Can ST25DV64k affect the internal temperature measurement of the host MCU

[ADSB]

Hello community,

I am trying to use the ST25dv64k with host MCU (stm32l031k6) to make a small project. I want to get a temperature measurement using the internal temperature sensor of the host MCU. However, I want to know if using ST25dvxxx actually affect the internal temperature sensor reading. I noticed that the values fluctuate when ever I try to read the tag with mobile phone or an industrial reader. The main reason I am asking this, is that I want to have a battery-less measurement of the temperature, but if the NFC field is causing the internal temperature reading to be useless, do you have any recommendation of cheap low power temperature sensor

Thank you for your help

[Ulysses HERNIOSUS]

​Hi ADSB,

Does the effect only depend on the reader field - even when  no ST25DV is connected. Maybe you want to check this.

With my software view: You could use the ST25DV to see if there is field (it has some FIELD_CHANGE feature) and only perform the temperature measurment when there is no field. So the MCU would need to cache the temperature and just return the value from the last valid measurement. But that would break your battery-less approach.

I also assume that the sensor of the L0 can only measure the junction temperature of the die which I expect to be increasing due to self-heating of the L0 by the time the measurement is executed. That time may also vary depening on the NFC reader operation. So I would experiment also to see if the effect is direct or actually a self-heating effect. Many thoughts / ideas....

But maybe some hardware guys can shed more light on the background of the hardware influence.

Regards, Ulysses

[JL. Lebon]

​Hello,

If internal L0 temperature sensor does not fit your needs, you can add an external temperature sensor.

ST has a wide thermal sensors offer, please see https://www.st.com/en/mems-and-sensors/temperature-sensors.html#products

For an example, you can have a look at the M24LR04E discovery board (https://www.st.com/content/st_com/en/products/evaluation-tools/product-evaluation-tools/st25-nfc-rfid-eval-tools/st25-nfc-rfid-eval-boards/m24lr-discovery.html#overview)

The M24LR04E discovery board is doing something very similar to what you want to do: it is a board with a micro (STM8), a NFC tag (M24LR04) an I2C temperature sensor (STTS751) and a LCD display. The M24LR04 is very similar to the St25DV (pin to pin compatible). The NFC reader is powering the tag, the micro, the sensor and the display using the Energy harvesting feature of the tag. You have source code and schematics available on ST web site.

Concerning the temperature of ST25DV itself, it may vary when a RF field is present, as it has to dissipate power provided by the RF field when it is too important. I don't have figures, but the temperature increase is quiet small with a smartphone as the RF fields power is not high.

Best regards.

[Brian TIDAL]

Hi,

the STM32L0 internal sensor will give the internal temperature of the STM32L0. This internal temperature may be different from the ambiant temperature. The activity of the various peripherals inside the micro controller will affect the internal temperature of the device. If you want a very precise ambiant temperature measurement , you may use an external temp sensor. Otherwise, you can imagine to keep the STM32L0 in STOP mode between each measurements, do the temp measurement just after wake up at low frequency and at 1.8V so that the internal temperature is very close to the ambiant temperature and enter in stop mode again.

Rgs

BT

[ADSB]

Hello Brian,

Thank you for answering my question. I am doing this already low frequency and voltage to do the measurement. I understand that inner temperature sensor is used more for measuring the MCU temperature not the ambient temperature. However, the problem that I am countering is not just a faulty measurement. it is literally garbage data (like -19 degC, 1.3degC, 55degC) while we are in room temperature, only when there is NFC field present. I want to know if the NFC field is the main issue, and if there is a way to mitigate it.

Thank you again for your help

[ADSB]

Hello,

Thank you again for your reply. I am trying to have as minimum components as possible that is why I want to use the embedded temperature sensor. In theory, I will expect the temperature will increase when the MCU is operating and when there is NFC field. However, I am encountering garbage measurement when there is NFC field, like very cold or very hot when there is NFC field, and it fluctuates.

[Brian TIDAL]

Hi,

do you mean that the temperature measurements are "correct" when no NFC field is present and inconsistant when the NFC field is present?

Can you share more details about your application, in particular what is the position of the antenna in regards with the MCU position? Would it make sense to try to add some shielding over the MCU? What is your VREF source?

Rgds

BT

[ADSB]

yes exactly,

When I have no NFC field, The measurements are correct, maybe a 2 degC higher, but consistent. When there is NFC field, some measurement are totally off and strange. I used a 3 volt coin cell battery, also I used a dc power generator as well. I got the same results

[Brian TIDAL · Accepted solution]

Hi,

if by your experiments, you observe that the temperature sensor is affected by the NFC field, I can only recommend to modify the placement of the antenna by few centimeters or to try some shielding over the STM32L0.

Rgds

BT.