I am open to a test setup that demonstrates the purpose of that resistor, especially if it differs from mine.

This simulation isn't different but it shows the effects of voltage and current offset at the output of the 741 model I have with a gain of 100.

The model input voltage offset is [V(1,2)] is 1.02mV (typical for the data sheet I have).
The model plus input bias current is 58nA and the minus input bias current is 102na for a current offset of 44nA (my data sheet shows typical values of 80nA for the bias and 20nA for the bias offset).

I switched a 10kΩ resistor in and out (Vsw high is resistor shorted) of the plus input to correct for the bias current (but it can't correct for the bias offset).
As you can see the output offset is less when the resistor is in the plus input.

I will leave the explanation of the Vout offset values displayed as an exercise for the reader.
Correct answer gets an attaboy.

 

This simulation isn't different but it shows the effects of voltage and current offset at the output of the 741 model I have with a gain of 100.

The model input voltage offset is [V(1,2)] is 1.02mV (typical for the data sheet I have).
The model plus input bias current is 58nA and the minus input bias current is 102na for a current offset of 44nA (my data sheet shows typical values of 80nA for the bias and 20nA for the bias offset).

I switched a 10kΩ resistor in and out (Vsw high is resistor shorted) of the plus input to correct for the bias current (but it can't correct for the bias offset).
As you can see the output offset is less when the resistor is in the plus input.

I will leave the explanation of the Vout offset values displayed as an exercise for the reader.
Correct answer gets an attaboy.

View attachment 101116

What is offset adjustment like in your circuit?
This is an artificial circuit, obeying artificial rules following artificial parameters you set up. This is not a reality. If you imagined a difference in the design there will be a difference.

 

This simulation isn't different but it shows the effects of voltage and current offset at the output of the 741 model I have with a gain of 100.

The model input voltage offset is [V(1,2)] is 1.02mV (typical for the data sheet I have).
The model plus input bias current is 58nA and the minus input bias current is 102na for a current offset of 44nA (my data sheet shows typical values of 80nA for the bias and 20nA for the bias offset).

I switched a 10kΩ resistor in and out (Vsw high is resistor shorted) of the plus input to correct for the bias current (but it can't correct for the bias offset).
As you can see the output offset is less when the resistor is in the plus input.

I will leave the explanation of the Vout offset values displayed as an exercise for the reader.
Correct answer gets an attaboy.

View attachment 101116

What is Vt and Vh?

 

What is offset adjustment like in your circuit?
This is an artificial circuit, obeying artificial rules following artificial parameters you set up. This is not a reality. If you imagined a difference in the design there will be a difference.

Building the circuit as you describe it (grounded input, no signal in), yes I see a difference. With the resistor I get 7 mV out. With the wire I get 22 mv out. If I input 25 mV I get -2.72 V (plus or minus a lot of drifting) out with or without the resistor. I can't measure the difference. A limitation of my ability to test the circuit, more than a difference in the circuit. But if I can't see a difference is there really a difference?

 

This simulation isn't different but it shows the effects of voltage and current offset at the output of the 741 model I have with a gain of 100.

The model input voltage offset is [V(1,2)] is 1.02mV (typical for the data sheet I have).
The model plus input bias current is 58nA and the minus input bias current is 102na for a current offset of 44nA (my data sheet shows typical values of 80nA for the bias and 20nA for the bias offset).

I switched a 10kΩ resistor in and out (Vsw high is resistor shorted) of the plus input to correct for the bias current (but it can't correct for the bias offset).
As you can see the output offset is less when the resistor is in the plus input.

I will leave the explanation of the Vout offset values displayed as an exercise for the reader.
Correct answer gets an attaboy.

View attachment 101116

Using your test setup with an LF351 I do not notice a change of 1 mV in the output. So is my conclusion the same? Yes, the resistor does make a difference in bipolar input op amps as bad as the LM741, even if the error is less than drift at a gain of 100. The resistor serves no purpose at all in FET input op amps, except perhaps at very high gain or very high input resistances (above, well above, 1 M ohm).
Are there any circuits that have meg ohm input resistances AND gains of 1,000?
Would we see a difference using an LF351 as a voltage comparator, using an analog voltage out?
In a microcontroller based system with a 10-bit ADC with a 5 V input you can't tell a difference smaller than 5 or 10 mv, considering noise, bobble and drift..
so what is the point of the resistor?

 

What is offset adjustment like in your circuit?
This is an artificial circuit, obeying artificial rules following artificial parameters you set up. This is not a reality. If you imagined a difference in the design there will be a difference.

The model I used had no offset adjustment. Few op amps use that anymore.

Depends upon how you define "artificial".
It's a simulated circuit using models for the components.
I do know it's not "reality". No surprise there.

