Correcting op-amp offset and input bias currents in one step

[eem2am]

op amp offset

Hello….

It is known that op-amps give slightly inaccurate output voltages due to …..

1) Input bias currents
2) Input offset

So, I realise that an op-amp has an input offset that means that when used with negative feedback, its output voltage will not be zero when it has zero volts between its inverting (V-) and non-inverting (V+) inputs.

I am aware that this offset is gotten-rid-of by using a pot at dedicated “offset-null” pins…..OR by putting some high value resistor between the positive supply of the op-amp and the inverting input.

Also…

… I am aware that an op-amps V+ and V- inputs draw input bias currents….and that these bias currents create voltages at the inputs, since they flow through whatever resistors are biasing the op-amp……
..in order to mitigate this problem with an inverting op-amp, a resistor of value R(feedback//R(in) is put from V+ to ground.

…..However, in reality, the op-amps “input offset” is actually caused by the fact that the op-amps “input bias currents” are actually NOT equal…(?).

…so in actual fact, “input offset” is just another problem caused by “input bias currents”. (?)

As such, if I want to mitigate (get rid of) these bad effects, then surely I only need to use ONE of the above correcting steps?……
…in other words……rather than use a resistor from V+ to ground AND the offset-null pot……….-I can simply just use the offset-null pot to deal with both the “input bias current” problem and the “input offset” problem.

Do you believe this is correct?

Thankyou for reading.

 

[LvW]

op amp offset null

Yes, I believe this is correct (in case you are going to make an offset null adjustement).
In many cases this is not necessary because of feedback (which reduces the output offset) . In additio, some effects caused by the bias currents are somewhat reduced (not eliminated) by this special "bias cancellation" resistor.

 

[FvM]

op amp input bias current

In addition, the said bias cancellation resistor is only meaningful with OP, that have a bias current with a specified sign. This is mostly the case with legacy bipolar (e.g. uA741) and a few modern high speed types. Others have effectively no bias curent (FET types) or bias current compensation (e.g. OP07 and many high speed bipolar types.)

 

[Audioguru]

input bias current compensation

If you don't like input offset voltage then buy an opamp where the manufacturer has nulled it for you.

If you don't like input offset current then buy an opamp with FET inputs.

I don't think there is a nulled FET-input opamp made.

There used to be some auto-zero opamps.

 

[Kral]

op amp offset current

eem2am,
Input offset voltage and input bias current are independent from each other, and have different causes. For example, in a bipolar op amp, the input offset voltage is caused by the Vbe mismatch between the two input transistors. The bias current is due to the fact that the Hfe of the input transistors is finite, and therefore, some base current must flow. If the Hfe (Betas) of the input transistors are not equal, the the base currents will not be equal, and an input offset current will exist as well. This will cause an additional output error equal to Ioff X Rf. Note that the offset current erro can be of either polarity.
.
As an example, take the case where the offset current is zero, and the ouput error due to bias current is zero because the resistances from the inverting and non-inverting inputs are equal. In this case there is no output error due to bias or offset currents. However, the input offset voltage might not be zero. This error can be eliminated using the external offset trim pot.
.
It is true that you can eliminate errors due to offset voltage, offset current and bias current by using the external trim pot, or by connecting a resistor from one of the supplies to the Inv input. However, this compensation will work at only one temperature. The errors due to the three causes will drift with temperature, and they may either cancel or add. For this reason, it is best to deal with the three causes separately. Balancing the resistances from the Inv, NI inputs eliminates the bias current errors at all temperatures. Using the offset voltage pin will not 0nly eliminate the offset voltage error, but will (for many modern op-amps) also reduce the offset error drift with temperature. With some early designs, such as the 741, the offset voltage drift gets worse if the offset adjust pin is used. Offset current errors can be eliminated at one temperature only, by injecting a current into the Inv input. However, to eliminate the offset current errors at more than one temperature, the injected current must be temperature compensate, since the offset current varies with temperature.
.
By compensating for the three sources of error separately, you have better control of the ouput errors over the complete operating temperature range of your circuit.
Regards,
Kral

 

[xulfee]

offset nulling of an opamp

for 741 put a potentiometer between pin 1 & 5 n apply voltage to it,then turn nob of potentiometer,put ur voltmeter on pin 6 n observe 0 output,this is how u can nullify the effect

 

[Audioguru]

op-amp eliminate error

for 741 put a potentiometer between pin 1 & 5 n apply voltage to it,then turn nob of potentiometer,put ur voltmeter on pin 6 n observe 0 output,this is how u can nullify the effect

No.
Connect the offset null parts as are shown in the datasshet of the opamp.

 

[eem2am]

op amp offset adjustment

thankyou very much for all these replies.

Does an instrumentation amplifier using opamps get rid of the input bias current and input offset voltages completely?

 

[Audioguru]

op amp input offset voltage

Does an instrumentation amplifier using opamps get rid of the input bias current and input offset voltages completely?

If an instrumentation amplifier uses FET inputs then its input bias current is very low but increases when the temperature increases. I don't know if anybody makes an intrumentation amplifier with FET inputs.

An instrumentation amplifier has matched resistors inside for excellent common-mode rejection. Its input offset voltage is laser trimmed.

An AD620A is a low cost isntrumentation amplifier.
Its max input bias current is 2nA at room temperature.
Its max input offset voltage is 125uV.

 

[FvM]

op amp offset nulling

Yes, you can get JFET and CMOS instrumentation amplifiers, that have very low bias currents, but higher noise voltage than bipolar types.

You can also see from the AD620 specification (as well from the functional schematic), that it's using bias current compensation, thus the remaining input currents are in an offset current order of magnitude.

 

[nice_star2008]

bias curent

i want to cancel the bias current & offset current by digitl methods(successive approximation algorithm) in simulink.can any one help me?