[PATCH] vf610_adc: Fix internal temperature calculation

[PATCH] vf610_adc: Fix internal temperature calculation

[Bhuvanchandra DV]

There is an observed temperature difference of ~20°C with the
internal temperature reading and the temperature measured on
SoC package. Existing calculations consider the typical values
provided in datasheet. Those typical values are valid for
VREFH_ADC at 3.0V. Voltage at 25°C is different for different
VREFH_ADC voltages. With VREFH_ADC at 3.3V, voltage at 25°C is
0.699V. Hence update the VTEMP25 to 0.699V which gives ADCR@Temp25
as 867 and the final temperature readings differs with ~5°C from
the external readings.

Formula for finding ADCR@Temp25:
ADCR@Temp25 = (ADCR@Vdd * V@TEMP25 * 10) / VDDconv

ADCR@Vdd for 12-Bit ADC = 4095
VDDconv = VREFH_ADC * 10

VREFH_ADC@3.3V
ADCR@Temp25 = (4095 * .699 * 10) / 33
ADCR@Temp25 ~= 867

| VREFH_ADC | V@TEMP25 | VDDconv | ADCR@Temp25 |
|   3.0V    | 0.696mV  |    30   |     950     |
|   3.3V    | 0.699mV  |    33   |     867     |

Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
---
 drivers/iio/adc/vf610_adc.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c
index f4df2a7..e7abc13 100644
--- a/drivers/iio/adc/vf610_adc.c
+++ b/drivers/iio/adc/vf610_adc.c
@@ -103,6 +103,13 @@
 
 #define DEFAULT_SAMPLE_TIME		1000
 
+/* V at 25°C of 696 mV */
+#define VF610_VTEMP25_3V0		950
+/* V at 25°C of 699 mV */
+#define VF610_VTEMP25_3V3		867
+/* Typical sensor slope coefficient at all temperatures */
+#define VF610_TEMP_SLOPE_COEFF		1840
+
 enum clk_sel {
 	VF610_ADCIOC_BUSCLK_SET,
 	VF610_ADCIOC_ALTCLK_SET,
@@ -636,11 +643,13 @@ static int vf610_read_raw(struct iio_dev *indio_dev,
 			break;
 		case IIO_TEMP:
 			/*
-			* Calculate in degree Celsius times 1000
-			* Using sensor slope of 1.84 mV/°C and
-			* V at 25°C of 696 mV
-			*/
-			*val = 25000 - ((int)info->value - 864) * 1000000 / 1840;
+			 * Calculate in degree Celsius times 1000
+			 * Using the typical sensor slope of 1.84 mV/°C
+			 * and VREFH_ADC at 3.3V, V at 25°C of 699 mV
+			 */
+			*val = 25000 - ((int)info->value - VF610_VTEMP25_3V3) *
+					1000000 / VF610_TEMP_SLOPE_COEFF;
+
 			break;
 		default:
 			mutex_unlock(&indio_dev->mlock);
-- 
2.5.0

Re: [PATCH] vf610_adc: Fix internal temperature calculation

[Jonathan Cameron]

On 23/09/15 14:43, Bhuvanchandra DV wrote:
> There is an observed temperature difference of ~20°C with the
> internal temperature reading and the temperature measured on
> SoC package. Existing calculations consider the typical values
> provided in datasheet. Those typical values are valid for
> VREFH_ADC at 3.0V. Voltage at 25°C is different for different
> VREFH_ADC voltages. With VREFH_ADC at 3.3V, voltage at 25°C is
> 0.699V. Hence update the VTEMP25 to 0.699V which gives ADCR@Temp25
> as 867 and the final temperature readings differs with ~5°C from
> the external readings.
> 
> Formula for finding ADCR@Temp25:
> ADCR@Temp25 = (ADCR@Vdd * V@TEMP25 * 10) / VDDconv
> 
> ADCR@Vdd for 12-Bit ADC = 4095
> VDDconv = VREFH_ADC * 10
> 
> VREFH_ADC@3.3V
> ADCR@Temp25 = (4095 * .699 * 10) / 33
> ADCR@Temp25 ~= 867
> 
> | VREFH_ADC | V@TEMP25 | VDDconv | ADCR@Temp25 |
> |   3.0V    | 0.696mV  |    30   |     950     |
> |   3.3V    | 0.699mV  |    33   |     867     |
> 
> Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
Looks fine to me, but I'll need an Ack from Fugang on this one
as I don't know or have the part I'm afraid.

