LPC54608 Vbat overcurrent @ VDD = 0V

[valeriapezzali]

Hi,

the Vbat of the LPC54608 mounted on my prototype is permanently supplied by a 0.33F capacitor while the other supply pins are switched off when the power supply is down.

The attached document "LPC54608 VBat.pdf" is the schematic of the power nets.

The table 18 of the datasheet declares 340nA of current consumption if VDD=VDDA=VREFP=0, but I measure an unacceptable value: 8uA.

I have carried out the test using a minimal program: the BOD detects the power-down and put the LPC in deep_sleep while it waits for the complete shut-down of the VDD.

//------------------------------------------------------------------------------

int main( void )
{
 POWER_DisablePD(kPDRUNCFG_PD_BOD_RESET);
 POWER_DisablePD(kPDRUNCFG_PD_BOD_INTR);
 SYSCON->BODCTRL = SYSCON_BODCTRL_BODRSTLEV(2) | SYSCON_BODCTRL_BODINTLEV(3);
 SYSCON->BODCTRL |= SYSCON_BODCTRL_BODINTENA(1) | SYSCON_BODCTRL_BODRSTENA(1);

 while(1);
 
 return 0;
}

//------------------------------------------------------------------------------

void WDT_BOD_IRQHandler(void)
{
 POWER_EnterPowerMode(kPmu_Deep_Sleep, 0);
 while(1);
}

//------------------------------------------------------------------------------

[adrianball]

Hi Valeria,

Note that in the LPC54608 the RESETN pin appears to be pulled up internally to the VBAT supply. This means that when VDD power is off, if there is ANYTHING connected to RESETN (such a JTAG pod, 100k pullup to VDD_TARGET as on the eval board, etc), you will drain battery power out the RESETN pin. On the eval board this amounts to more than 30uA. With the RESETN pin disconnected, I measure the RTC current drain in 'deep power-down' mode as 0.7uA.

Regards,

Adrian.

[valeriapezzali]

Hi Adrian,

thank you for the support.
I found the solution of my problem in the revision 2.4 of the datasheet (note 2 of section 6.2.2), this information was not included in the document that I used to design the board.

Best regards.

Valeria

[victorjimenez]

Hello Valeria, 

The measures that appear in the datasheet such as the Ibat were made with only the MCU, not with the MCU mounted on a board. So, the value of Ibat will change if you have the MCU mounted on your own prototype since you will have more connections to the MCU that are requiring current.

Hope it helps!

Victor.

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[valeriapezzali]

Hello Victor,

a lot of applications requires only the power supply of the RTC while the power of the whole device is down.

The LPC17xx I used until now fits perfectly this need: the VBAT supplies only the RTC, the other parts of the device are not supplied, a capacitor can supply the RTC for many days.

The schematic of the OM13092 is like mine: the external battery (J10-11) supplies only the VBAT pin and VDD is supplied by the main power regulator.

But if I understand correctly your answer, this application is not allowed because the VBAT supply current is unpredictable.

Even if I disable the pull-ups and set all the LPC54608 pins the low-level outputs, the current leak from VBAT to the board is kept to 0 while the VDD is applied, but it becomes unpredictable when VDD is down.

If the above statements are correct, the lacks of essential information about the supply management in the NXP documentation, and the evaluation board schematic lead to expensive design errors.

Please let me know if there is a solution or I have to redesign my application.

Kind regards.

Valeria Pezzali

Sol.E. snc

Via Jacopo da Pecorara, 8

29122 - Piacenza

tel. +390523615748

mobile +393336859882

skype valeria.pezzali.sol-e.it

Da: victorjimenez <admin@community.nxp.com>

Inviato: venerdì 18 maggio 2018 19:37

A: Valeria Pezzali <valeria.pezzali@sol-e.it>

Oggetto: Re: - Re: LPC54608 Vbat overcurrent @ VDD = 0V

<https://community.freescale.com/resources/statics/1000/35400-NXP-Community-Email-banner-600x75.jpg> NXP Community

Re: LPC54608 Vbat overcurrent @ VDD = 0V

reply from Victor Jimenez <https://community.nxp.com/people/victorjimenez?et=watches.email.thread> in LPC - View the full discussion <https://community.nxp.com/message/1016112?commentID=1016112&et=watches.email.thread#comment-1016112>

[victorjimenez]

Hello Valeria,

In order to give you a more accurate information regarding your design I will need to see the whole schematic of your application to see where the current leak comes from. 

Regards, 

Victor Jimenez.

[valeriapezzali]

Hello Victor,

Herewith there is the schematic.

Regards.

Valeria

Da: victorjimenez <admin@community.nxp.com>

Inviato: lunedì 21 maggio 2018 17:51

A: Valeria Pezzali <valeria.pezzali@sol-e.it>

Oggetto: Re: - Re: LPC54608 Vbat overcurrent @ VDD = 0V

<https://community.freescale.com/resources/statics/1000/35400-NXP-Community-Email-banner-600x75.jpg> NXP Community

Re: LPC54608 Vbat overcurrent @ VDD = 0V

reply from Victor Jimenez <https://community.nxp.com/people/victorjimenez?et=watches.email.thread> in LPC - View the full discussion <https://community.nxp.com/message/1016478?commentID=1016478&et=watches.email.thread#comment-1016478>

[victorjimenez]

Hello Valeria, 

I missed something important you mentioned at the beginning. You said that you are sending the MCU to deep_sleep mode, however, on the table 18 of the data sheet it's mention that one of the conditions is to have the MCU in deep power-down mode. This is a completely different mode than the one you are currently using. I recommend you check the user manual (UM10912) chapter 8 to see the differences between the power modes. 

Also, we have an application note for the Low power modes and wake up times using the LPC541XX. You can use this as a guide for your project. AN11799.zip

Best regards, 

Victor.

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[valeriapezzali]

Hello Victor,

I have carried out the test sending the MPU to deep power_down mode, but if Vbat is on and VDD is down, the Ibat value doesn't change in deep power_down or deep sleep.

The AN11799 is not applicable to the LPC54608 because it is made for the LPC5411x, this MCU is not provided with the Vbat pin: it is not possible to supply the RTC while the power supply of the other parts is down.

When the LPC54608 is send in deep power_down, the pins are tri-stated, the internal peripheral are not powered, but if VDD is 0V, there is a high current leak from Vbat to the other parts of the MCU.

I'm still thinking that the declaration of the table 18 of the LPC546xx data sheet and the presence of the Vbat pin let thinking that the LPC54608 is applicable like the LPC17xx, but it is erroneous: the table 18 should be removed and the right connection of the Vbat pin should be declared: if the RTC is used, Vbat has to be connected to VDD.

Best regards.

Valeria