• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 51 subscribers
  • Views 1076 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.

LM324: Terminating of Unused Amplifiers

Tai Wei Chua
Prodigy 80 points
Part Number: LM324
Other Parts Discussed in Thread: TINA-TI, , LMC6484, LMC6482

Good day,

I came across an application note from TI regarding the termination of unused op-amps.   

I ran the above through TINA-TI but, could not detect the issues highlighted in the article.     As such, we built a laboratory prototype to test the issues highlighted in the Technical note and also did not find the problems indicated in the technical note.

The LM324 is a Rail-Rail input op amp, as such, driving its input to the 0V rail does not seem to give the amplifiers any problems.    We also tried saturating the LM324, and the current increase was not particularly high.    As it turns out, the second configuration (+/- input grounded) draws less current than the first configuration (unity gain voltage follower).

Please advise if it is ok for LM324 and LMC6484 to be operated with its inputs shorted as in the second configuration above.
Please also advise, for the first configuration (unity gain voltage follower), is it ok to drive the input to GND.

Thank you and have a good day,
Tai Wei

  • 0
    Guru 243220 points

    The image did not attach. Please use the "Insert Media" button.

  • 0 in reply to Clemens Ladisch
    Prodigy 80 points

    Sorry about that, this is the drawing.

  • 0 in reply to Clemens Ladisch
    Prodigy 80 points

    Sorry about that, this is the drawing.

  • 0 in reply to Tai Wei Chua
    Guru 243220 points

    The Spice model does not necessarily implement all behavour under abnormal conditions.

    How did you create the 100 mV voltage source in the real circuit? Or did you actually ground the input?

  • 0 in reply to Tai Wei Chua
    Guru 140200 points

    Hi Tai,

    have you performed the same measurements with the LMC6482 as well? Didn't you again notice any relevant increase in supply current?

    Kai

  • 0
    TI__Guru*** 140026 points

    Tai Wei,

    For LM324 see Design Guidelines for Devices with LM324/LM358 Cores . I have never seen problems with any unused set up with the exception for direct to ground connection (no series resistor) when the ground bus was poorly laid and had high current transients. 

    For LMC6484, I have no data for this device when used in a non linear mode; open loop, comparator, or output driven to V- or V+. Therefore, I suggest to keep the unused amplifier biased properly as the unused amp application note suggests.

    The simulation model may not accurately reflect non linear operation.

  • 0 in reply to Ron Michallick
    Prodigy 80 points

    I have not tested the LMC6484 yet.

    For the LM324, using voltage follower configuration:

    Seems like connecting the positive input to ground reduces current.


    We also tested the case whereby

    both inputs are shorted to ground, i.e. the not recommended configuration:

    Ok, we checked the output voltage, the opamp output is zero, so we tried another variation, i.e. we offset the positive input using a power supply to m

    ake the opamp saturate to VCC:

    It appears that for LM324 type amplifiers, shorting the input pins to the ground does not appear to raise the current significantly.   Also, ground level is within the input range of the LM324.    As such, we would like your opinion, if perhaps the conditions mentioned in the application note on termination of unused opamps, may have special cases such as LM324, that the rules do not apply?

  • 0 in reply to Tai Wei Chua
    Guru 243220 points

    The app note recommends 10 kΩ resistors only to protect against situations in which the inputs are below GND. As mentioned by Ron, this does not happen in most actual circuits.

    When both inputs are at the same voltage (here: 0 V) without feedback, the output is determined by the offset voltage. If the offset voltage happens to be (near) zero, then the output might try to move around. Therefore, the appnote recommends circuits that force the two inputs to have a difference larger than the worst-case offset voltage.

    I do not know if your chip has a positive or negative offset voltage. Therefore, not all of your measurements might be universally valid.

    43 mA or 57 mA sounds too high. I guess the actual units are 0.43 mA and 0.57 mA (which corresponds with the datasheet typical values)?

    To answer your question: the 0 V voltage follower has the lowest supply current, so this is what you should use. (The app note confirms that VOL uses less supply current than VOH.)

    While the app note recommends a voltage in the linear operation range, it never says that his would reduce the current. This recommendation is not explained; as far as I can see, the only negative effect of saturation is increased overload recovery time, but that does not matter for an unused op-amp.

  • 0 in reply to Clemens Ladisch
    Prodigy 80 points

    Thanks,

    Yes, should be 0.43 and 0.57mA.

    I see, so the best termination for rail-rail input op-amps could be to use a voltage follower and ground the signal input as the signal low is lower current.

    Thanks and best regards,
    Tai Wei

  • 0 in reply to Tai Wei Chua
    Guru 140200 points

    Hi Tai,

    I always do it this way (picture from another thread):

    Kai