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.

Inquiry regarding LMP2021 CG equation

Other Parts Discussed in Thread: LMP2021, OPA188

Hello, all

Now we are evaluating LMP2021, then would like to ask you regarding CG equation.

Please see the item below and feedback us with your comment.

On page 18 of this datasheet, it is described that CG could be calculated by RG value.

However, we could not find this detailed equation on the datasheet.

Please let us clarify how CG value could be calculated.

We thank you in advance for your information.

Best regards,

  • Hello Atsushi,

    In short - the LMC2021 is NOT suitable for high (>100Kohm) source impedance.

    What the section is basically trying to explain is that the LMP2021 (which is a "chopping" amplifier), when used with a high impedance source, will not have "conventional" DC bias current.

    Because the input is "chopped" by a MOSFET switch matrix at 35KHz, there are large transient currents that flow in-and-out of the inputs on each clock cycle due to gate charge injection in the MOSFET switches. This creates essentially an AC impulse current noise on the inputs - independent and uncorrelated with the actual DC bias current.

    For non-chopper amplifiers, to get the lowest DC offset, one must balance all the DC resistances at the inputs for the best bias current cancellation across the source resistances.

    For a chopper, you must also balance the capacitance on the inputs. If one input has less capacitance, then the charges created by the transients will not match and do not cancel. These current pulses are then integrated by the input capacitance and eventually appear as a DC bias current offset.

    This creates a rather strange “bias current vs input capacitance” factor. That is what these graphs are trying to explain.

    What the graph is really showing is the how the calculated input bias current changes as input capacitance changes. The "bias current" shown was back-calculated by the change in measured offset using a 1Gohm sense resistor.

    In reality - if you wish to add this "cancellation" capacitor - the true capacitor value can only truly be found through experimentation in the actual circuit and layout. THe graphs are only good for exactly 1Gohm and for that particular device. Strays will play a big factor in the results. You must also be careful, because adding this capacitor can reduce the overall circuit bandwidth and/or add instabilities (use to peaking).

    In order to use the graphs, you must know what the capacitence is at the input pin of your circuit, including strays, fixture, sensor and cable capcitence in order to calcuate the "cancellation" capacitor. Even if you calculate the value, the needed value will differ because of device to device variations in which the effects can change several percent.

    If your input impedance is higher than 100Kohm, then look at the OPA188 or OPA33x series of Auto-Zero amplifiers. They use higher clock frequencies and do a little better job of cancelling the input current transients.

    Regards,