• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 48 subscribers
  • Views 403 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.

Original question:

TLV3501: What is frequency equation

TLV3501: capacitor value error

Osman
Prodigy 15 points
Part Number: TLV3501
Other Parts Discussed in Thread: LMV7239, TLC555

Tool/software:

I implemented this circuit. But I have a problem. I measure the value of the capacitor with a high-precision LCR meter. And then I insert it into its place in the circuit. When I power on the circuit and mathematically calculate the value of the capacitor over the frequency produced, I find a different capacitance value. In other words, there is a difference between the calculated value and the value I measured on the LCR meter. For example, the capacitor I measured as 102pF is calculated as 114pF. is it not normal to have such a big difference? I encounter this situation with different capacitors as well.
What is the problem?

  • 0
    Guru 318490 points

    Does your equation take the propagation delay into account? The traces and components also have a capacitance, and I'd guess the LCR meter isn't infinitely precise. 12 pF sounds like a reasonable total error.

  • 0 in reply to Clemens Ladisch
    Prodigy 15 points

     Considering the structure of the board, I don't think it's normal to have so many errors. The error rate is almost 10%. 

  • 0 in reply to Osman
    TI__Guru**** 167506 points

    Osman,

    The capacitance is low enough that percentage is not the way to look at this error. 12pf is not a lot. You could try a faster comparator. 

  • 0 in reply to Ron Michallick
    Prodigy 15 points

    Thanks Ron.

    But I need to standardize the error. When I add a larger capacitor, I have difficulty establishing a meaningful relationship between the measured and calculated value. This makes the relationship between frequency and capacitor difficult. I need to establish a mathematical relationship between frequency and capacitor value. I can't determine any offset or gain. I also tested LMV7239

  • 0 in reply to Osman
    TI__Guru**** 167506 points

    Osman,

    This oscillator is similar to what TLC555 timer does. So I'll use the equation in that data sheet.

    For (RA+ RB) use R1 + [output resistance of VOH]. 

    For (RB + ron) use R1 + [output resistance of VOL]. 

    tplh & tphl are comparator propagation delays. Output frequency is 1 / [tc(H) + tc(L)]

    Because this formula is complicated and require values that may be hard to get, most designers measure actual result then tweak value or R1 to fine tune the frequency.

  • 0 in reply to Ron Michallick
    Prodigy 15 points

    Hi Ron,

    Thanks for your detailed explanation.

    I want to clarify my real goal: I'm not trying to fine-tune the frequency to a target value.
    My actual goal is to determine the actual capacitance value based on the measured frequency.

    In other words, I measure the output frequency and want to accurately calculate the capacitance from it.

    Considering that the oscillator's internal resistances (VOH, VOL) and propagation delays affect the measurement,
    what method would you recommend to correctly back-calculate the capacitance value from the measured frequency?

    Should I characterize the output resistance and delay first, or should I approach it differently?

    Thanks for any advice.

  • 0 in reply to Osman
    TI__Guru**** 167506 points

    Osman,

    In that case greatly increase R1 for the range that includes 100pF. Higher make internal drive resistance les impactful. Lower frequency makes propagation delay less impactful. 

    Is final C1 always there but is variable or do different caps get attached?

  • 0 in reply to Osman
    Guru 318490 points

    I do not know your cost constraints, but you will probably have to calibrate with a known, precise capacitance.