SAMB11 Clock

Go To Last Post
5 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have a question about the clock of SAMB11. I have the SAMB11 Xplained board and works fine with the Atmel examples, like the blinking example.

I have also designed a circuit board with a separate SAMB11 chip and when I program this chip with the same example software, there' s a problem with timing.

When I program a delay of 200 ms and use the program at the Xplained board, the output is toggled every 200 ms. But with the same program on my own circuit board it is about 240 ms. The time is not constant and varies with addition of extra software.

I have measured the frequency of the crystal on my board and it is exactly 26 MHz with a spectrum analyzer, so that's not the problem.

I used the reference diagram of the SAMB11 datasheet for the design. The only difference is that I don' t use a RTC clock, because I don' t need this for my application. Also the baudrate of the UART differs from the XPlained board, so it seems it is working at the internal RC oscillator?

Does someone recognize this, what is wrong?

 

Last Edited: Tue. Jan 17, 2017 - 04:06 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I am having the same issue but have not yet resolved it.

I have had some details provided by Atmel but not enough to give you really good advice.

 

Here's what I currently understand.  The external RTC crystal is used by Atmel's firmware stack to calibrate the internal 32kHz RC oscillator so what you are seeing are the effects of the RC oscillator not being calibrated.  Apparently there is a way to configure the firmware stack to not do this calibration and run the module only with the 26MHz XO by writing to a particular eFuse in the module and additionally tieing the AO_GPIO_0 pin to VCC but the compromise that this will keep the module from entering ULP.

 

Other than the above which is still being worked out, we found that adding the RTC crystal back in to the circuit fixed the issue.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thank you for your response. The problem is solved by writing to bit 14 of block 3 of bank 5 and also tieing AO_GPIO_0 to VCC. I don't need the ULP mode so this workaround is sufficient for now.

Do you have information available about the configuration bits of the Efuses? I cannot find any details of these bits.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi marpan70,

 

I believe what you are describing with the bit 14 of block 3 of bank 5 write is the configuration with the eFuses.   Currently I'm still waiting for more information from Atmel and the only thing I know is that possibly a command line tool 'EfuseBlockProgram.exe' is required.  Could you please describe how you did that write, or provide the code you used, or provide the document you have to know that, as I haven't been able to find any information about this.  Thanks.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Just a question, how are you timing the 200ms delay ?

 

Currently I'm using a dualtimer (as using a timer and using ULP crashes the SAMB11 most of the times) with my own millisecond counter.