Total current required by OPAMP

[Bjtpower]

Hi,

I am trying to do paper calculation for current requirement for OPAMP AD8022

https://www.analog.com/media/en/technical-documentation/data-sheets/ad8022.pdf

I would like to make sure how much current required by OPAMP when it is in operation mode.
I have captured below snapshot from PAGE No: 3

I have supply voltage: +-5V
So 55 mA is the correct current which will draw from my power supply? or 55mA+11mA=66mA (Quicent current of both opamp) or something else?

I

 

[D.A.(Tony)Stewart]

Input Impedance must be defined from 1 port with input only from that port.

Yet there are 3 ports and 1 is actively correcting Vin+ input errors whilst creating confusion of where is the input impedance with 2 source currents passing thru the input resistor. (Source and Error negative feedback.) We only care about input V/I and not Vin/ (Iin+I.fb)

Therefore only one of the 3 methods is correct for z11 and the output must be active in the linear region for this to be correct. Thus the CM mode test is the only way to independant measuring voltage drop from a matched source. Alternatively to verify add a model of this result to each source of the diff amp but driven from one source. This "matched source" will yield a flat response exactly at half power at both input Ain+/- validating the CM method Z(f)=V/I (f)

 

[FvM]

Input Impedance must be defined from 1 port with input only from that port.

Yet there are 3 ports and 1 is actively correcting Vin+ input errors whilst creating confusion of where is the input impedance with 2 source currents passing theu the input resistor. (Source and Error negative feedback.)

Therefore only one of the 3 methods is correct for z11.

That's simply the definion of Z11, as quoted in post #57. But it's not the impedance seen by a source connected to differential amplifier input. I'm not describing different measurement methods, I'm describing different source configurations. Each is creating different effective input impedance values.

Taking only Z11 as real means to ignore the effects of source configuration. The source is however loaded with actual input impedance.

I see your intention to keep things simple, but it doesn't work.

 

[D.A.(Tony)Stewart]

The 3 port cannot be computed as a 2 port. This is where the error rises from the negative feedback current.

But you can test each input as a 1 port device with a matched source = load

 

[FvM]

The 3 port cannot be computed as a 2 port. This is where the error rises from the negative feedback current.

Misses the point. Yes it's a 3-port (two amplifier inputs, one output), but V1 and V2 don't depend on I3. Z11, Z12 ,Z21, and Z22 respectively don't change when you add an output port to the analysis. I'm omitting it just for simplicity.

--- Updated Dec 17, 2023 ---

O.k. reducing the setup to 2-port is based on prior knowledge of circuit behaviour (particularly assumption of ideal OP in the frequency range of interest). If you can't be sure, apply full 3-port analysis.

If we restrict analysis to separate 1-port measurement of AIN+ and AIN-, we see that AIN- impedance (Z22 in the suggested impedance matrix) is still affected by negative feedback.

The common mode input case (left schematic in post #58) which you describe as "Each input is matched" doesn't fulfill the specification that the input to the port must be "only from that port", it's actually a 2-port setup. 6.01k impedance of AIN- is only observed when AIN+ is driven with the same voltage. True 1-port measurement reads 4.81k, by the working of feedback.

 

[danadakk]

This paper solves for the individual input expressions, and one can set
the solutions to be equal to show the conditions for the equal case.


Regards, Dana.

 

[D.A.(Tony)Stewart]

Misses the point. Yes it's a 3-port (two amplifier inputs, one output), but V1 and V2 don't depend on I3. Z11, Z12 ,Z21, and Z22 respectively don't change when you add an output port to the analysis. I'm omitting it just for simplicity.

Yes but z11 is only = Rin for each port is only equal to the series input no matter what the negative feedback current is and regardless of which input is the signal source or differential. So Rfb affects input current and voltage without affecting impedance.

If the resistors are matched on the input and shunts matched with any ratio ,

Can you reason why z11+ + z11- = z11diff is true (or false?)

I say it is true only iff R1=R3 and R2=R4

otherwise when not matched impedance gain on inversion makes z11- ≠ z11+
yet with same gain ratio

z11diff is still =R1+R3.

More importantly, the tolerance error of the input resistors in high gain will affect the z11 for each input greatly due to impedance inversion of negative feedback.

https://tinyurl.com/ysegh8vf

Here I intentionally mismatched the input R1,R3 then with a differential input adjusted the source resistors to match the 50% node amplitude.

 

[KlausST]

We only care about input V/I and not Vin/ (Iin+I.fb)

We always measured the current exactly at the input node and the voltage at this node.
This is the definition of input resistance and input impedance. R_IN = V_IN / I_IN

We focussed at the very input and nothing else.

Or in other words:
There is a signal source connected to the input. We used the voltage of the source and the current delivered by the source.
And we calculate what resistance (or impedance) the source has to drive.

This is true for this difference amplifier circuit and also well known for the standard inverting OPAMP circuit.

****
Standard inverting OPAMP circuit: uses two resistors. R1 (input) and R2 (feedback).
It is very well known the input resistance is just R1.
Independent of R2 and no textbook adds the feedback current to the input current. Nor do we with the difference amplifier circuit.

Klaus

 

[D.A.(Tony)Stewart]

Here the ratios are matched but Rin are mismatched, yet still z11 + z11- = z11 diff

https://tinyurl.com/yvd66k4h

--- Updated Dec 17, 2023 ---

We always measured the current exactly at the input node and the voltage at this node.
This is the definition of input resistance and input impedance. R_IN = V_IN / I_IN

This does not work with negative feedback from Port3 (output). So I added source impedance to match input for 50% reduction on EACH port to match the unequal input R's

See the last sim above and observe the results. at the box interface the Vdiff is 50% meaning source matches the load for each single ended input R1+R3=differential input impedance. and remove the 60 Hz and see each + or - input is also 50% of source.

--- Updated Dec 17, 2023 ---

OK I concede

https://tinyurl.com/ysv3znsu

 

[KlausST]

Hi,

thanks dana for providing the document in post#65.
They named it "obvious" .. I personally don´t call it obvious .. and the disscussion (ours and theirs) is an example for that it´s not that obvious.

It shows the math behind.
It shows that the input impedaces (indeed only the inverting one) depend on operation mode.

The impedances are identical on common mode. Only on common mode.

They did the math for the case when (AIN-) = -(AIN+)

They provided all formulas for calculating different cases, like:
* floating differential input (math is also shown in the document in the following discussion with Ken)
* one input grounded
* common mode
* whatever you want.

Klaus

 

[FvM]

They don't say the calculation is obvious, just the fact that + and - input impedances are different and depending on input voltage ratio.

I agree with the concept of looking at effective input impedances for different source configurations, applying the calculation scheme to the circuit discussed in this thread confirms my results in post #49.