[PATCH 0/2] iio: ak8975: Fix calculation formula and add device name

[PATCH 0/2] iio: ak8975: Fix calculation formula and add device name

[Beomho Seo]

 Three months ago Lars-Peter revised description of device driver[1].
Correctly, device's sensitivity is 0.3 uT/LSB typ and One micro tesla equal 0.01 gauss[2].
So I have fixed calculation formula,And then add RAW_TO_GAUSS macro.
Scale are returned as VAL_INT_PLUS_MICRO.
Additionally, I have add device name.

Tested on exynos4412 board, with following result:

-sh-4.1# cat name
ak8975
-sh-4.1# cat in_magn_x_raw
80
-sh-4.1# cat in_magn_x_scale
0.003609

Calculate the result 80 x 0.003609 = 0.28872 Gauss.
0.28872 G equal 288.72 mG

I think this magnitude value is proper level[3].

[1] http://marc.info/?l=linux-iio&m=138220306822747&w=2
[2] http://en.wikipedia.org/wiki/Gauss_(unit)
[3] http://en.wikipedia.org/wiki/Orders_of_magnitude_(magnetic_field)

Beomho Seo (2):
  iio: ak8975: Fix calculation formula for microtesla to gauss unit
  iio: ak8975: Add device name

 drivers/iio/magnetometer/ak8975.c |   17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

-- 
Best Regards,

Re: [PATCH 0/2] iio: ak8975: Fix calculation formula and add device name

[Peter Meerwald]

Hello,

>  Three months ago Lars-Peter revised description of device driver[1].

me, not Lars-Peter

> Correctly, device's sensitivity is 0.3 uT/LSB typ and One micro tesla equal 0.01 gauss[2].
> So I have fixed calculation formula,And then add RAW_TO_GAUSS macro.
> Scale are returned as VAL_INT_PLUS_MICRO.
> Additionally, I have add device name.

+#define RAW_TO_GAUSS(asa) ((((asa) + 128) * 3000) / 256)

  *
  * Since 1uT = 0.01 gauss, our final scale factor becomes:
  *
- * Hadj = H * ((ASA + 128) / 256) * 3/10 * 100
- * Hadj = H * ((ASA + 128) * 30 / 256
+ * Hadj = H * ((ASA + 128) / 256) * 3/10 * 1/100
+ * Hadj = H * ((ASA + 128) * 0.003) / 256
  *
  * Since ASA doesn't change, we cache the resultant scale factor into the
  * device context in ak8975_setup().
  */
-       data->raw_to_gauss[0] = ((data->asa[0] + 128) * 30) >> 8;
+       data->raw_to_gauss[0] = RAW_TO_GAUSS(data->asa[0]);

thank you for following up on this; I don't have the hardware, so I didn't 
dare to change the driver...

the RAW_TO_GAUSS() macro doesn't match the comment Hadj = H * ... anymore

maybe put this explanation in the patch comment, not the cover letter?

regards, p.

-- 

Peter Meerwald
+43-664-2444418 (mobile)

Re: [PATCH 0/2] iio: ak8975: Fix calculation formula and add device name

[Beomho Seo]

On 01/29/2014 05:53 PM, Peter Meerwald wrote:
> Hello,
> 
>>  Three months ago Lars-Peter revised description of device driver[1].
> 
> me, not Lars-Peter
> 

I'm really sorry, my mistake.

>> Correctly, device's sensitivity is 0.3 uT/LSB typ and One micro tesla equal 0.01 gauss[2].
>> So I have fixed calculation formula,And then add RAW_TO_GAUSS macro.
>> Scale are returned as VAL_INT_PLUS_MICRO.
>> Additionally, I have add device name.
> 
> +#define RAW_TO_GAUSS(asa) ((((asa) + 128) * 3000) / 256)
> 
>   *
>   * Since 1uT = 0.01 gauss, our final scale factor becomes:
>   *
> - * Hadj = H * ((ASA + 128) / 256) * 3/10 * 100
> - * Hadj = H * ((ASA + 128) * 30 / 256
> + * Hadj = H * ((ASA + 128) / 256) * 3/10 * 1/100
> + * Hadj = H * ((ASA + 128) * 0.003) / 256
>   *
>   * Since ASA doesn't change, we cache the resultant scale factor into the
>   * device context in ak8975_setup().
>   */
> -       data->raw_to_gauss[0] = ((data->asa[0] + 128) * 30) >> 8;
> +       data->raw_to_gauss[0] = RAW_TO_GAUSS(data->asa[0]);
> 
> thank you for following up on this; I don't have the hardware, so I didn't 
> dare to change the driver...
> 
> the RAW_TO_GAUSS() macro doesn't match the comment Hadj = H * ... anymore
> 
> maybe put this explanation in the patch comment, not the cover letter?
> 

OK. I will resend explanation in the patch comment.

> regards, p.
> 


-- 
Best Regards,

Beomho Seo, Assistant Engineer
System S/W Lab., Software Center, Samsung Electronics