• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 47 subscribers
  • Views 282 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.

TLV9154: Nonlinear Distortion Problem

sberegov
Prodigy 20 points
Part Number: TLV9154
Other Parts Discussed in Thread: TL072H, TLV9164, OPA4197

Tool/software:

Hello,

I'm using this op amp as an analog front-end (unity gain) buffer and it works according to spec for frequencies less than a few kHz. 

Above 10-15 kHz, the part's output voltage becomes dependent on the output load - the output signal becomes distorted.

Also, if I use an isolation resistor on the noninverting inputs, output ringing occurs that is also load-dependent.

Please see the attached schematic and traces:

The yellow trace below is the input signal and the blue trace is the output signal.

Here, at 20 kHz, the THD is low and the output is not distorted.

At 40 kHz there appears quite a large distortion at the zero-crossings. This is with a 160-ohm resistive load to GNDA.

Without the 160-ohm load, there is no distortion.

There is also some ringing with the 13k-ohm input resistors. This ringing goes away when the outputs are loaded.

No ringing with the 160-ohm load, but the signal is distorted.

The datasheet states that TLV9154 has a GBW of 4.5 MHz and a slew rate of 20 V/us so this is not what I expected.

If I swap the TLV9154 with a TL072H, the problem goes away, but, of course, TL072H is not a RRIO op amp.

Is this an expected behaviour for this part and if so, does an equivalent part exist that can tolerate up to 30 V of input supply voltage?

Thank you in advance!

  • 0
    TI__Guru* 77221 points

    Hi sberegov, 

    sberegov said:
    If I swap the TLV9154 with a TL072H, the problem goes away, but, of course, TL072H is not a RRIO op amp.
    sberegov said:
    Is this an expected behaviour for this part and if so, does an equivalent part exist that can tolerate up to 30 V of input supply voltage?

    Yes, this is expected behavior and it is not well documented in the datasheet. The issues are TLV9154's slew rate depends on input voltage signal's amplitude. For small signal close to 1V range, the slew rate is poor and the output may be distorted. The slew rate of 20V/usec is slew-boosted and only activated in a large input signal. 

    Please try TLV9164 RRIO, which it may work better (this is low cost general purpose op amp, 16V op amps). Or you can try TLV4197-Q1 (higher cost) or OP07x (not RRIO, your supply is +7.5V/-2.5V dual supply rails).

    Image below is TLV9164's slew rate behavior. 

    What is the driving application? If you provide me a price range for a quad op amp, I may be able to find other lower cost alternatives. For instance, OPA4197 is the same family as TLV4197-Q1, but it is higher cost due to lower Vos. 

    If you have other questions, please let me know. 

    Best,

    Raymond

  • 0 in reply to Raymond Zhang1
    Prodigy 20 points

    Hi Raymond Zhang1,

    thank you for the feedback.

    Raymond Zhang1 said:
    TLV9154's slew rate depends on input voltage signal's amplitude

    Oh, I've missed this part of the datasheet. Strange behaviour indeed.

    I'm using the op amp as an optional buffer for a low-impedance differential input 13-bit ADC. 

    The ADC's common-mode input voltage is +7.5/-2.5 V so this is why I'm looking for a (possibly) RRIO op amp. The supply voltage of the whole system is around +8 V. 

    The differential-mode input voltage is around 1.5 V - large input signals are not an option.

    Also, I need the amp to be able to withstand the clamping voltage of those SMAJ-s - at least 26 V max Vcc.

    The bandwidth is not a big concern, I need a maximum of 100 kHz, but the signal behaviour should remain linear at least upto 1 MHz so transients are not causing any distortion in the measured signal.

    And, if that's not enough, the chip should be operational in the industrial (-40 C/+60 C) temperature range.

    The price isn't the biggest factor, I guess anything below $ 3.5 per chip is tolerable.

    TLV9154 really sounded like the perfect amp for the purpose, but sadly the slew rate is too low.

    I've looked into the OPA4197's specs and it sounds promising! Do you think it will be fit for the purpose, or maybe you can recommend me another part?

    Best Regards,

    sberegov

  • +1 in reply to sberegov
    TI__Guru* 77221 points

    Hi sberegov,

    If you have able to use OPA4197 precision op amp, it will be the best option. It is designed for ADC interfacing - mux-friendly op amps, which are the same family as OPA4196 and OPA4192. You may be able to use OPA4196, which is rated at 2.5MHz. 

    https://www.ti.com/lit/ab/sbot040c/sbot040c.pdf?ts=1748360968005

    If you have other questions, please let me know. 

    Best,

    Raymond