Jonathan
> ---
>  drivers/iio/adc/vf610_adc.c | 19 ++++++++++++++-----
>  1 file changed, 14 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c
> index f4df2a7..e7abc13 100644
> --- a/drivers/iio/adc/vf610_adc.c
> +++ b/drivers/iio/adc/vf610_adc.c
> @@ -103,6 +103,13 @@
>  
>  #define DEFAULT_SAMPLE_TIME		1000
>  
> +/* V at 25°C of 696 mV */
> +#define VF610_VTEMP25_3V0		950
> +/* V at 25°C of 699 mV */
> +#define VF610_VTEMP25_3V3		867
> +/* Typical sensor slope coefficient at all temperatures */
> +#define VF610_TEMP_SLOPE_COEFF		1840
> +
>  enum clk_sel {
>  	VF610_ADCIOC_BUSCLK_SET,
>  	VF610_ADCIOC_ALTCLK_SET,
> @@ -636,11 +643,13 @@ static int vf610_read_raw(struct iio_dev *indio_dev,
>  			break;
>  		case IIO_TEMP:
>  			/*
> -			* Calculate in degree Celsius times 1000
> -			* Using sensor slope of 1.84 mV/°C and
> -			* V at 25°C of 696 mV
> -			*/
> -			*val = 25000 - ((int)info->value - 864) * 1000000 / 1840;
> +			 * Calculate in degree Celsius times 1000
> +			 * Using the typical sensor slope of 1.84 mV/°C
> +			 * and VREFH_ADC at 3.3V, V at 25°C of 699 mV
> +			 */
> +			*val = 25000 - ((int)info->value - VF610_VTEMP25_3V3) *
> +					1000000 / VF610_TEMP_SLOPE_COEFF;
> +
>  			break;
>  		default:
>  			mutex_unlock(&indio_dev->mlock);
> 

Re: [PATCH] vf610_adc: Fix internal temperature calculation

[Bhuvanchandra]

Hi Fugang,

On 09/27/2015 09:23 PM, Jonathan Cameron wrote:

> On 23/09/15 14:43, Bhuvanchandra DV wrote:
>> There is an observed temperature difference of ~20°C with the
>> internal temperature reading and the temperature measured on
>> SoC package. Existing calculations consider the typical values
>> provided in datasheet. Those typical values are valid for
>> VREFH_ADC at 3.0V. Voltage at 25°C is different for different
>> VREFH_ADC voltages. With VREFH_ADC at 3.3V, voltage at 25°C is
>> 0.699V. Hence update the VTEMP25 to 0.699V which gives ADCR@Temp25
>> as 867 and the final temperature readings differs with ~5°C from
>> the external readings.
>>
>> Formula for finding ADCR@Temp25:
>> ADCR@Temp25 = (ADCR@Vdd * V@TEMP25 * 10) / VDDconv
>>
>> ADCR@Vdd for 12-Bit ADC = 4095
>> VDDconv = VREFH_ADC * 10
>>
>> VREFH_ADC@3.3V
>> ADCR@Temp25 = (4095 * .699 * 10) / 33
>> ADCR@Temp25 ~= 867
>>
>> | VREFH_ADC | V@TEMP25 | VDDconv | ADCR@Temp25 |
>> |   3.0V    | 0.696mV  |    30   |     950     |
>> |   3.3V    | 0.699mV  |    33   |     867     |
>>
>> Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
> Looks fine to me, but I'll need an Ack from Fugang on this one
> as I don't know or have the part I'm afraid.

Can you please give your ACK if the patch looks fine.

Best regards,
Bhuvan

>
> Jonathan
>> ---
>>   drivers/iio/adc/vf610_adc.c | 19 ++++++++++++++-----
>>   1 file changed, 14 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c
>> index f4df2a7..e7abc13 100644
>> --- a/drivers/iio/adc/vf610_adc.c
>> +++ b/drivers/iio/adc/vf610_adc.c
>> @@ -103,6 +103,13 @@
>>   
>>   #define DEFAULT_SAMPLE_TIME		1000
>>   
>> +/* V at 25°C of 696 mV */
>> +#define VF610_VTEMP25_3V0		950
>> +/* V at 25°C of 699 mV */
>> +#define VF610_VTEMP25_3V3		867
>> +/* Typical sensor slope coefficient at all temperatures */
>> +#define VF610_TEMP_SLOPE_COEFF		1840
>> +
>>   enum clk_sel {
>>   	VF610_ADCIOC_BUSCLK_SET,
>>   	VF610_ADCIOC_ALTCLK_SET,
>> @@ -636,11 +643,13 @@ static int vf610_read_raw(struct iio_dev *indio_dev,
>>   			break;
>>   		case IIO_TEMP:
>>   			/*
>> -			* Calculate in degree Celsius times 1000
>> -			* Using sensor slope of 1.84 mV/°C and
>> -			* V at 25°C of 696 mV
>> -			*/
>> -			*val = 25000 - ((int)info->value - 864) * 1000000 / 1840;
>> +			 * Calculate in degree Celsius times 1000
>> +			 * Using the typical sensor slope of 1.84 mV/°C
>> +			 * and VREFH_ADC at 3.3V, V at 25°C of 699 mV
>> +			 */
>> +			*val = 25000 - ((int)info->value - VF610_VTEMP25_3V3) *
>> +					1000000 / VF610_TEMP_SLOPE_COEFF;
>> +
>>   			break;
>>   		default:
>>   			mutex_unlock(&indio_dev->mlock);
>>

