• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 3 answers
  • Subscribers 47 subscribers
  • Views 257 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.

OPA2992: Drawing b=very high and fluctuating i/p current from power supply

Suraj Sharma
Prodigy 60 points

Part Number: OPA2992
Other Parts Discussed in Thread: OPA2387, , TLE2141, OPA836, LMV793, OPA2991

image.png  I/p to opa2387 O/p of opa2387



This is the high level block diagram for my circuit.Just after LC tank i'm using opamp as a buffer before demodulator TS5A23159RSE. Earlier i tried to use OPA2387 but the signal having frequency 1MHz became triangular from sinusoidal.The reference or dc value of sine i/p is 2.048V.Vpp is 1.92V and Vcc of opa2387 is at 3V.I have uploaded the i/p and o/p to the opamp opa2387.I calculated the slew rate and get to know the slew rate to opa2387 is very less so i ordered OPA2992 but with this opamp,the i/p current from the dc power supply is continuously fluctuating randomly from 16mA to 70mA which was 16mA with OPA2387.Please help me to make my design perfect

  • 0
    TI__Mastermind 27778 points

    Suraj,

    The issue that you are seeing where the 1MHz input sinusoidal wave starts to look like a triangle is due to slew-rate induced distortion.  You can calculate the maximum peak waveform that has minimal slew induced distortion using the equation Vp = SR/(2*pi*f).  Here "SR" is slew-rate and "f" is the applied input frequency.  This relationship is captured in the Maximum Output voltage versus frequency curve (see below for OPA2992).  I would caution you that this curve provides the limit where gross visible distortion due to slew limitations happens.  Also, sometimes slew boost limits the accuracy of this calculation.  The curve indicates you are ok with the OPA2992, but I recommend that you choose an higher slew rate for more margin (TLE2141 slew rate of 45V/us, or TLE2888 slew rate of 40V/us).  Note that when an op amp slews or even drives a high frequency non-slewing signal the power supply current will be higher because you are charging and discharging all the bus capacitance.  Check out op amp precision labs section on slew rate for theory on this subject.

    I hope this helps.  Best regards, Art

  • 0
    Prodigy 60 points

    Thanks for the response.

  • 0 in reply to Art Kay
    Prodigy 60 points

    Hi Art kay,
    Thanks for the respose.The waveforms i uploaded was for OPA2387 not OPA2992. As per my understanding the slew rate of OPA2992 is 32 which is good for my case where Vpp is 1.68V and 1MHz frequency.I'm uploading the i/p and o/p to the opamp OPA2992 now.Every thing is proper just the current drawing by the circuit is not stable.It keeps fluctuating from 16mA to 70mA .Little bit distortion in the o/p of opa2992 is as per my understanding is okay as it is coonected to demodulator TS5A23159RSE.My concern is the unstable current.
    Please help me regarding this.
    Thanks

    I/p to OPA2992




    O/P of OPA2992

  • 0 in reply to Suraj Sharma
    TI__Mastermind 27778 points

    Suraj,

    1. I assume when you say "O/P" and "I/P" you mean input signal and output signal.
    2. The signal labeled "O/P of OPA2992" has gross distortion that is visible by inspection.  You say "every thing is proper" with the 1MHz signal on the OPA2992.  I am saying that this signal is exhibiting the signs of slew induced distortion, and you should choose an op amp with a higher slew rate.
      1. When an amplifier slews it will draw higher current.  This is likely the "unstable" current that you mention.
      2. The fact that I see two slopes on the distorted sine wave is indicative of slew boost.  As I mention, slew boost also can impact the point at which slew induce distortion happens.
      3. If you look at your supply current over frequency, you will probably notice that the "unstable current" happens when the waveform becomes distorted.  This is when you reach the slew rate limit.
      4. Note that slew rate is generally specified as as typical value (as with OPA2992).  This parameter will very due to process variation (typically +/-20%).
    3. The point at which slew induced distortion happens depends on the op amp gain.  I need a schematic to comment further.
    4. Long story short.  To get rid of the "unstable current" and gross distortion, use an op amp with higher slew rate like TLE2141 slew rate of 45V/us, or TLE2888 slew rate of 40V/us.

