MKL16Z64VLH4 sleep mode issue

[robertkrisjocso]

Hi All,

We are currently using MKL16Z64VLH4 in our products and we are encountering in several and increasing in numbers of our units with high sleep current. Our good units during sleep mode (LLS) only consume between 20 to 30 uA while the bad unit are consuming 150 to 190uA which is quite high.

We are using the following interfaces.

1. TSI (12 electrodes)

2. SPI interface (interface to RF module)

3. GPIOs (LEDs and buzzer control)

There are 25 unused GPIOs which are not defined.

Our wake-up location is at TSI0.

Mask Set of all the chips that we are using is 1N15J.

We have tried to swap MCUs from Good board (board with good MCU) to bad board (board with bad MCU), the result is that the good board with bad MCU consume high sleep current (150 to 190uA), while the Bad board with Good MCU, consume low sleep current (20 to 30uA) which simply show that the problem is inherent to the MCU.

We have check the erratas, there are no related erratas for the specific Mask set that is related to sleep mode.

Does any have any thoughts about this?

Your comments, suggestions and recommendation are high appreciated.

Krgrds,

Kris

[Hui_Ma]

Hi Kris,

Could you do a test with bad MCU to disable the TSI module before enter into LLS low power mode?

What's the power consumption with this situation?

Please let us know the result. Thanks.

Have a great day,
Ma Hui

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

[robertkrisjocso]

Hi Hui_Ma,

Thank you for your response.

We have tried your request to disable the TSI module before entering LLS low power mode on the bad MCU and the measured sleep power consumption below is the results:

First power up: 150uA to 170uA (first time the power is connected to the board and let the MCU enter sleep mode)

second power up: 180uA to 190uA (from the first power up, the supply was disconnected for few seconds and reconnected to the board)

The sleep current is still high.

Krgrds,

Kris

[Hui_Ma]

Hi Kris,

Thank you for the info.

I had checked with Kinetis product team, there without record about KL16 power consumption exceed datasheet provided data.

How many chips from your side with this issue?

Please refer AN4503 chapter 2.6 Kinetis Energy Savings – Tips and Tricks about how to get a low power consumption.

And I also recommend customer to refer attached FRDM-KL26Z sample code [low_power_demo] example.

Thank you for the attention.

Have a great day,
Ma Hui

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

[robertkrisjocso]

Hi Ma Hui,

We have already seen the APP note AN4503 and compare are code which is generated by processor expert into it.

I don't have yet the numbers of affect chips in our production, what I have are RMA from our customer which are report as fast drain units 2 out 7 units manifest this issue. We are planning to add additional testing in production to filter out units with this kind of issue.

Could you reattached the sample code of  FRDM-KL26Z? looks like you missed to attached it.

Krgrds,

Robert Kris

[Hui_Ma]

Hi Robert,

Ok, Please check the attached FRDM-KL26Z_SC file.

Have a great day,
Ma Hui

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

[robertkrisjocso]

Hi Ma Hui,

Sorry, but I could not see any attachment, could you send it to my email: kris.jocson@ionics-ems.com.

By the way, we have made some evaluation on the bad units (units with bad mcu - high sleep currrent), we have tried to set the different scenario on the unused gpios below are the summaries:

1. Set all unused GPIOs as input prior to entering LLS. The sleep current is still the same 150uA to 190uA.

2. Set all unused GPIOs as output, value is "low", prior to entering LLS. The sleep current is still the same 150uA to 190uA.

3. Set all unused GPIOs as output, value is "high", prior to entering LLS. The sleep current is still the same 50uA to 60uA.

Scenario # 3, shows a significant reduction on the sleep current of the bad units, which is almost near as the target sleep current which is 20uA to 30uA.

We have search the net, on how to proper way to define the unused GPIOs and all of the references that we have seen said that the proper way is to set the unused GPIOs as output with value of "low", but this is different from the result of our evaluation.

Any comments on this one?

Krgrds,

Robert Kris

[Hui_Ma]

Hi,

That's really strange, I could find the attachment:

I will check with IT about attachment issue.

About the unused pin setting, I find KL17 chip provide the recommendation at KL17 datasheet.

I think customer also could refer it at page61 chapter5.3.

Please following the KL17 recommended way and check the result.

Have a great day,
Ma Hui

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