RE: [PATCH] vf610_adc: Fix internal temperature calculation

[Duan Andy]

RnJvbTogSm9uYXRoYW4gQ2FtZXJvbiA8amljMjNAa2VybmVsLm9yZz4gU2VudDogU3VuZGF5LCBT
ZXB0ZW1iZXIgMjcsIDIwMTUgMTE6NTMgUE0NCj4gVG86IEJodXZhbmNoYW5kcmEgRFY7IGxpbnV4
LWlpb0B2Z2VyLmtlcm5lbC5vcmcNCj4gQ2M6IHN0ZWZhbkBhZ25lci5jaDsgbWFpdHlzYW5jaGF5
YW5AZ21haWwuY29tOyBEdWFuIEZ1Z2FuZy1CMzg2MTE7DQo+IGtuYWFjay5oQGdteC5kZTsgbGFy
c0BtZXRhZm9vLmRlOyBwbWVlcndAcG1lZXJ3Lm5ldDsgc2hhd24uZ3VvQGxpbmFyby5vcmc7DQo+
IGxpbnV4LWtlcm5lbEB2Z2VyLmtlcm5lbC5vcmc7IER1YW4gRnVnYW5nLUIzODYxMQ0KPiBTdWJq
ZWN0OiBSZTogW1BBVENIXSB2ZjYxMF9hZGM6IEZpeCBpbnRlcm5hbCB0ZW1wZXJhdHVyZSBjYWxj
dWxhdGlvbg0KPiANCj4gT24gMjMvMDkvMTUgMTQ6NDMsIEJodXZhbmNoYW5kcmEgRFYgd3JvdGU6
DQo+ID4gVGhlcmUgaXMgYW4gb2JzZXJ2ZWQgdGVtcGVyYXR1cmUgZGlmZmVyZW5jZSBvZiB+MjDC
sEMgd2l0aCB0aGUgaW50ZXJuYWwNCj4gPiB0ZW1wZXJhdHVyZSByZWFkaW5nIGFuZCB0aGUgdGVt
cGVyYXR1cmUgbWVhc3VyZWQgb24gU29DIHBhY2thZ2UuDQo+ID4gRXhpc3RpbmcgY2FsY3VsYXRp
b25zIGNvbnNpZGVyIHRoZSB0eXBpY2FsIHZhbHVlcyBwcm92aWRlZCBpbg0KPiA+IGRhdGFzaGVl
dC4gVGhvc2UgdHlwaWNhbCB2YWx1ZXMgYXJlIHZhbGlkIGZvciBWUkVGSF9BREMgYXQgMy4wVi4N
Cj4gPiBWb2x0YWdlIGF0IDI1wrBDIGlzIGRpZmZlcmVudCBmb3IgZGlmZmVyZW50IFZSRUZIX0FE
QyB2b2x0YWdlcy4gV2l0aA0KPiA+IFZSRUZIX0FEQyBhdCAzLjNWLCB2b2x0YWdlIGF0IDI1wrBD
IGlzIDAuNjk5Vi4gSGVuY2UgdXBkYXRlIHRoZSBWVEVNUDI1DQo+ID4gdG8gMC42OTlWIHdoaWNo
IGdpdmVzIEFEQ1JAVGVtcDI1IGFzIDg2NyBhbmQgdGhlIGZpbmFsIHRlbXBlcmF0dXJlDQo+ID4g
cmVhZGluZ3MgZGlmZmVycyB3aXRoIH41wrBDIGZyb20gdGhlIGV4dGVybmFsIHJlYWRpbmdzLg0K
PiA+DQo+ID4gRm9ybXVsYSBmb3IgZmluZGluZyBBRENSQFRlbXAyNToNCj4gPiBBRENSQFRlbXAy
NSA9IChBRENSQFZkZCAqIFZAVEVNUDI1ICogMTApIC8gVkREY29udg0KPiA+DQo+ID4gQURDUkBW
ZGQgZm9yIDEyLUJpdCBBREMgPSA0MDk1DQo+ID4gVkREY29udiA9IFZSRUZIX0FEQyAqIDEwDQo+
ID4NCj4gPiBWUkVGSF9BRENAMy4zVg0KPiA+IEFEQ1JAVGVtcDI1ID0gKDQwOTUgKiAuNjk5ICog
MTApIC8gMzMNCj4gPiBBRENSQFRlbXAyNSB+PSA4NjcNCj4gPg0KPiA+IHwgVlJFRkhfQURDIHwg
VkBURU1QMjUgfCBWRERjb252IHwgQURDUkBUZW1wMjUgfA0KPiA+IHwgICAzLjBWICAgIHwgMC42
OTZtViAgfCAgICAzMCAgIHwgICAgIDk1MCAgICAgfA0KPiA+IHwgICAzLjNWICAgIHwgMC42OTlt
ViAgfCAgICAzMyAgIHwgICAgIDg2NyAgICAgfA0KPiA+DQo+ID4gU2lnbmVkLW9mZi1ieTogQmh1
dmFuY2hhbmRyYSBEViA8Ymh1dmFuY2hhbmRyYS5kdkB0b3JhZGV4LmNvbT4NCj4gTG9va3MgZmlu
ZSB0byBtZSwgYnV0IEknbGwgbmVlZCBhbiBBY2sgZnJvbSBGdWdhbmcgb24gdGhpcyBvbmUgYXMg
