Unreasonable power consumption of LPC1114/302 in Deep Power-down mode

[lpcware]

Content originally posted in LPCWare by ut_conrad on Fri Jun 14 00:25:20 MST 2013

Hi,

I need to use LPC1114/302 in Deep Power-down mode in order to conserve battery power. However, the current consumption is still about 1.5mA when Deep Power-down mode is enabled. The following is one of my test programs:

[collapse title=Deep Power-down Mode Test Program]

#ifdef __USE_CMSIS
#include "LPC11xx.h"
#endif

#include <cr_section_macros.h>
#include <NXP/crp.h>

__CRP const unsigned int CRP_WORD = CRP_NO_CRP ;

int main(void)
{
    LPC_SYSCON->PDSLEEPCFG = 0x18ff;
    LPC_SYSCON->PDRUNCFG = 0xedf0;
    LPC_SYSCON->PDAWAKECFG = LPC_SYSCON->PDRUNCFG;

    LPC_PMU->PCON = 0x02;
    SCB->SCR |= (1 << 2);    // Set SLEEPDEEP bit
    __WFI();
    SCB->SCR &= ~(1 << 2);    // reset SLEEPDEEP bit

    for (;;) __NOP();

    return 0 ;
}
[/collapse]

I have also changed system_LPC11xx.c to use IRC instead of crystal

[collapse]

#define SYSPLLCLKSEL_Val      0x00000000
#define SYSPLL_SETUP          0
#define SYSPLLCTRL_Val        0x00000000
#define MAINCLKSEL_Val        0x00000000
[/collapse]

During testing, the only connections to the MCU LPC1114/302 are the SWCLK, SWDIO, VDD and VSS pins. I have also tried setting all I/O pins to GPIO (except SWCLK and SWDIO) and output at low level. The result is still the same. Note that the program execution stops at __WFI(). Without the __WFI(), the current is much higher.

I have run similar program using LPCXpresso LPC11U14 board (LPC11U14/201), and the current is less than 100uA.

Can anyone tell me what can be the problem? My program or the MCU.

Thank you in advance.

-- Conrad

[lpcware]

Content originally posted in LPCWare by LabRat on Tue Nov 05 06:27:26 MST 2013

---

Quote: Rickster
Should I be using a TI microcontroller or a PIC?

---

[color=#f00]OMG, don't do that!!![/color]

There are so many wonderful ATMEL micros....

Not sure which sample we are talking about, but 'LPC1100 Self Wake-up Sample Application' (='selfwakeup' sample of AN AN11027) in LPCXpresso6 / Release is consuming 1/4 mA here in Deep Sleep with a simple separated original LPCXpresso1114 Target Board...

[lpcware]

Content originally posted in LPCWare by Rickster on Mon Nov 04 20:00:53 MST 2013
I am using AN11027 I get 2.1ma.  AN11027 says that it is not necessary to change the port pins in deep power down mode.

Also Deep Sleep does not work for me.  I have cut the 1343 USB away from my LPC1114 expresso board and I still can get
no lower than 2.6ma there.

I've ordered a new LPC1114 expresso board.   In some event the part is defective or bad.

I was willing to accept that perhaps this chip cannot cut it for a software wake-up that could achieve anything less than
one 1ma.  I was going to order a Dallas DS1372 watchdog chip and wake up using the deep power down mode but
I can't even get the deep power down mode to work below 2.1ma. 

What is wrong?  Should I be using a TI microcontroller or a PIC?  This particular product has a potential use of 9000 LPC1114
if it would work for me.

[lpcware]

Content originally posted in LPCWare by cararra on Thu Jul 18 06:26:57 MST 2013
Your problem sounds like a GPIO pin problem I had a while back.
GPIO pins all have to be inputs to get the low battery drain.
If they are set as outputs, even if not connected, they consume a lot
of power in power down mode.

The way I got around the issue is to set the GPIO pin
to input, set the programmable pull up or pull down for the input depending
on whether I wanted the input to appear as logic 1, or
logic 0, and then drive a FET such as 2N7002 with it.
Works wonders as the FET draws no current.
Though it is sensitive to pick up and FET does not switch fast.

[lpcware]

Content originally posted in LPCWare by bkastel1 on Sun Jul 14 13:52:41 MST 2013
Hi,

sorry to hijack this thread but I have strange experience with normal sleep. When I out the LPC1114/302 FBD48 to sleep (using Systick to periodically (every 500 msec) wake it up and do some reading on I2C) the current lowers to 1,22mA in average,but after some 19 second the current slowly rises to 1,8mA and then drops back to 1,22mA for another 10 seconds. And this goes on forever. Does anybody have an idea what this my be or how can I trace what is causing this current rise? The current rises to the value that LPC has if I do not put it to sleep. So most probably it has something to do with the sleep mode but I have no clue what.

Borut

[lpcware]

Content originally posted in LPCWare by ut_conrad on Mon Jun 17 01:01:01 MST 2013

I did some more tests. It seems that MCU reset itself whenever it trys to enter deep sleep / deep power down mode. Furthermore, the MCU consumes about 1.x mA more than others running the same program under normal execution. It should be a defective chip.

[lpcware]

Content originally posted in LPCWare by ut_conrad on Sun Jun 16 20:56:27 MST 2013

Thank you for your reply.

I have tested using the sample code in AN11027 with the modifications to use IRC instead of crystal. The current is still about 1.5mA. I have tested another MCU in the same lot, the result is also the same.

On the other hand, I used my test program on the same MCU with different package (LPC1113FHN33/302 instead of LPC1113FBD48/302) and different date code, the current is only 0.0001mA.

The marking on the defective MCU is "LPC1114F/302 07.107 ZSD13 061B". Can you find out whether this problem apply to the whole lot or I am just unlucky :-(

[lpcware]

Content originally posted in LPCWare by tha on Fri Jun 14 19:13:57 MST 2013

Just to confirm it's not a software issue, can you try using the sample code in AN11027?

<a href="http://www.lpcware.com/content/nxpfile/an11027-using-lpc1100-low-power-modes-and-wake-times-lpcxpres...">http://www.lpcware.com/content/nxpfile/an11027-using-lpc1100-low-power-modes-and-wake-times-lpcxpres...</a>

If that code still gives you a high current draw, you may have a defective chip.