    Best regards, Art

  • 0
    Prodigy 60 points

    Hi Art Kay,

    Thanks for the response.Now I got the real issue.I have 

  • 0 in reply to Art Kay
    Prodigy 60 points

    Hi Art,

    Thanks for the response.Now I got the real issue.Our power supply is 3V and 2V.

    Could you please help me to find out a better opamp as I'm using it as a buffer where the i/p to the opamp is the output from LC tank tunned at 1MHZ and the o/p of the opamp is connected to the demodulator TS5A23159RSE. I was searching for some good opamp for my use case and found OPA836IDBVT.Will it be better?

  • 0 in reply to Suraj Sharma
    TI__Mastermind 27778 points

    Suraj,

    When you say your supply is 3V and 2V do you mean that the positive supply is 3V and the negative supply is -2V?  If you are running on a 2V supply you will need to be careful about you op amp minimum supply rating.  The OPA836 seems like overkill (slew rate = 560V/us, and bw = 205MHz).  Also this device has a minimum supply of 2.5V.  I did a search for op amps with slew rate > = 35V/us AND Vs(min) <= 2V.  I do not see any op amps that meet this criteria.  Here is a list of op amps with a minimum supply rating of 2V or less:  Op amp Vs<=2V .  If you need the 2V minimum supply, the option with the best slew rates are LMV793 and OPA232.

    Going back to the original question.  Minimum slew rate can be determined with the equation SR = 2*pi*f*Vp.  Where Vp is the peak sinusoidal voltage and f is the applied frequency.  For a 3Vpk 1MHz sine wave the slew rate needs to be 19V/us.  Keep in mind that this is the peak voltage, not the peak to peak.  In your case you used the OPA2992 device.  This device has slew boost and sometimes this can affect the point at which distortion happens.  Also, gain will effect the point at which this distortion happens.  The relationship (SR = 2*pi*f*Vp) is designed for a buffer configuration.  To properly address this we need the following:

    1. Circuit diagram.

    2. Power supply voltages.

    3. Max frequency, and Max peak output voltage.

    It may be that the OPA2992 shows an issue because of slew boost.  It may also be that the amplifier configuration is part of the issue.  However, by inspection of the output waveform it is clear that the issue is related to slew rate.

    Best regards,  Art

  • 0 in reply to Art Kay
    Prodigy 60 points

    Hi Art,
    Positive supply is 3V and negative is 0V.We are using one LDO also for 2V power generation so if required then we can use that one also.I have attached a part of the schematic. Please review it.

    Regards, Suraj

  • 0 in reply to Suraj Sharma
    TI__Mastermind 27778 points

    Suraj,

    I think you should use the 3V supply.  This opens up more device possibilities.  The circuit looks good.  The amplifier is in a buffer, so this is the best configuration to minimize slew induced distortion.  The required slew rate assuming a 1MHz 1.5Vpk input signal is:  SR = 2*pi*1MHz*1.5V = 9.4V/us.   Since the OPA2992 has a specified slew rate of 32V/us, I would not expect to see slew induced distortion at 1MHz.  The OPA2992 does have slew boost and for some devices this can cause slew induced distortion to happen at a frequency lower than the frequency predicted by the equation above.  Since your input is generated by an LC tank can you verify the frequency, amplitude, and signal integrity?  We have seen issues with the OPA2991 slew boost causing unexpected slew induced distortion, but I am not aware of issues with the OPA2992.  I leaving on a business trip tomorrow, so I will pass this question on to one of my colleagues.  We may want to make some measurements on OPA2992.  Sorry to stretch out your question, but I think we can resolve this soon.

    Best regards, Art