SSBkb24ndA0KPiBrbm93IG9yIGhhdmUgdGhlIHBhcnQgSSdtIGFmcmFpZC4NCj4gDQo+IEpvbmF0
aGFuDQoNClRoZSBwYXRjaCBpcyBmaW5lIGZvciBtZS4gU29ycnkgZm9yIHRoZSBsYXRlIHJlc3Bv
bnNlIGR1ZSB0byB2YWNhdGlvbi4NCg0KQWNrZWQtYnk6IEZ1Z2FuZyBEdWFuIDxCMzg2MTFAZnJl
ZXNjYWxlLmNvbT4NCg0KPiA+IC0tLQ0KPiA+ICBkcml2ZXJzL2lpby9hZGMvdmY2MTBfYWRjLmMg
fCAxOSArKysrKysrKysrKysrKy0tLS0tDQo+ID4gIDEgZmlsZSBjaGFuZ2VkLCAxNCBpbnNlcnRp
b25zKCspLCA1IGRlbGV0aW9ucygtKQ0KPiA+DQo+ID4gZGlmZiAtLWdpdCBhL2RyaXZlcnMvaWlv
L2FkYy92ZjYxMF9hZGMuYyBiL2RyaXZlcnMvaWlvL2FkYy92ZjYxMF9hZGMuYw0KPiA+IGluZGV4
IGY0ZGYyYTcuLmU3YWJjMTMgMTAwNjQ0DQo+ID4gLS0tIGEvZHJpdmVycy9paW8vYWRjL3ZmNjEw
X2FkYy5jDQo+ID4gKysrIGIvZHJpdmVycy9paW8vYWRjL3ZmNjEwX2FkYy5jDQo+ID4gQEAgLTEw
Myw2ICsxMDMsMTMgQEANCj4gPg0KPiA+ICAjZGVmaW5lIERFRkFVTFRfU0FNUExFX1RJTUUJCTEw
MDANCj4gPg0KPiA+ICsvKiBWIGF0IDI1wrBDIG9mIDY5NiBtViAqLw0KPiA+ICsjZGVmaW5lIFZG
NjEwX1ZURU1QMjVfM1YwCQk5NTANCj4gPiArLyogViBhdCAyNcKwQyBvZiA2OTkgbVYgKi8NCj4g
PiArI2RlZmluZSBWRjYxMF9WVEVNUDI1XzNWMwkJODY3DQo+ID4gKy8qIFR5cGljYWwgc2Vuc29y
IHNsb3BlIGNvZWZmaWNpZW50IGF0IGFsbCB0ZW1wZXJhdHVyZXMgKi8NCj4gPiArI2RlZmluZSBW
RjYxMF9URU1QX1NMT1BFX0NPRUZGCQkxODQwDQo+ID4gKw0KPiA+ICBlbnVtIGNsa19zZWwgew0K
PiA+ICAJVkY2MTBfQURDSU9DX0JVU0NMS19TRVQsDQo+ID4gIAlWRjYxMF9BRENJT0NfQUxUQ0xL
X1NFVCwNCj4gPiBAQCAtNjM2LDExICs2NDMsMTMgQEAgc3RhdGljIGludCB2ZjYxMF9yZWFkX3Jh
dyhzdHJ1Y3QgaWlvX2Rldg0KPiAqaW5kaW9fZGV2LA0KPiA+ICAJCQlicmVhazsNCj4gPiAgCQlj
YXNlIElJT19URU1QOg0KPiA+ICAJCQkvKg0KPiA+IC0JCQkqIENhbGN1bGF0ZSBpbiBkZWdyZWUg
Q2Vsc2l1cyB0aW1lcyAxMDAwDQo+ID4gLQkJCSogVXNpbmcgc2Vuc29yIHNsb3BlIG9mIDEuODQg
bVYvwrBDIGFuZA0KPiA+IC0JCQkqIFYgYXQgMjXCsEMgb2YgNjk2IG1WDQo+ID4gLQkJCSovDQo+
ID4gLQkJCSp2YWwgPSAyNTAwMCAtICgoaW50KWluZm8tPnZhbHVlIC0gODY0KSAqIDEwMDAwMDAg
Lw0KPiAxODQwOw0KPiA+ICsJCQkgKiBDYWxjdWxhdGUgaW4gZGVncmVlIENlbHNpdXMgdGltZXMg
MTAwMA0KPiA+ICsJCQkgKiBVc2luZyB0aGUgdHlwaWNhbCBzZW5zb3Igc2xvcGUgb2YgMS44NCBt
Vi/CsEMNCj4gPiArCQkJICogYW5kIFZSRUZIX0FEQyBhdCAzLjNWLCBWIGF0IDI1wrBDIG9mIDY5
OSBtVg0KPiA+ICsJCQkgKi8NCj4gPiArCQkJKnZhbCA9IDI1MDAwIC0gKChpbnQpaW5mby0+dmFs
dWUgLSBWRjYxMF9WVEVNUDI1XzNWMykgKg0KPiA+ICsJCQkJCTEwMDAwMDAgLyBWRjYxMF9URU1Q
X1NMT1BFX0NPRUZGOw0KPiA+ICsNCj4gPiAgCQkJYnJlYWs7DQo+ID4gIAkJZGVmYXVsdDoNCj4g
PiAgCQkJbXV0ZXhfdW5sb2NrKCZpbmRpb19kZXYtPm1sb2NrKTsNCj4gPg0K

