• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 80 subscribers
  • Views 798 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.

noisy single-slope with battery powered 5519

Sylvain Bernard
Prodigy 110 points
Other Parts Discussed in Thread: TPS61220, MSP430F5519, LM317

I've got a custom MSP430F5519 board which uses a TPS61220 to provide power from 4 x AA (in parallel, is an option to power from a 24VAC using an LM317.)

The DCO is used to generate a 4MHz clock from XT1 (32KHz) on SMCLK.

I'm implementing a simple single-slope ADC using comparator-B and timer-A0 clocked from SMCLK.  When power from AC, the ADC samples within 0.1%; however, when powered from batteries, samples jitter +1%.  Using a scope, VCC seems clean.

Any ideas on what I should look at next?

  • 0
    Expert 1255 points

    You don't describe your test setup, but I wonder if you don't have a ground connection in place when you're using 24VAC that you don't have when using your batteries.   Is the test signal you're using to drive your ADC setup AC coupled to your board?

  • 0
    Expert 2740 points

    Your power supply layout and component selection should follow the EVM (TPS61220EVM-31). Battery on one side, output on the other.

    You can also add LC  filter on the supply output. 

    Peter

  • 0
    Expert 1610 points

    Hi,

    since the TPS61220 is a switching regulator (by the way I wonder why 4xAA and a step up regulator instead of a linear ldo) I could think of noise at first place that is catched up somewhere. Did you follow the layout guides in the TPS61220 datasheet? There is almost always ripple on a switching requlators output. And what kind of inductor do you use? There are unshielded and shielded ones. Also keep your power supply away from the MSP and especially from the crystal, maybe you loose some clock ticks - that can be checked by monitoring ACLK.  

  • 0 in reply to Jan Kesten
    Prodigy 110 points

    Thanks to all for your inputs.  We finally got to the bottom of these...

    And we've been chasing ghosts!  it turns out that the serial debug port, whose tx/rx traces are exposed to an offboard 232 driver, was trying to pull VCC beyond what the TPS61220 was capable of.  The 24VAC + regulator had no such limitations.

    We've now replaced the external driver with the appropriate part :)

**Attention** This is a public forum