AT32UC3A1512-AUR PLL will not Lock

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

I have been using this same processor for the past 6 years.  In all that time I have never used the Power Manager in ASF to set up my clocks, but rather System Clock Services.  I even posted a how to in May of 2017 (  But for some reason, the good times are over.  I cannot get past the dreaded PLL not Locked. 


I have tried previous software projects that worked before, I have tried different processor boards - some used to work, I just made two new ones - one failed immediately, the other ran for a while before it started failing.


I like using System Clock Services - it is easy to understand - setup the clock values in conf_clock.h and call sysclk_init();  No mess, no fuss - and now no PLL LOCK!


Here is the code I am using:

//conf_clock.h set cpu to 66 MHz


/* Fbus = Fsys / (2 ^ BUS_div) */
#define CONFIG_SYSCLK_CPU_DIV         0
#define CONFIG_SYSCLK_PBA_DIV         0
#define CONFIG_SYSCLK_PBB_DIV         0


#define CONFIG_USBCLK_DIV             1

#define CONFIG_PLL0_SOURCE          PLL_SRC_OSC0
#define CONFIG_PLL0_MUL             16 //(48000000UL / BOARD_OSC0_HZ)
#define CONFIG_PLL0_DIV             3

#define CONFIG_PLL1_SOURCE            PLL_SRC_OSC0
#define CONFIG_PLL1_MUL               (48000000UL / BOARD_OSC0_HZ)
#define CONFIG_PLL1_DIV               1

#ifndef USER_BOARD_H
#define USER_BOARD_H

#include <conf_board.h>

/* Clock Speed */
#define APPLI_CPU_SPEED	66000000
#define APPLI_PBA_SPEED 66000000

/* These are documented in services/basic/clock/uc3b0_b1/osc.h */
#define BOARD_OSC0_HZ           12000000
#define BOARD_OSC0_STARTUP_US   17000
#define BOARD_OSC0_IS_XTAL      true
#define BOARD_OSC32_HZ          32768
#define BOARD_OSC32_STARTUP_US  71000
#define BOARD_OSC32_IS_XTAL     true
#define FOSC0					12000000

#define CLOCK_PBA_FREQ          66000000

#define Test_Pin        AVR32_PIN_PA19      

#endif // USER_BOARD_H
#include <asf.h>
#include <board.h>
#include <conf_board.h>
#include <user_board.h>

void IO_Init(void);

void board_init(void)

void IO_Init(void)


	#include <asf.h>
	#include <user_board.h>

int main (void)
	int c;


	while (1)

As I mentioned - one of the new assembled boards ran for a while before failing to the no PLL - the output pin was producing a waveform with a 90ns pulse.


The only thing I have not tried is changing the JTAGICE - I have a second one, so I guess I should give it a try since I have been using the current one for 5 years - that would eliminate one more possibility. 


I, as always, would welcome some thoughts on this strange, fustrating, problem.


Kind regards,


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

The datasheet says that the Vco frequency range is 80 MHz to 240 MHz, but my version of ASF thinks that it should be 40 MHz to 240 MHz.
If the reality is a minimum of 80 MHz, then try adjusting your mul and div values.

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

Hi mikech - as always I appreciate your suggestions.  I had tried your suggestion prior to your making it - but did it again using mul values of 1 to 18 with a div of 1 and of 0 - and with several of the two different versions of my board.  I got nothing.


I tried a different JTAGICE with no success.  I have tried another project board (that the CPU board plugs into) and finall a different computer. 


A little more playing and a newer board came to life with an old program.  I tried the same program on the newer computer and and it ran.  Tried the same board in a different project board and it ran.  Changed programs and BOOM - nothing.  I have compared conf_clock.h, board.h, init.c, main.c and I can find no difference!


At this point, I am going to search for why one program works and the other does not - and then why one board works and another does not.  All the capacitors for the power suppies on the CPU are mpunted under the processor and are 0402 size - I may have a solder issue on one of them.  After all, I will be 70 in 4 months and my eyesight is not what it used to be!


Below is an image of the front of the CPU board for comparision of the the back of the board and the caps under the processor.  If you have anymore thoughts, please feel free to share them!

Thanks again,