Re: [PATCH] vf610_adc: Fix internal temperature calculation

[Stefan Agner]

Hi Bhuvan,

On 2015-09-23 06:43, Bhuvanchandra DV wrote:
> There is an observed temperature difference of ~20°C with the
> internal temperature reading and the temperature measured on
> SoC package. Existing calculations consider the typical values
> provided in datasheet. Those typical values are valid for
> VREFH_ADC at 3.0V. Voltage at 25°C is different for different
> VREFH_ADC voltages. With VREFH_ADC at 3.3V, voltage at 25°C is
> 0.699V. Hence update the VTEMP25 to 0.699V which gives ADCR@Temp25
> as 867 and the final temperature readings differs with ~5°C from
> the external readings.

I am not entirely happy with the message. The proposed changes seem not
to alter the offset which would explain a ~20°C difference... Where do
this 20°C come from?

--
Stefan

> 
> Formula for finding ADCR@Temp25:
> ADCR@Temp25 = (ADCR@Vdd * V@TEMP25 * 10) / VDDconv
> 
> ADCR@Vdd for 12-Bit ADC = 4095
> VDDconv = VREFH_ADC * 10
> 
> VREFH_ADC@3.3V
> ADCR@Temp25 = (4095 * .699 * 10) / 33
> ADCR@Temp25 ~= 867
> 
> | VREFH_ADC | V@TEMP25 | VDDconv | ADCR@Temp25 |
> |   3.0V    | 0.696mV  |    30   |     950     |
> |   3.3V    | 0.699mV  |    33   |     867     |
> 
> Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
> ---
>  drivers/iio/adc/vf610_adc.c | 19 ++++++++++++++-----
>  1 file changed, 14 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c
> index f4df2a7..e7abc13 100644
> --- a/drivers/iio/adc/vf610_adc.c
> +++ b/drivers/iio/adc/vf610_adc.c
> @@ -103,6 +103,13 @@
>  
>  #define DEFAULT_SAMPLE_TIME		1000
>  
> +/* V at 25°C of 696 mV */
> +#define VF610_VTEMP25_3V0		950
> +/* V at 25°C of 699 mV */
> +#define VF610_VTEMP25_3V3		867
> +/* Typical sensor slope coefficient at all temperatures */
> +#define VF610_TEMP_SLOPE_COEFF		1840
> +
>  enum clk_sel {
>  	VF610_ADCIOC_BUSCLK_SET,
>  	VF610_ADCIOC_ALTCLK_SET,
> @@ -636,11 +643,13 @@ static int vf610_read_raw(struct iio_dev *indio_dev,
>  			break;
>  		case IIO_TEMP:
>  			/*
> -			* Calculate in degree Celsius times 1000
> -			* Using sensor slope of 1.84 mV/°C and
> -			* V at 25°C of 696 mV
> -			*/
> -			*val = 25000 - ((int)info->value - 864) * 1000000 / 1840;
> +			 * Calculate in degree Celsius times 1000
> +			 * Using the typical sensor slope of 1.84 mV/°C
> +			 * and VREFH_ADC at 3.3V, V at 25°C of 699 mV
> +			 */
> +			*val = 25000 - ((int)info->value - VF610_VTEMP25_3V3) *
> +					1000000 / VF610_TEMP_SLOPE_COEFF;
> +
>  			break;
>  		default:
>  			mutex_unlock(&indio_dev->mlock);

