Depending on the nRF52832 supply voltage, can the quiescent current vary?~

[SHMoon]

hello.

We are considering an external LDO that supplies VDD to the nRF52832.

I'm reviewing an LDO called TPS7A02, but there's a phenomenon I don't understand.

When using TPS7A0228 (LDO output 2.8V), the standby current of the board with nRF52832 is measured to be very low, at 1uA ~ 20uA.

On the other hand, when using TPS7A0230 (LDO output 3.0V), 20uA is added and 21uA ~ 40uA is measured.

TPS7A0228 : https://www.mouser.kr/ProductDetail/Texas-Instruments/TPS7A0228PDBVR?qs=sGAEpiMZZMug9GoBKXZ751mAF56Mq1BpFfw3x6s4ZMIdLDEJ2sm0Ow%3D%3D

TPS7A0230 : https://www.mouser.kr/ProductDetail/Texas-Instruments/TPS7A0230PDBVR?qs=sGAEpiMZZMug9GoBKXZ751mAF56Mq1BpAesZnDPtYLWRF%2FNPd%2FJnPQ%3D%3D

There is no current difference in the TPS7A02 LDO Datasheet.

Does the standby current consumed by the nRF52832 vary depending on the VDD voltage (LDO output)?

[hmolesworth]

TPS7A02 Iq is indeed flat and unvarying in Dropout with increasing input voltage, but when not in Dropout Iq increases significantly with input voltage, though it is difficult to estimate the exact increase

Edit: Not sure about the two conflicting graphs; Ven =1.1V above shows the increase but Ven=Vin below shows the flat Iq value which I thought was in Dropout but maybe not so; maybe ensure Ven=Vin to test this as if the Enable pin is driven by the nRF52 it will be less than Vin

Official Iq data vs input voltage for nRF52 internal LDO is not released, I don't think; easy to measure though ..

[hmolesworth]

I just measured the standby current on a bare nRF52832 (no port i/o pins connected to anything) at both 2.800 volts and 3.000 volts, no change in current measured with a 5-digit meter with 10nA resolution:

 1.12uA  at 2.8 and 3.0 volts - some errata applied in SystemInit()

 1.01uA  at 2.8 and 3.0 volts - no errata applied

I should add that at this low level of current the hysteretic mode of the LDO is enabled, and there is no information that I know of regarding the current at which LDO hysteretic mode is disabled and normal continuous LDO mode takes over (this is not related to DC/DC mode, which is not enabled in this test).

[hmolesworth]

I forced the LDO into normal continuous LDO mode by driving output ports low against the internal pull-up using high drive H0H1 and find there is no increase when the consumption is a steady 1mA approx (5 output ports driven low against internal 13.33k pull-up). Since current into such a constant resistance depends on voltage, lowering the voltage lowers the current, but it is as expected with no change in LDO Iq.

#define NUMBER_PINS_TO_DRIVE 5 //32
static void TestPower(void)
{
   // Set up to 32 output pins
   for (uint32_t i=0; i<NUMBER_PINS_TO_DRIVE; i++)
   {
        // Configuration      Direction    Input            Pullup       Drive Level      Sense Level
        // ===============    ==========   ==============   ==========   ==============   =============
        NRF_P0->PIN_CNF[i] = (PIN_OUTPUT | PIN_DISCONNECT | PIN_PULLUP | PIN_DRIVE_H0H1 | PIN_SENSE_OFF);
        NRF_P0->OUTCLR = (1UL << i);
   }
   // Sleep
   while (1)
      {
         __WFE();
         __NOP(); __NOP(); __NOP(); __NOP();
      }
}