diff options
| author | Bill Richardson <wfrichar@chromium.org> | 2013-07-09 15:38:21 -0700 |
|---|---|---|
| committer | ChromeBot <chrome-bot@google.com> | 2013-07-09 18:56:36 -0700 |
| commit | d67f83a63cc0a03994a2cc5c0565c7b51f773816 (patch) | |
| tree | 65b4c8b8b866fea008c5d2a47070bada75488471 | |
| parent | 3bc27f0627637cae6e4edb8cd26716d46e5e67d2 (diff) | |
| download | chrome-ec-d67f83a63cc0a03994a2cc5c0565c7b51f773816.tar.gz | |
Use sample averaging on ADC to increase accuracy
The AC adapter detection isn't measuring the expected values, most likely
because we're not providing enough current for the ADC circuitry to sample.
This tells the ADC to take 64 samples for every reading and average them
together. That seems to fix it.
BUG=chrome-os-partner:19594
BRANCH=none
TEST=manual
To test, use a scope to probe the voltage pin 6 of the AC adapter connector.
Compare that with the measured result from the ADC, by using the EC
console's "adc" command.
Before this change:
Adapter Probed(mV) ADC(mV) error
45W 469 415 11.5%
65W 600 530 11.6%
90W 794 701 11.7%
After this change:
Adapter Probed(mV) ADC(mV) error
45W 469 452 3.6%
65W 600 590 1.6%
90W 794 773 2.6%
Close enough.
Change-Id: I5ab5526934a5a2711969a90a26631d54c7863da7
Signed-off-by: Bill Richardson <wfrichar@chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/61346
Reviewed-by: Randall Spangler <rspangler@chromium.org>
| -rw-r--r-- | chip/lm4/adc.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/chip/lm4/adc.c b/chip/lm4/adc.c index eb6f640680..8e17854104 100644 --- a/chip/lm4/adc.c +++ b/chip/lm4/adc.c @@ -229,6 +229,9 @@ static void adc_init(void) task_enable_irq(LM4_IRQ_ADC0_SS2); task_enable_irq(LM4_IRQ_ADC0_SS3); + /* 2**6 = 64x oversampling */ + LM4_ADC_ADCSAC = 6; + /* Initialize ADC sequencer */ for (i = 0; i < ADC_CH_COUNT; ++i) lm4_adc_configure(adc_channels + i); |