Re: [PATCH] vf610_adc: Fix internal temperature calculation

[Jonathan Cameron]

On 08/10/15 03:01, Duan Andy wrote:
> From: Jonathan Cameron <jic23@kernel.org> Sent: Sunday, September 27, 2015 11:53 PM
>> To: Bhuvanchandra DV; linux-iio@vger.kernel.org
>> Cc: stefan@agner.ch; maitysanchayan@gmail.com; Duan Fugang-B38611;
>> knaack.h@gmx.de; lars@metafoo.de; pmeerw@pmeerw.net; shawn.guo@linaro.org;
>> linux-kernel@vger.kernel.org; Duan Fugang-B38611
>> Subject: Re: [PATCH] vf610_adc: Fix internal temperature calculation
>>
>> On 23/09/15 14:43, Bhuvanchandra DV wrote:
>>> There is an observed temperature difference of ~20°C with the internal
>>> temperature reading and the temperature measured on SoC package.
>>> Existing calculations consider the typical values provided in
>>> datasheet. Those typical values are valid for VREFH_ADC at 3.0V.
>>> Voltage at 25°C is different for different VREFH_ADC voltages. With
>>> VREFH_ADC at 3.3V, voltage at 25°C is 0.699V. Hence update the VTEMP25
>>> to 0.699V which gives ADCR@Temp25 as 867 and the final temperature
>>> readings differs with ~5°C from the external readings.
>>>
>>> Formula for finding ADCR@Temp25:
>>> ADCR@Temp25 = (ADCR@Vdd * V@TEMP25 * 10) / VDDconv
>>>
>>> ADCR@Vdd for 12-Bit ADC = 4095
>>> VDDconv = VREFH_ADC * 10
>>>
>>> VREFH_ADC@3.3V
>>> ADCR@Temp25 = (4095 * .699 * 10) / 33
>>> ADCR@Temp25 ~= 867
>>>
>>> | VREFH_ADC | V@TEMP25 | VDDconv | ADCR@Temp25 |
>>> |   3.0V    | 0.696mV  |    30   |     950     |
>>> |   3.3V    | 0.699mV  |    33   |     867     |
>>>
>>> Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
>> Looks fine to me, but I'll need an Ack from Fugang on this one as I don't
>> know or have the part I'm afraid.
>>
>> Jonathan
> 
> The patch is fine for me. Sorry for the late response due to vacation.
> 
> Acked-by: Fugang Duan <B38611@freescale.com>
Applied to the fixes-togreg branch of iio.git and marked for stable.
Thanks,
Jonathan
> 
>>> ---
>>>  drivers/iio/adc/vf610_adc.c | 19 ++++++++++++++-----
>>>  1 file changed, 14 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c
>>> index f4df2a7..e7abc13 100644
>>> --- a/drivers/iio/adc/vf610_adc.c
>>> +++ b/drivers/iio/adc/vf610_adc.c
>>> @@ -103,6 +103,13 @@
>>>
>>>  #define DEFAULT_SAMPLE_TIME		1000
>>>
>>> +/* V at 25°C of 696 mV */
>>> +#define VF610_VTEMP25_3V0		950
>>> +/* V at 25°C of 699 mV */
>>> +#define VF610_VTEMP25_3V3		867
>>> +/* Typical sensor slope coefficient at all temperatures */
>>> +#define VF610_TEMP_SLOPE_COEFF		1840
>>> +
>>>  enum clk_sel {
>>>  	VF610_ADCIOC_BUSCLK_SET,
>>>  	VF610_ADCIOC_ALTCLK_SET,
>>> @@ -636,11 +643,13 @@ static int vf610_read_raw(struct iio_dev
>> *indio_dev,
>>>  			break;
>>>  		case IIO_TEMP:
>>>  			/*
>>> -			* Calculate in degree Celsius times 1000
>>> -			* Using sensor slope of 1.84 mV/°C and
>>> -			* V at 25°C of 696 mV
>>> -			*/
>>> -			*val = 25000 - ((int)info->value - 864) * 1000000 /
>> 1840;
>>> +			 * Calculate in degree Celsius times 1000
>>> +			 * Using the typical sensor slope of 1.84 mV/°C
>>> +			 * and VREFH_ADC at 3.3V, V at 25°C of 699 mV
>>> +			 */
>>> +			*val = 25000 - ((int)info->value - VF610_VTEMP25_3V3) *
>>> +					1000000 / VF610_TEMP_SLOPE_COEFF;
>>> +
>>>  			break;
>>>  		default:
>>>  			mutex_unlock(&indio_dev->mlock);
>>>

