drivers: adc: sam0: Adjust resolution with the oversampling factor #75201
Conversation
|
Hello @Robibobo1, and thank you very much for your first pull request to the Zephyr project! |
zephyrbot
added
the
platform: Microchip SAM
zephyrbot
requested review from
attie-argentum,
mnkp,
nandojve,
pdgendt and
stephanosio
|
Sorry for my wrong commit, this is my first time contributing |
|
Check the contribution guidelines, I think you should squash all the commits, and avoid merge commits. |
Robibobo1
force-pushed
the
Oversampling
branch
2 times, most recently
from
1f2eb7f to
2fdc829
Compare
|
@anangl @nandojve please take a look @Robibobo1 please address the failed compliance check |
Robibobo1
force-pushed
the
Oversampling
branch
2 times, most recently
from
86cc5ff to
ec978ea
Compare
drivers/adc/adc_sam0.c
Outdated
There was a problem hiding this comment.
please make the commit title clear that this is specific to sam, ie it should start with drivers: adc_sam0:
nandojve
changed the title
|
Hi @attie-argentum , Could you check this change? |
|
This pull request has been marked as stale because it has been open (more than) 60 days with no activity. Remove the stale label or add a comment saying that you would like to have the label removed otherwise this pull request will automatically be closed in 14 days. Note, that you can always re-open a closed pull request at any time. |
The old SAM0 adc drivers forces you to put 12 bits when using oversampling. The adc_raw_to_millivolts_dt function is not working properly because if we uses an oversampling of 8, we get 16 bits of resolution. The function uses directly the resolution in the DT, so the voltage will not be correct. To counter this, i forced the user to put the right resolution for his oversampling factor. Fixes: zephyrproject-rtos#74607 Signed-off-by: Robin Carrupt <[email protected]>
Robibobo1
force-pushed
the
Oversampling
branch
from
ec978ea to
0e49a16
Compare
| LOG_ERR("Oversampling requires minimum 13 bit resolution"); | ||
| return -EINVAL; | ||
| } | ||
| ADC_RESSEL(adc) = ADC_RESSEL_12BIT; |
There was a problem hiding this comment.
The ADC resolution is 12bit max.
However, if you want to Accumulate, Average or Oversampling, see
Check below sections:
45.6.2.9 Accumulation
45.6.2.10 Averaging
45.6.2.11 Oversampling and Decimation
This change must work for all sam0 series.
This possible it related to #76963
|
This pull request has been marked as stale because it has been open (more than) 60 days with no activity. Remove the stale label or add a comment saying that you would like to have the label removed otherwise this pull request will automatically be closed in 14 days. Note, that you can always re-open a closed pull request at any time. |
|
This pull request has been marked as stale because it has been open (more than) 60 days with no activity. Remove the stale label or add a comment saying that you would like to have the label removed otherwise this pull request will automatically be closed in 14 days. Note, that you can always re-open a closed pull request at any time. |
9 participants
When using the oversampling factor on the SAMD21 family, the value that you get from the conversion is 16 bits instead of 12.
Using a factor of 8 adds 3 bits, a factor of 6 adds 2 bits, etc. I corrected this by dividing the factor by 2 and adding it to the resolution variable.
I didn't find other MCU families that use hardware oversampling. If there are others, we need to check if this is correct for that type of ADC.
Fixes: #74607
Signed-off-by: Robin Carrupt [email protected]