Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 49 subscribers
  • Views 1941 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.

OPA2340: Instable issue

Cera Wei
Intellectual 2010 points
Part Number: OPA2340

Hi team,

My customer meets an instable issue when using our OPA2340. This issue occurs about 1% (4/400) possibility.

The circuit and output waveform are as below, the input is -12V DC, and we want to get a 2.4V DC output.

To exclude the influence from the output, we removed R669 and C383, it's still instable.

We tried the following things:

1. Simulated by TINA, it's stable;

2. Increase the value of R667 and R666 to 100k and 500k, all chips are oscillated.

Do you have any idea what may cause this issue?

Thanks

Cera

  • 0
    Guru 140200 points

    Hi Cera,

    yes, phase stability analysis shows that your circuit is not very stable. Make R682 = 0R and put a phase lead capacitance across R667 of at least 10pF. Why is R682 in your circuit?

    Also, R666 and R667 should sit close to the OPAmp. There isn't any cabling, is it?

    Antoher issue is the supply decoupling. You have taken a ferrite bead with a rather high low frequency inductance. I calculated 2.9µH from the impedance plot of datasheet. This will give a very sharp and heavy resonance at about 300kHz with the 100nF decoupling cap. Increase the decoupling capacitance and/or add a resistance in series to the ferrite bead of about R >= SQRT(2L/C):

    An alternative is to put a cap with considerable ESR ("snubber") in parallel C382.

    Kai

  • 0 in reply to kai klaas69
    TI__Intellectual 2010 points

    Hi Kai,

    Thanks for your timely help.

    R682 is to balance the bias current of Vin+ and Vin-. Actually we tried to put a 0ohm resitor on R682, it's still instable. This is a common circuit for an amplifier application, but you said the phase stability analysis shows that your circuit is not very stable. Is this depends on the input capacitance (it's only 6pF according to datasheet)? Could you share the simulation file or result?

    I did the simulation, and it seems to have enough phase margin.

    I will try to  put a phase lead capacitance across R667 of at least 10pF.

     Yes, R666 and R667 sits very close to the OPAmp.

     As to the supply decoupling, I understand, we will try your suggestions, thanks.

    What confused us is why only 4 chips out of 400 have this issue, which parameter difference may cause this issue?

    Thanks,

    Cera

  • 0
    TI__Guru 64160 points

    Cera,

    R682 should ONLY be used to cancel IB related error in bipolar op amps without IB cancelation (where IB in both inputs flows in the same direction) but NOT CMOS op amp like OPA2340 where IB may have opposite polarity-please remove it because it only increases thermal noise.

    However, the instability comes from the fact that using a large input resistor, R666, together with the input capacitance,Cin, introduces a zero within the effective bandwidth of your close-loop transfer function, Acl=-R667/(R666||Zcin), which causes a shift in the phase margin.  In order to fix the stability issue, please add a CF capacitor across R667 feedback so R667*CF=R666*Cin.  Since Cin=Ccm+Cdiff=6pF+3pF=9pF, this means to cancel the zero you must add 45pF RF cap across R667.

  • 0 in reply to Cera Wei
    Guru 140200 points

    Hi Cera,

    I have made a simplified phase analysis for you. R1 is the open loop output impedance of OPA3240 which I found by simulation. First, your circuit with the 16k2 resistor and without feedback capacitance:

    Then, without the16k2 resistor but with 10pF feedback capacitance:

    This circuit should already work. By increasing the feedback capacitance you can furtherly increase stability. The following simulation shows the performance with a feedback capacitance of 45pF, as suggested by Marek:

    You will notice that for this feedback capacitance the phase is almost zero for all frequencies.

    Kai

  • 0 in reply to kai klaas69
    TI__Genius 11135 points
    Cera

    We haven't heard back from you so we assume this resolved your issue. If not, post a reply below or create a new thread if this on has timed-out.

    Thanks
    Dennis
  • 0 in reply to Dennis Goeke
    TI__Intellectual 2010 points
    This problem is solved, thanks for your kindly and timely help.