CC2340R5: Very high current draw after enabling internal DCDC

Rob Oppen

Part Number: CC2340R5
Other Parts Discussed in Thread: SYSCONFIG,

Tool/software:

I'm using a custom PCB. Here's the CC2340R5 portion of the schematic:

When I enable the internal DCDC the idle current draw with no radio activity is 50mA at 3V! If I change the voltage regulator in sysconfig from DCDC to the GLDO, the current draw is ~0mA at 3V (I'm using a power supply that can't measure under 1mA).

I was under the impression that the DCDC is more efficient than the GLDO?

over 1 year ago

0 FI over 1 year ago

TI__Mastermind 21610 points

Hi,

Have you tried example code from the SDK to verify?

Try the rfPacketTx project from the SDK.

https://dev.ti.com/tirex/explore/node?node=A__ANZwbR0Z-XV0RRPrHbvggg__com.ti.SIMPLELINK_LOWPOWER_F3_SDK_PROPRF_MODULE__58mgN04__LATEST

regards

0 Rob Oppen over 1 year ago in reply to FI

Prodigy 160 points

I tried the pwmled project, which doesn't do anything but blink the LED and that has the same behavior - much higher power when using the DCDC

0 FI over 1 year ago in reply to Rob Oppen

TI__Mastermind 21610 points

Hi,

pwmled1 project puts the device to sleep for 50msec, so you will not notice sleep current on regular power supply indicator.

/* Sleep time in microseconds */
uint32_t time = 50000;

Change the sleep time to 5 sec and see it that works.

You don't have to change any sysconfig parameters. The default SDK projects are designed to showcase the device features.

Regards.

0 Rob Oppen over 1 year ago in reply to FI

Prodigy 160 points

I changed the pwmled source to just sleep and nothing else. With the DCDC in use the current draw is 50mA. With the GLDO in use the current draw is less than 1mA. Both of these tests are at 3V. There must be a connection problem in hardware, but we literally copied the LP-EM-CC2340R5 schematic so I'm failing to see what we did wrong.

0 FI over 1 year ago in reply to Rob Oppen

TI__Mastermind 21610 points

Hi,

From the shared schematic, nothing looks wrong. What else do you have on the board?

Do you disconnect the debugger when doing measurements?

regards,

0 Rob Oppen over 1 year ago in reply to FI

Prodigy 160 points

I really appreciate the help FI!

There are a couple of ambient light sensors and humidity sensors on a shared I2C bus, a push button, and an RG LED that's always off right now. Everything is powered off Vbatt, which in this case is a power supply set to 3V. The board isn't very complicated.

I've measured with and without the debugger in both cases and when the DCDC is enabled, the power draw is significantly higher (50mA vs <1mA). When stepping through the code with the debugger, the current draw immediately spikes up to 50mA with this one line enabling the DCDC:

PMCTLSetVoltageRegulator(PMCTL_VOLTAGE_REGULATOR_DCDC);

__STATIC_INLINE void PMCTLSetVoltageRegulator(uint32_t regulator)

{

// Write to VDDR regulator register while preserving other configuration

// fields

uint32_t tmp = HWREG(PMCTL_BASE + PMCTL_O_VDDRCTL) & ~PMCTL_VDDRCTL_SELECT_M;

HWREG(PMCTL_BASE + PMCTL_O_VDDRCTL) = tmp | regulator;

}

0 FI over 1 year ago in reply to Rob Oppen

TI__Mastermind 21610 points

Hi,

Please do the following to rule out HW issue:

1. Cut off power from all sensors (cut VBAT trace). That is, just power the CC2340 portion of the board.

2. Load and run empty project form the SDK and measure current:

You will have to disconnect debugger for the device to go to sleep mode.

https://dev.ti.com/tirex/explore/node?node=A__AOxSzUOfE8KQLyklCeEobw__com.ti.SIMPLELINK_LOWPOWER_F3_SDK__58mgN04__LATEST

Can you share the device picture?

Regards,

0 Rob Oppen over 1 year ago in reply to FI

Prodigy 160 points

My only next option is to cut a trace on my brand new PCB? That feels extreme. Is there a step I can take before cutting traces to help diagnose?

Bluetooth®︎

Bluetooth forum

CC2340R5: Very high current draw after enabling internal DCDC