Powered nRF52832 from primary Lithium-Thionyl Chloride

[Francesco_M]

Dear,

we are developing a beacon product with nRF52832 powered from Li-SOCL2 primary battery. I understood that We cannot connect directly battery to the nRF52832 beacause the open voltage of the battery exceeds the maximum reccommended operating voltage Vdd of 3.6V (open circuit voltage of Li-SOCL2 battery is about 3.7V depending on temperature). NRF52832 is configured to use built-in internal DCDC converter and to stay in sleep mode with typical 1.5uA current consumption and periodically goes in TX mode (about 7-9mA for some msec. every 100msec.). To extend battery life we intend to powered the nRF52832 with a voltage level of about 2.5V or less (2.2V) therefore we need to use a regulator that reduce voltage from battery voltage (3.7V) to 2.5(2.2V). We need to select external LDO or DCDC (buck) converter. If we not considere the difference in term of cost between LDO and DCDC converter (LDO is much cheaper) I think that external DCDC converter is the right choice to minimize power consumption and extend battery life (a buck converter similar to TPS62840 from TI has very low quiescent current of about some hundred nA and high efficiency tipically 85% with very light load of about 1uA). Can you confirm this assumption?

Best Regards

[Turbo J]

Francesco_M said:

To extend battery life we intend to powered the nRF52832 with a voltage level of about 2.5V or less (2.2V) therefore we need to use a regulator that reduce voltage

That will not extend your battery life, unless there are other consumers present. Rememer: The NRF chip supports a DCDC switching mode regulator, and will thus consume less operating current at higher voltage.

After reading a datasheet from such Li-SOCL2 cell, I recommend to use the cell voltage directly. The voltage goes down to 3.6V as soon as there is anything used (CPU core, radio) - only the sleeping current might be a tiny bit higher than datasheet numbers, as this is the only state where the battery voltage can exeed 3.6 Volts. This is still "far enough" away from the 3.9V absolute maximum.

[Andreas]

Hi,

 

Recommended supply range is up to 3.6V, but absolutely highest is 3.9V, so at least in theory running directly off the battery should be fine. You should make sure though that the battery can deliver enough current for all practical use cases (e.g. radio + CPU) without the cell voltage dropping >100mV, absolutely not more than 300mV.

As mentioned by Turbo the nRF52 DCDC is most power efficient with a high drop over the regulator. With this, it might not be too bad to use an external LDO, from VBAT to say 3.3V or something.

 

Another solution could be to look at nRF52833 or nRF52840, these support up to 5.5V supply.

 

Best regards,

Andreas

[Francesco_M]

Dear Turbo, many thanks for your answer!

In our application the nRF52832 stays almost all the time in sleep mode (about 1.5uA current consumption) and therefore there is a large amount of time in which voltage applied to the nRF52832 exceeds max limit for Vdd. I found, in other posts, that if Vdd exceeds, for long time, the value of 3.6V, the SoC could be damaged. In my sentence "To extend battery life we intend to powered the nRF52832 with a voltage of about 2.5V or less (2.2V) therefore we need to use a regulator that reduce voltage" I mean that to use all the battery capacity, voltage at the input of regulator must be a value that is almost the value of cut-off voltage for Li-SOCL2 battery, i.e. the voltage for which the battery is completely depleted. With this consideration, I thought I cannot use regulator that reduce voltage from 3.6V to 3.3V or 3V beacause a these voltages the battery has still residual capacity.