Re: [PATCH] vf610_adc: Fix internal temperature calculation

[Jonathan Cameron]

On 11/10/15 15:05, Jonathan Cameron wrote:
> On 08/10/15 03:01, Duan Andy wrote:
>> From: Jonathan Cameron <jic23@kernel.org> Sent: Sunday, September 27, 2015 11:53 PM
>>> To: Bhuvanchandra DV; linux-iio@vger.kernel.org
>>> Cc: stefan@agner.ch; maitysanchayan@gmail.com; Duan Fugang-B38611;
>>> knaack.h@gmx.de; lars@metafoo.de; pmeerw@pmeerw.net; shawn.guo@linaro.org;
>>> linux-kernel@vger.kernel.org; Duan Fugang-B38611
>>> Subject: Re: [PATCH] vf610_adc: Fix internal temperature calculation
>>>
>>> On 23/09/15 14:43, Bhuvanchandra DV wrote:
>>>> There is an observed temperature difference of ~20°C with the internal
>>>> temperature reading and the temperature measured on SoC package.
>>>> Existing calculations consider the typical values provided in
>>>> datasheet. Those typical values are valid for VREFH_ADC at 3.0V.
>>>> Voltage at 25°C is different for different VREFH_ADC voltages. With
>>>> VREFH_ADC at 3.3V, voltage at 25°C is 0.699V. Hence update the VTEMP25
>>>> to 0.699V which gives ADCR@Temp25 as 867 and the final temperature
>>>> readings differs with ~5°C from the external readings.
>>>>
>>>> Formula for finding ADCR@Temp25:
>>>> ADCR@Temp25 = (ADCR@Vdd * V@TEMP25 * 10) / VDDconv
>>>>
>>>> ADCR@Vdd for 12-Bit ADC = 4095
>>>> VDDconv = VREFH_ADC * 10
>>>>
>>>> VREFH_ADC@3.3V
>>>> ADCR@Temp25 = (4095 * .699 * 10) / 33
>>>> ADCR@Temp25 ~= 867
>>>>
>>>> | VREFH_ADC | V@TEMP25 | VDDconv | ADCR@Temp25 |
>>>> |   3.0V    | 0.696mV  |    30   |     950     |
>>>> |   3.3V    | 0.699mV  |    33   |     867     |
>>>>
>>>> Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
>>> Looks fine to me, but I'll need an Ack from Fugang on this one as I don't
>>> know or have the part I'm afraid.
>>>
>>> Jonathan
>>
>> The patch is fine for me. Sorry for the late response due to vacation.
>>
>> Acked-by: Fugang Duan <B38611@freescale.com>
> Applied to the fixes-togreg branch of iio.git and marked for stable.
> Thanks,
> Jonathan
Actually, change of plan.  I'll hold this until Stefan's query is answered.
(naughty me - I hadn't read the whole thread!)
>>
>>>> ---
>>>>  drivers/iio/adc/vf610_adc.c | 19 ++++++++++++++-----
>>>>  1 file changed, 14 insertions(+), 5 deletions(-)
>>>>
>>>> diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c
>>>> index f4df2a7..e7abc13 100644
>>>> --- a/drivers/iio/adc/vf610_adc.c
>>>> +++ b/drivers/iio/adc/vf610_adc.c
>>>> @@ -103,6 +103,13 @@
>>>>
>>>>  #define DEFAULT_SAMPLE_TIME		1000
>>>>
>>>> +/* V at 25°C of 696 mV */
>>>> +#define VF610_VTEMP25_3V0		950
>>>> +/* V at 25°C of 699 mV */
>>>> +#define VF610_VTEMP25_3V3		867
>>>> +/* Typical sensor slope coefficient at all temperatures */
>>>> +#define VF610_TEMP_SLOPE_COEFF		1840
>>>> +
>>>>  enum clk_sel {
>>>>  	VF610_ADCIOC_BUSCLK_SET,
>>>>  	VF610_ADCIOC_ALTCLK_SET,
>>>> @@ -636,11 +643,13 @@ static int vf610_read_raw(struct iio_dev
>>> *indio_dev,
>>>>  			break;
>>>>  		case IIO_TEMP:
>>>>  			/*
>>>> -			* Calculate in degree Celsius times 1000
>>>> -			* Using sensor slope of 1.84 mV/°C and
>>>> -			* V at 25°C of 696 mV
>>>> -			*/
>>>> -			*val = 25000 - ((int)info->value - 864) * 1000000 /
>>> 1840;
>>>> +			 * Calculate in degree Celsius times 1000
>>>> +			 * Using the typical sensor slope of 1.84 mV/°C
>>>> +			 * and VREFH_ADC at 3.3V, V at 25°C of 699 mV
>>>> +			 */
>>>> +			*val = 25000 - ((int)info->value - VF610_VTEMP25_3V3) *
>>>> +					1000000 / VF610_TEMP_SLOPE_COEFF;
>>>> +
>>>>  			break;
>>>>  		default:
>>>>  			mutex_unlock(&indio_dev->mlock);
>>>>
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-iio" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [PATCH] vf610_adc: Fix internal temperature calculation

