Series of articles catalog
1. Component foundation
2. Circuit design 3. PCB design 4. Component soldering 6. Program design
This so-called balance resistance needs to be analyzed in specific circumstances. It may not be possible. Do not apply it mechanically. According to the razor rule: if it is not necessary, do not increase the entity.
The first is the ideal op amp: the ideal op amp amplifies the voltage difference between the same direction and the reverse input terminal . The simplest non-directional amplifier is shown in the figure below (ignore Rg' in the dashed box)
. The input impedance of the ideal op amp is infinite, so there will be no current flowing in and out of the two input terminals. At this time, if a balancing resistor Rg' is added to the input in the same direction , There will be no current flowing through Rg', which will not have any effect on the entire circuit, so this balancing resistor is meaningless and unnecessary .
Let's take a look at the simplest inverting amplifier. Similarly, there is no current flowing in or out of the input terminal of the op amp, and there will be no voltage drop on the balance resistance Rg' added to the same direction terminal, and it can be completely removed and grounded directly.
Now look at what the actual op amp is like. The actual input structure of the op amp is as follows: the input stage of the op amp is a differential pair. For the case of the NPN tube in the figure, if the transistor is to work in the linear region, then the transistor must be correctly biased, and there must be current flowing into the op amp. The two input terminals of the amplifier (for the PNP tube are outflow), and the input terminal of the operational amplifier must not be left floating, and the external circuit must provide the correct static operating point for the input stage .
At this time, there will be a problem: the equivalent resistances RP and RN seen from the input terminals of the op amp, if there is RP Ib+≠RN Ib-, an error voltage will be generated at the two input terminals of the op amp, and this error voltage will vary with The input signal is amplified together, and finally reflected in the output.
This is why it is said in many places that op amp circuits, whether in the same direction or inverting amplifier, need to add balancing resistors to make the equivalent resistance seen from the two input terminals of the op amp equal , so as to ensure the input bias current of the op amp. No error voltage will be generated.
It sounds very reasonable, right, but to talk about toxicity aside from dosage is a hooliganism .
Let’s look at a common op amp OPA209, its input bias current is no more than 15nA (ignore the problem of ± sign first, this is caused by other circuits, see the last), suppose our circuit is seen from the op amp input The equivalent resistance difference of 10kΩ (for conventional circuits is already very large), the error voltage caused by the input bias current is, uh...150uV, which is equivalent to the input offset voltage of the op amp, how big is this error in your circuit influences?
Look at another op amp AD797. In the worst case, there is an input bias current of 3uA. In the above case, the input error voltage has increased to 3mV, which is much larger than the input offset voltage of the op amp, which is one larger than the above example. Many orders of magnitude, how much influence does this error have in your circuit?
Therefore , whether or not this balancing resistor is required depends on the actual situation of the circuit . Of course some people will say, I don't know what the actual situation is, so it doesn't hurt to put one in there, right? Then I will talk about the disadvantages of this balancing resistor.
Resistance has resistance thermal noise. This noise is an inherent property of resistance and is only related to resistance and temperature. The noise density of a 10kΩ resistor at room temperature is about 12.8nV/√Hz, the input noise density of AD797 is 1.2nV/√Hz, and the input noise density of OPA209 is 2.2nV/√Hz, which is the noise caused by a balancing resistor. It is far higher than the input noise of the op amp selected. How much influence does this noise have in your circuit? Does the noise have a big impact on your circuit or the DC error has a big impact on your circuit?
to sum up
The function of the balance resistor
(1) Provide a suitable static bias for the transistor inside the chip.
The circuit inside the chip is usually directly coupled, it can automatically adjust the static operating point, but if an input pin is directly connected to the power or ground, its automatic adjustment function is not normal, because the transistor inside the chip The voltage of the ground wire cannot be raised, and the voltage of the power supply cannot be pulled down, which causes the chip to fail to meet the conditions of virtual short and virtual disconnection, and the circuit needs to be analyzed separately.
(2) Eliminate the influence of the static base current on the output voltage, and the size should be balanced with the equivalent resistance of the external DC path of the two input terminals, which is also the reason for its name.
What are the consequences if the balance resistance of the non-inverting amplifier circuit of the operational amplifier is incorrect?
(1) The same and opposite phase ends are unbalanced. When the input is 0, there will be output. When the signal is input, the output value is always larger (or smaller) than the theoretical output value by a fixed number.
(2) The error caused by the input bias current cannot be eliminated.
To add, I saw that the input bias current of OPA209 is very small and has ±, and the input bias current of AD797 is much larger, and it is only a + value. This is because the input terminal of OPA209 is designed with an input offset cancellation circuit, this circuit cancels most of the bias current of the input transistor, so the current seen from the input terminal is very small, and then the current of the offset cancellation circuit It may be slightly larger or slightly smaller than the bias current of the input transistor, and it may be flowing in or flowing out from the outside. The input pole of AD797 does not have such a bias cancellation circuit, and the input stage is heavily biased to reduce noise, so the input bias current of AD797 is very large, and it only flows in the direction of the amplifier.