MKE02Z64M4 at 80degree C

[neilporven]

I am experiencing a lock-up issue when the MKE02Z64M4 is operated at 80deg. Celcius.

Here is my code:

ICS_C2 = 0x20; //BDIV=2, divide by 2
ICS_C1 = 0x04; //Set internal reference clock to FLL

/**************************************************************/
/* Trim Value is entered in the debuger configuration menu */
/* From the Debug tab, click on Advance Options. In advance */
/* Options, click on Calculate Trim. The value is stored in */
/* a memory location designated by Freescale (0x000003FF). */
/**************************************************************/

cTempIndex = 0x000003FF; //The address location of Trim value

ICS_C3 = *cTempIndex; //Load Trim Value = 35.15625 kHz, BUS Frequency is 18MHz

while (!(ICS_S & ICS_S_LOCK_MASK)) //Lock Status (Wait for the FLL to be stable)
{
BACKLIGHT_ON;
}

BACKLIGHT_OFF;

The spec's for the processor state operating temperature -40C to 105C.

Is there an errata for this?

Thank you,

Neil

[kerryzhou]

Hi Neil Porven,

     From your code, it seems you are using the internal IRC and trimmed it.

     From the datasheet, you can get that:

  The deviation is up to 2% over full voltage and the temperature range.

  When the deviation is larger, and it is out of the FLL DCO range, it will caused the FLL lock up failed.

   I suggest you choose the external crystal which can support 80 degree C, then try it again.

Wish it helps you!

Have a great day,
Kerry

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

[neilporven]

Thank you Kerry for responding.

Kerry can you place a link to that document or tell me what is called so that I can do a search on the NXP site?

From the datasheet spec you showing above, I don't see where it is stating that my ICS_S_LOCK bit does not get set.

It just specify about a percentage off the output of the DCO which at this point is not critical to my application.

I went ahead and as a test, I lowered my frequency from 18MHz to 9MHz and the ICS_S_LOCK bit started to get set and 

I didn't get locked in my while loop.  I feel that running the processor at the max bus frequency at or above 80 degrees C could

be causing the issue, but I would have to look at the datasheet you found.

Please let me know your input on this?

Regards,

Neil

[neilporven]

Update Kerry,

After bringing the temperature down to room temperature and performing the test again, I got different results.

At approx 73 degrees Celcius, I started locking up again.  Apparently lowering the clock down to 9MHz did not

correct the problem with the lock-up.  Please let me know the link or the datasheet name your using to get the 

processors specs.

Regards,

Neil

[kerryzhou]

Hi Neil Porven,

   Sorry for my mistake, the picture I post is from the older datasheet.

Now the new datasheet can be downloaded from this link:

http://www.nxp.com/assets/documents/data/en/data-sheets/MKE02P64M40SF0.pdf 

Page 22

Now, there has two points you need to take care:

1. Trim configuration

  I don't know which code you are using? And how you get the trim data.

Normally, we use the Codewarrior or the KDS PE code,  like this:

Then use the PE debugger to trim it, it will save the trim data in address 0x3ff, and 0x3fe.

I also attached my test project for your reference.

You can trim the 31.25Khz which is the default trim data, please also take care, the trim temperature should in 25C, VDD=5V.

2. Make sure your FLL out (DCO) is in the range of 32Mhz to 40Mhz.

Please trim it in temperature 25C, 5V VDD at first, then add the temperature to 80C, check whether it still have this problem.

Wish it helps you!

Have a great day,
Kerry

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

[neilporven]

Hi Kerry,

Our trim value at 25C 5V VDD did not work at temperatures 80C and above.  

So as a test, I decided to use the factory default values and not load anything or

check off any of the boxes on the debug configuration menu that uses the TRIM

value registers.

My code looks like this:

ICS_C2 = 0x20; //BDIV=2, divide by 2
ICS_C1 = 0x04;

//Lock Status (Wait for the FLL to be stable)
while (!(ICS_S & ICS_S_LOCK_MASK))
{
FGPIOA_PDOR |= 0x00000400;         <---------------------- sets a pin HIGH
}

FGPIOA_PDOR &= ~0x00000400;     <---------------------- sets a pin LOW

The results are the same, below the 80C  looking at my scope, I get a sharp pulse.

Above the 80C, its a HIGH pulse, indicating that it never leaves the while loop.

Can you please verify if you get the same?

Regards, 

Neil

[kerryzhou]

Hi Neil Porven,

    I don't have the test tool which can change the temperature .

   Now, please give me your oscilloscope test wave when in 25C and 80C, then I will check it with our according department.

   Besides, did you have another board? You also can check another KE02 board, just use the internal IRC, and test it, whether it has the same result?

Waiting for your test pictures.
Have a great day,
(my name)

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

[neilporven]

Hi Kerry,

Sorry I haven't been able to reply sooner.

I have many other boards, since we are in production.  I tested other boards and they display the same behavior.

Can you please pass this information with the appropriate department, timing has become critical.

ICS_C2 = 0x20; //BDIV=2, divide by 2
ICS_C1 = 0x04;

//Lock Status (Wait for the FLL to be stable)
while (!(ICS_S & ICS_S_LOCK_MASK))
{
FGPIOA_PDOR |= 0x00000400;         <---------------------- sets a pin HIGH
}

FGPIOA_PDOR &= ~0x00000400;     <---------------------- sets a pin LOW

Can you pass this information and please let me know who will be working with me on this?

Regards,

Neil

[kerryzhou]

Hi Neil,

  Please add this code:

 SIM_SOPT |= SIM_SOPT_CLKOE_MASK;

Then test the bus clock in default trimmed factory IRC in 80 degree C and under 80 degreeC.

  Whether you can get 16MHz in PTH2 or not?

We need to check the frequency.

Have a great day,
Kerry

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

[neilporven]

Hi Kerry,

I also opened up a ticket for this issue.  I have someone by the name of Vincent also helping me out.

I explained to him the situation and what was happening.  I was also able to find out that the interrupt

NMI was being triggered at the 80C which I found odd.  I don't know if the NMI has anything to do with 

what I am experiencing, but its just another observation as to what is happening as the temperature reaches

and exceeds the 80C mark.

Thank you,

Neil

[kerryzhou]

Hi Neil,

   Can you debug your code at 80 degree C?

   After the board working at 80 degree C, please run the code for a while, then use the debug to check the register , whether  the FLL states:ICS_S[Lock] =1?

  Please test it on your side, then give me the result!

  Please give me reply and the test result, this is very import point for the problem solution.

Waiting for your reply!

Have a great day,
Kerry

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

[kerryzhou]

Hi Neil Porven,

    Thank you very much for your information, the NMI information is very important to us, Vincent is our colleague.

    Now, please try to add 4.7K to 10K external pull-up resistor in the NMI pin(PTB4), because the internal pull up is weak, normally, we recommend the customer to add the external pull-up, I check your schematic, you didn't add it.

    I don't know how you get the 80 degree C, if that heating tool have a lot of EMI, and you didn't add the external pull up in the NMI pin, the interference signal may around the NMI pin, it may cause the NMI interrupt.

   Please add the NMI external pull-up, then try again in 80 degree C.

 Waiting for new test information, besides, please also check the bus frequency as I told you before.

Have a great day,
Kerry

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

[kerryzhou]

Hi Neil,

  Sorry for my later reply because of Chinese Spring Festival.

  I already help you check it with our according department, when I get the reply, I will let you know ASAP.

 Please wait patiently.

  Thanks a lot for your understanding.

Kerry