[Bhuvanchandra]

Hi Stefan,

Sorry! for the delay.

On 10/08/2015 11:46 AM, Stefan Agner wrote:

> Hi Bhuvan,
>
> On 2015-09-23 06:43, Bhuvanchandra DV wrote:
>> There is an observed temperature difference of ~20°C with the
>> internal temperature reading and the temperature measured on
>> SoC package. Existing calculations consider the typical values
>> provided in datasheet. Those typical values are valid for
>> VREFH_ADC at 3.0V. Voltage at 25°C is different for different
>> VREFH_ADC voltages. With VREFH_ADC at 3.3V, voltage at 25°C is
>> 0.699V. Hence update the VTEMP25 to 0.699V which gives ADCR@Temp25
>> as 867 and the final temperature readings differs with ~5°C from
>> the external readings.
> I am not entirely happy with the message. The proposed changes seem not
> to alter the offset which would explain a ~20°C difference... Where do
> this 20°C come from?

The ~20°C difference is observed at higher temperatures when VTEMP25
at VREDH_ADC of 3V0 is considered. Will update the comment according
to the updated VTEMP25 constant instead of the 20°C difference which
is confusing wrt the changes done here.

Best regards,
Bhuvan

> --
> Stefan
>
>> Formula for finding ADCR@Temp25:
>> ADCR@Temp25 = (ADCR@Vdd * V@TEMP25 * 10) / VDDconv
>>
>> ADCR@Vdd for 12-Bit ADC = 4095
>> VDDconv = VREFH_ADC * 10
>>
>> VREFH_ADC@3.3V
>> ADCR@Temp25 = (4095 * .699 * 10) / 33
>> ADCR@Temp25 ~= 867
>>
>> | VREFH_ADC | V@TEMP25 | VDDconv | ADCR@Temp25 |
>> |   3.0V    | 0.696mV  |    30   |     950     |
>> |   3.3V    | 0.699mV  |    33   |     867     |
>>
>> Signed-off-by: Bhuvanchandra DV<bhuvanchandra.dv@toradex.com>
>> ---
>>   drivers/iio/adc/vf610_adc.c | 19 ++++++++++++++-----
>>   1 file changed, 14 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c
>> index f4df2a7..e7abc13 100644
>> --- a/drivers/iio/adc/vf610_adc.c
>> +++ b/drivers/iio/adc/vf610_adc.c
>> @@ -103,6 +103,13 @@
>>   
>>   #define DEFAULT_SAMPLE_TIME		1000
>>   
>> +/* V at 25°C of 696 mV */
>> +#define VF610_VTEMP25_3V0		950
>> +/* V at 25°C of 699 mV */
>> +#define VF610_VTEMP25_3V3		867
>> +/* Typical sensor slope coefficient at all temperatures */
>> +#define VF610_TEMP_SLOPE_COEFF		1840
>> +
>>   enum clk_sel {
>>   	VF610_ADCIOC_BUSCLK_SET,
>>   	VF610_ADCIOC_ALTCLK_SET,
>> @@ -636,11 +643,13 @@ static int vf610_read_raw(struct iio_dev *indio_dev,
>>   			break;
>>   		case IIO_TEMP:
>>   			/*
>> -			* Calculate in degree Celsius times 1000
>> -			* Using sensor slope of 1.84 mV/°C and
>> -			* V at 25°C of 696 mV
>> -			*/
>> -			*val = 25000 - ((int)info->value - 864) * 1000000 / 1840;
>> +			 * Calculate in degree Celsius times 1000
>> +			 * Using the typical sensor slope of 1.84 mV/°C
>> +			 * and VREFH_ADC at 3.3V, V at 25°C of 699 mV
>> +			 */
>> +			*val = 25000 - ((int)info->value - VF610_VTEMP25_3V3) *
>> +					1000000 / VF610_TEMP_SLOPE_COEFF;
>> +
>>   			break;
>>   		default:
>>   			mutex_unlock(&indio_dev->mlock);