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.

OPA2863: Amplifiers forum

Part Number: OPA2863
Other Parts Discussed in Thread: OPA2837

Dear TI:

I am using part number OPA2863RUNR as a buffer amplifier.  This amp features a power-down pin.  The amplifier IC package is powered with 5VDC. 

I am observing that when the power-down pin is above 3.9VDC the output buffer follows the 4.096VDC reference as expected with a 350-Ohm load, and a short circuit current of 64mA.

However, when the power-down pin is anywhere between 0.00VDC and 3.90VDC with respect to ground, the output does not shut off but remains at 1.864V with a 350-Ohm load, and with a short circuit current of 18mA.

Why is a power-down pin voltage below the turn-on threshold not causing a Hi-Z condition on the output?

Thanks

 

  • Hi,

    section 8.3.3 of datasheet shows that there are two antiparallel protection diodes between the two inputs. When wiring the OPAmp as voltage follower with a load at the output, as you did, the voltage at the +input minus two silicon diodes' voltage drops appears at the output of OPAmp. Or by other words, when the output becomes high ohmic and you load the output by forcing a source current to flow, then the input protection diodes of OPAmp become forward biased.

    Keep in mind that this is merely a power down feature to reduce the power consumption of OPAmp. It is not intended to be used for output multiplexing purposes.

    Kai

  • Hi,

    I agree with Kai's analysis. When powered down, the amplifier's output does become High-Z but the other path from the low-impedance 4.096V signal via the two back-to-back diodes between the inputs can still drive the output resulting in this observation. There can be a few solutions to this problem:

    1. When the amplifier is powered down, if the 4.096V reference device could also be powered down, then this additional low-impedance source can be prevented from driving the circuit at the output of the OPA2863.

    2. Using a feedback resistor in the range of 500-1000 ohm with the OPA2863. When the amplifier is active, the output will follow the non-inverting input in buffer configuration, but when it is disabled, this additional resistor in the feedback will limit the current from the 4.096V source. This feedback resistor is necessary to use in such configuration to keep the current in the back-to-back ESD diodes between the amplifier's inputs below 10 mA (see max. input current rating of 10 mA for the input ESD diodes in datasheet).

    3. If (1) or (2) are not feasible, another alternate is to use the OPA2837 with an internal switch at its inverting input. When this amplifier is disabled, this switch opens disconnecting the inverting input components and these back-to-back diodes from the output, hence this path is cut-off and suitable for such circuits. More details on this in Section 7.3.4 of the OPA2837 datasheet. The OPA2837 is also offered in this RUN-10 package with the same pinout.

    Thanks,

    Bharat Agrawal

  • Thank you Kai.

  • Thank you Bharat for the explanation and the list of solutions you provide.  I considered using the OPA2837 as an candidate with the same space saving package; however, I believe that this amp's common mode input spec will not accommodate the 4.096Vdc input range.

  • Hi,

    Agreed that the OPA2837 can't support input of 4.096 V with a 5V supply. This will either need the OPA2837 input to be reduced from 4.096 V to within its common-mode input range and setting the amplifier in a gain, or the above discussed options (1) or (2) using the OPA2863.

    Thanks

    Bharat Agrawal