• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 49 subscribers
  • Views 579 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.

LM331 Pin 2 oscillation

andrea chierici
Prodigy 50 points
Other Parts Discussed in Thread: LM331

Hi all,

I'm using a LM331 as a voltage to frequency converter and I observed a weird oscillation on PIN2.

I put the LM331 on a solderless board using the component values shown in the datasheet (http://www.ti.com/lit/ds/symlink/lm331.pdfand adding power supply decoupling capacitors of 100nF-1uF.

I grounded the input in order to avoid comparator triggering. The datasheet shows a typical current range from PIN2 of 10uA to 500uA, so if a 1.9 V reference is used, then Rs maximum value should be 190Kohm. 

I put the circuit inside a grounded/shielded aluminium case to avoid EMI.

In this condition if I use a Rs value higher than 10Kohm, then oscillation appears on PIN2. This oscillation is at 65-100Khz, and seems pretty regular once Rs value is fixed. The amplitude is around 100-200mV.

If I let Rs to float, then the terminal connected to PIN2 oscillates, while the other is steady at 1.9V. (antenna?)

I'd like to know if this behaviour is normal or not (I suppose not).

  • 0
    TI__Genius 16890 points
    Hi Andrea,

    Some hints on making this circuit work:

    Rs: Make this value between 12k to 16k. Making this resistor too small can take too much current from the internal circuit. 14k may be a good value to start with.

    Ct should be a NP0 type.

    CL (the 1uF) needs to be a mylar (best) or ceramic (ok) type of capacitor. This cannot be an electrolytic capacitor.

    With the values that you have in your circuit, when you apply a voltage to the input do you get the correct frequency out, according to the formula in the datasheet?

    Mike
  • 0 in reply to Michael Stout
    Prodigy 50 points
    Hi Mike,

    thank for your response. Right now I stopped using the simple VtF and I'm using a "precision VtF converter": pin 1 on the inverting input of the OpAmp integrator.

    As you suggested I used NP0 ceramic (Ct) and PP (instead of mylar for Cf).

    With the values I used I got 8700 Hz while I should be around 10000 Hz according to the formula. This could be caused by components tolerance, but it's a bit too much because I used 1% resistors and 5% capacitors. However frequency output is linear so this is not really a problem.

    The real problem is that, if I use a value higher than 10K (14k for example) for Rs a small oscillation (the oscillation amplitude grows if I use higher values for Rs, for example is around 200mV with a 100k resistor, while the frequency is around 65-100Khz) appears on Pin2. This oscillation appears even with no input signal and no comparator trigger.
    Pin2 is the reference current pin, and is the inverting input of an op amp configured as a charge pump. I really dont know if this is caused by an instability problem of this charge pump, or if it is a normal behaviour.

    This cause a 200-300Hz oscillation when I'm around 10Khz with a fixed input signal, probably because if the current from pin2 oscillates then the current from pin 1 oscillates as well (when the comparator is triggered).

    Andrea
  • 0 in reply to andrea chierici
    TI__Genius 16890 points
    Hi Andrea,

    No, you shouldn't see an oscillation on pin 2. Can there be something else causing it, a layout or scope issue? Could you send a picture of the oscillation?

    By the way, you mentioned you tried Rs at 100k. It is not recommended to use a value that large. You should stay near the value recommended in the datasheet (12k - 15K). This sets the current at about 100uA. When a larger resistor is used, for example 100k, this will give a current of 1.9V/100k = 19uA. This gives only 1.9V across RL which is not enough.

    Mike
  • 0 in reply to Michael Stout
    Prodigy 50 points
    Hi Mike,

    thank again for your reply.

    Tomorrow I'll be back to the lab and I'll send a picture of the oscillation and the circuit.

    I will try a Rs value in the suggested range, modifyng Ct and Rt in order to change my frequency to voltage gain a bit (I can't change Rin). What's the minimum value for Rt*Ct time constant? In the "precision voltage to frequency converter" the OpAmp integrator Gain BandWidth could set the limit if it is low enough. I read somewhere that the pulse duration "should not be too short", but I was not able to quantify it.

    Andrea