• State Resolved
  • Replies 6 replies
  • Subscribers 49 subscribers
  • Views 169 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

OPA818: Parallel an OPA818 (7 V/V stable) & an OPA817 (unity stable) in an alternative Power-Down scheme

Chia Hung Lai
Prodigy 110 points
Part Number: OPA818
Other Parts Discussed in Thread: OPA817, THS3091, THS3491, OPA837, TIDA-01565

Hi TI expert,

I am designing a gain configurable precision wideband amplifier, the OPA81x family seems meeting my requirement. However, I need to switch between different gains while maintaining the bandwidth as high as possible, e.g. x5 & x50.

As the OPA817 & OPA818 has a Power-Down mode, is it feasible to connect the two op-amps in parallel, i.e. common the +IN & OUT pins, and toggle the /PD pin respectively for different gain?

  • Gain = x5, enable OPA817, disable OPA818
  • Gain = x50, disable OPA817, enable OPA818

The exact power-down behavior of the OPA817 & OPA818 is not mentioned in the datasheet, do advise. Thanks.

  • 0
    TI__Genius 11585 points

    Hello Chia,

    My initial thoughts are that you can do this, but you need to make sure both devices are not on at the same time.  You must wait for one device to be fully off before powering on the other.  The enable/shutdown timings are in the device datasheets.  Additionally, you may find that if there is a large enough signal on the shared output line, the "off' device may turn "on."

    Do you know what your input signals, output signals, and amplifier load would be for this project?

    Best,

    Alec

  • 0 in reply to Alec Saebeler
    Prodigy 110 points

    Hi Alec,

    Thanks for the advice. The combined output of the OPA817/818 pair is intended to drive a CFB amplifier (THS3491 or THS3091) at a fixed gain, so the amplifier load is not an issue here.

    The output signal level is intended to be 7.06 Vpp sine max, just within the output capability of OPA817/818 for Vs = +/- 5 V.

    Op-Amp Gain (V/V) Max Input (Vpp) Max Output (Vpp)
    OPA817 5 1.412 7.06
    OPA818 50 141.2 m 7.06

    I will check the datasheet of OPA817/818 for the power-down timing.

    Please elaborate further on the statement:

    Alec Saebeler said:
    if there is a large enough signal on the shared output line, the "off' device may turn "on."

    Thanks.

  • 0 in reply to Chia Hung Lai
    TI__Genius 11585 points

    Hello Chia,

    I am glad I was able to help.  I may look into your overall circuit more, specifically including the CFB amplifier as the final stage.

    Please let me know if any concerns or issues as they occur.  I am going to see if your max values would cause any ESD diodes to activate on these devices.

    Best,

    Alec

  • +1 in reply to Alec Saebeler
    TI__Genius 11585 points

    Hello Chia,

    I realize I did not include an explanation for my earlier statement.  The OPA817 and OPA818 do not have high impedance outputs when in power down (PD).  The OPA837 does have high impedance outputs when in PD, and is used for muxing in TIDA-01565.  The OPA837 creates an open circuit at the inverting input and has a high-impedance output when in PD.

    For the OPA818 and OPA817, if the current to the ESD diodes exceeds 10mA the diodes will conduct at the inputs.  As these devices both have low output impedances over frequency, there is not assurance the devices will remain in power down or behave normally when sharing the same output line.  Series input and output resistors would need to be used to current-limit the signals going into either amplifier.

    I am asking across my team to learn more about how to support your design.  Your 7.06Vpp maximum output may affect the power down of the off channel, but it may also be okay depending on your output load and current drive.  The inputs may be fine in your design, but I am checking if the output of one amplifier could turn-on the other.  

    Best,

  • 0 in reply to Alec Saebeler
    Prodigy 110 points

    Hi Alec,

    Thanks for the detail explanation. I am abandoning the parallel OPA817||OPA818 outputs w/ PD design for now, as I realized that there's an extreme scenario where as much as +/- 6.3 Vpeak may be present to the output of OPA817||OPA818 outputs, where it is obviously beyond the supply capability of both OPA817 & OPA818. Thanks.

  • 0 in reply to Chia Hung Lai
    TI__Genius 11585 points

    Hello Chia,

    Thank you for the update.  

    Please ask a new question or reach out over private message if you need to discuss your overall design or want to share details offline.  I can send my email over message if you ever need me as a contact.

    I am marking this thread as "TI Thinks Resolved" for now; you can mark my previous message on PD output impedance as "Resolved" and close this thread.  A new related question can always be opened.

    Best,

    Alec