps when using a regulator which doesn't like being turned off
and back on quickly.
|
|
|
|
|
|
|
|
- changed column order to (model, mcu, name)
- changed column sizes (auto-sized w/ 2 spaces between columns)
- made it handle hex digits in model numbers
- reserved 1900 to 2199 for years only, not model numbers
- noted gChart and thefreeman sharing a brand ID
|
|
|
|
The 0144 model number is reserved for the successor to the Meteor M44.
This is Hank's first AA light, so it's assigned as 0161:
- 01: Emisar
- 6: product line 6
- 1: model 1
|
|
|
|
(users complained, and extended simple UI probably needs to be off by
default in all factory builds anyway)
|
|
|
|
It had no minimum thermal stepdown defined, so it used MAX_7135 by default,
which was level 149/150 ... thus virtually no regulation.
I also made it easier to switch between internal and external temperature
sensors by just commenting out a few lines in the hwdef.h.
|
|
(the MCU's internal sensor works too, but external is a bit better I guess?)
|
|
- upgraded to DSM: lower lows, *much* smoother ramp
- made lows more efficient with underclocking
- fixed party strobe being too blurry
- calibrated UI speed / bogomips multiplier
- added readme to document this hardware's multiple quirks
|
|
strobe
|
|
|
|
(also enabled smooth steps on BLF GT, but had to remove stuff to make room)
|
|
|
|
- fixed t1616 Vref values getting clobbered sometimes,
wrapped setting those in a #define'd function for ease and consistency
- moved some DAC definitions from hw/ to arch/ to reduce repetition
- fixed thefreeman's other builds
- switched from PWM_TOPS to PWM2_LEVELS (I'm trying to phase out _TOPS)
|
|
(realtime voltage colors, and no simple UI by default)
|
|
internal volt unit
Before this branch, `voltage` was 6 bits: Volts * 10
A couple patches ago, I upgraded it to 16 bits: 65535 * Volts / 10.24
That costs too much extra ROM on attiny85 though, for extra precision
it doesn't even use... so I switched back to an 8-bit value. It's still
more precise than before though: Volts * 40
... and battcheck displays an extra digit now, on devices with ROM for it.
... and battcheck waits a second to get a more accurate measurement
before displaying the first value. It has *much* less variation
between first and later readings now.
Also:
- got t1616 builds working again (tested fc13 and thefreeman-boost-fwaa)
- upgraded t1616 voltage and temp to 12-bit (10 bits + 4x oversampling)
- removed expensive temp conversion from t1616 ADC interrupt
- recalibrated t1616 bogomips again; runs faster after interrupt fix
- increased t1616 internal VDD measurement resolution by 36% (1.5V Vref, not 1.1V)
- fixed sloppy setting of Vref bits
I still need to test / update other t1616 builds,
and fix all the t85 + t1634 code and build targets.
|
|
This patch changes the ADC code to use two internal standard units, and
everything else must convert to these units:
- FSM Volts: centiVolts << 6 (range 0 to 10.24 V per cell)
- FSM Kelvin: Kelvin << 6 (range 0 to 1024 K)
UI-level voltage is still "Volts * 10", and temperature is still Celsius.
FSM expects functions to be provided, to convert from the hardware's
raw ADC measurements to these internal units: `voltage_raw2cooked()`
and `temp_raw2cooked()`. Defaults will be provided by arch/*.[ch] for
each MCU type, or the hwdef can make its own.
Anyway, gotta go fix all the other MCUs and builds now. :(
|
