• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 80 subscribers
  • Views 736 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MSP430F5528 won't run at 24MHZ

Sid Levingston
Prodigy 70 points
Other Parts Discussed in Thread: MSP430F5528, MSP430F5529

Hello,

Thank you so much for reading and responding to this post. I am stuck with this problem and would appreciate any insight offered.

 

  1. I have a prototype board with am MSP430G5529 PN package on it. I used the 5529 for the prototype because it had leads that I could solder myself. I am finished with firmware development and hardware debug and everything is working.

     

  2. I had a new board fabricated with the same hardware except, I used the MSP430F5528 RGC as I wanted a smaller package and I did not need the extra I/O and ADC channels of the 5529 All the port usage, power supply’s, crystals, etc are identical for the two boards. (if P1.1 is used for a certain function on the prototype, it is the also P1.1 on the new board)

     

  3. On the new board the 5528 starts, then immediately jumps to random addresses and locks up. I have traced this to the UCS, I think. I am using the 32768 REFCLK for ACLK. I then source the FLL from the REFCLK and multiply it by 731 to get ~ 24MHz. With the 5529, no issues. With the 5528, it will not work. If I reduce MCLK to 12MHZ or less, the 5528 will work just fine, but I need the max clock speed from my application

I next wrote a small program that does nothing but initialize the clocks and ports and then toggles a pin. It will run up to about 12MHz but no higher.

Sid Levingston

  • 0
    Guru 14640 points

    Sid

    Maybe you have tried, but did you load the example code onto your new board?

  • 0 in reply to Todd Anderson78572
    Prodigy 70 points

    Hi Todd,

    Which example code do you mean? My clock initialization code is pretty much copied from the MSP430F5529  User Experience examples, but if you know of a very simple example I can try that might be useful.

    Sid

  • 0 in reply to Sid Levingston
    Prodigy 70 points

    Hi Todd,

    OK, I found the example code. I ran the UCS example 10 which sets MCLK to 25MHz sourced from REFCLK and it behaves as my code does, it will not work. So, I dug around on the old stuff shelf and tried the example code on my 5528 eval board and it works. I am thinking I must have a hardware issue on the new PCB.

    Thanks,

    Sid

  • 0 in reply to Sid Levingston
    Guru 14640 points

    Sid:

    Sorry, at the time, I did not have the examples in front of me. I have run UCS example 10 both on the TI 5529 board, and I have also run it on the 5528-based board that I am using, and both of them did work...

    Sometimes the hardware issues are directly connected to the crystal.  You may want to see if you can run at 16 MHz.

    Just to try it out, you could try running with REFO instead of the external crystal.

  • 0
    Guru 92251 points

    Sid Levingston said:
    I next wrote a small program that does nothing but initialize the clocks and ports and then toggles a pin. It will run up to about 12MHz but no higher.

    Some things to check that may cause failure to operate at 24MHz:

    1. Is the supply voltage in the range 2.4V to 3.6V with a PMMCOREVx setting of 3?

    [See Figure 1. Maximum System Frequency in the MSP430F5528 datasheet]

    2. Is the CVCORE recommended capacitor at VCORE the nominal value of 470nF ?

  • 0 in reply to Chester Gillon
    Prodigy 70 points

    Hi All,

    To answer the questions that have been asked:

    1. I am not using XT2 or XT1. I am using the DCO. But, I did try using XT2 and it behaved the same way.

    2. It will run on the DCO or XT2 up to about 12MHz. Any frequencies higher than this fail, both on XT2 and the DCO.

    3. VCC is 3V, Vcore is set to level 3 in a 1,2,3 step as the example code shows.

    4. The caps are supposed to be correct, but I did not build the board. I am having my tech install the correct ones now just in case.

    Thanks,

    Sid

  • 0 in reply to Sid Levingston
    Prodigy 70 points

    Thank you everyone for the suggestions. The capacitor on Vcore must have been incorrectly stuffed. My tech installed a 470nF and it all started working.

    Sid

  • 0
    Guru 227245 points

    Ceramic caps with a larger capacitance usually have a large tolerance. Depending on used dielectric material, capacitance can be +-50% of the nominal value. Maybe the one stuffed did have significantly less than 470nF while being 'nominally' the correct one.
    Unfortunately, this is analog tech and not digital. So a closer look on teh part datasheets is required. Even for something as simple as a capacitor.
    Of course, it also could have been a simple mistake. Especially since the SMT ceramic caps are usually not labeled.

**Attention** This is a public forum