• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 79 subscribers
  • Views 1806 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Proper use of FLL_CTL0 and XCAPXPF

Chris62565
Prodigy 100 points
Other Parts Discussed in Thread: MSP430FG4618

Hello,

I am experimenting with an MSP430FG4618/F2013 board and I am curious about the FLL_CTL0 |= XCAP18PF; setting.  I am also reading the MSP430 Microcontroller Basics and downloaded the Family Guide (slau056j) and the data sheet.  From looking at other forums I have figured how to determine from the datasheet the capacitance that you can choose from XCAP0PF, XCAP10PF, XCAP14PF, and XCAP18PF.  The example uses XCAP18PF.  I saw in another post that if your crystal is rated a certain capacitance you should choose the capacitance closest to the crystal. I'm slowly piecing this together most of what I have read I probably should have been able to understand but sometimes just need a better explanation.

Thanks,

Chris

  • 0
    Guru 227245 points

    A crystal works like a pendulum. An electric charge swaps from the capacitor on one side throught he crystal to the capacitor on the other side and back. The crystal itself is a frequency-dependent capacitor with a resonance frequency where it has the lowest impedance. On one side, the MSP gets the signal output, on teh other side it is injecting the inverted signal to keep the oscillation running and compensate for losses. If the MSPs feedback is too fast, the impedance of the crystal raises, the charges flows slower and the output is delayed, so the whole loop is slowed down again. Of course that's just a simplified explanation. (Crystals are a really complex breed)

    The two capacitors on both crystal sides need to match the crystals internal capacitance at the resonance point. So if the crystal is specified for a load capacitance of 9pF, it needs 18pF on each side (including any parasitic capacitance by processor pins and PCB). It is twice, since the two capacitors are seen 'in series' from the crystals view (crystal-capacitor-GND-capacitor-crystal) which halves the effective capacitance. The capacitor-split across GND is necessary for the oscillation circuit used in the MSP. In theory, a single ideal capacitor of 9pF parallel to the crystal woudl be perfect, but then you couldn't start the oscillation nor measure it :)

    The MSP provides the necessary capacitors for a typical low-power 32768Hz watch crystal internally. But do not forget the parasitic capacitance od the MSP pins, the socket, the PCB (especially the breakout connectors) etc.

  • 0 in reply to Jens-Michael Gross
    Prodigy 100 points

    Hey, I appreciate it.   I clicked on the verify answer cause to say that answered my question.  Although I'm not sure if that was the correct usage for the forum, but thank you very much for the answer.  There is probably more on this subject buried in my EE books but sometimes some body can explain it with based on their experience a lot better than reading it in text. 

  • 0 in reply to Chris62565
    Guru 227245 points

    Chris35513 said:
    I'm not sure if that was the correct usage for the forum

    That's exactly what this funciton was made for: mark the answer(s) that solved your problem (partially or completely) :)

    Crystals are indeed a very complex field and there are many good and more bad books and articles around. I had to read through some of them some time ago and I'm still far from being an expert, but after some reading, the basics started to become clearer. I still wouldn't dare to design my own driver circuit or calculate a safety factor, but for the use with the MSP and its internal drivers, it should be enough.

    Chris35513 said:
    sometimes some body can explain it with based on their experience a lot better than reading it in text. 

    It's still text :) but you're right, a different explanation from someone with a different background often makes the difference (well, this sentence could be considered a pleonasmus)
    Sometimes it's just the viewing angle that lets the figure appear.

**Attention** This is a public forum