What do you mean "imagined a difference"?

 

...............
so what is the point of the resistor?

For the umpteenth time, if the bias correction resistor corrects the offset less than your smallest signal of interest, then you don't need it.
How many times does that need to be said.

 

The model I used had no offset adjustment. Few op amps use that anymore.

Depends upon how you define "artificial".
It's a simulated circuit using models for the components.
I do know it's not "reality". No surprise there.

What do you mean "imagined a difference"?

Re: "Few op amp do that anymore"
Very true. The LM741 does / did. FET input (those newer) op amp don't need it ... or the resistor of our topic.

Re: simulation

The simulation was designed to fail. You used mythical data for a mythical chip using a mythical circuit and the part failed, just as designed. The circuit you chose, with no input signal, had no application in the real world.
THE BIAS PROBLEM IS, IN DEED, REAL. I will admit that, but it is only a problem with older bipolar op amps and most often isn't detectable. In the inverting op amp with a voltage applied I can't measure the difference. A failure more of my ability to measure, I suspect. With a microcontroller using a 10-bit ADC with a 5 V level, the increments of the ADC are about the same as the LM741 bias errors. An 8-bit ADC has 20 mV per step. So, who cares? The LM741 is only used in school to demonstrate the op amp errors. Does anybody but a hobbyist like myself actually use an LM741 any more?
What happens if you repeat your simulation using an LF351 or some newer op amp with pA bias?

 

..................
The simulation was designed to fail. You used mythical data for a mythical chip using a mythical circuit and the part failed, just as designed. The circuit you chose, with no input signal, had no application in the real world.

Where do you get this mythical crap?
The offset and bias values in the model are near nominal values for the old 741.
The circuit is a typical inverting op amp configuration with the inputs grounded to show the offset (the same as if you had a zero input signal).
I didn't design it to fail (why do you say it failed?). It just demonstrates the offsets that a real circuit could see.
It does indeed have an application in the real world.
Are you a troll?

 

I vaguely remember some books suggesting a resistor be added to an inverting amplifier design, but I never understood why. I remember it suggesting a value for R1 to be equal to the parallel resistance of the other two resistors. In retirement now I have the time to pursue answers to questions I never found an answer to. So what is R1 supposed to accomplish? I built a circuit with and without R1 and see no difference.
I used an LM741 with an offset pot if it makes a difference.
Is it not apparent in my circuit?
Is there an equivalent resistor in a non-inverting amplifier?

Well the offset pot allows you correct for the problem, so since you have one, you needn't worry about R1.

To answer your question though, as to why it makes no difference, it is likely because the values of the resistors you are using in your amplifier are relatively small. If you try to make an amplifier with a gain of 2 using a 5Meg and a 10Meg resistor the bias currents may become more apparent. Though again, it's just going to create a DC offset in the output, so if you're using an offset pot it doesn't really matter.

 

@hp1729
I don't think you are a troll. I do think you are struggling with something. Maybe you live in an 8 bit world and refuse to accept that the world is a lot more vast then that. Maybe you are angry about something. You are not stupid, but you are stubborn. The world will not yield to your self-indulgence. Please, take a breath.

Oh, by the way, I just bought my grandson a USB mixer that has a 16 bit converter in it. Professional audio regularly uses 24 bit A/D converters.

Also, stop jousting with @crutschow , he has done a lot to show you aspects of bias current that you will not get anywhere else.

 

If you try to make an amplifier with a gain of 2 using a 5Meg and a 10Meg resistor the bias currents may become more apparent.

Tough to see how that could be an every day thing.

You are actually saying the same thing but with different emphasis. And if i may dissect it for you, it would go something like this:
1) bias current does make a difference, and
2) most of the times, that difference doesn't matter.

Some people may emphasize different aspect of the two statements but they are actually very consistent with each other.

 

Where do you get this mythical crap?
The offset and bias values in the model are near nominal values for the old 741.
The circuit is a typical inverting op amp configuration with the inputs grounded to show the offset (the same as if you had a zero input signal).
I didn't design it to fail (why do you say it failed?). It just demonstrates the offsets that a real circuit could see.
It does indeed have an application in the real world.
Are you a troll?

Nominal values are not REAL values. The circuit you suggested has no real application. It is specifically used to exaggerate the bias problem. It does demonstrate the problem but it does not show a real result in a circuit.

 

Instead of arguing with every suggestion that is made, why don't you just conclude that none of this is important, that the text books have no relation to real life, and that the component manufacturers just enjoy wasting time and money characterizing their parts for no good reason. Then stick your head in the sand and ignore the issue. As long as you stay with safe hobbyist circuits that are way away from the bounds where this stuff becomes important, you may well never get bit. When and if you do get bit, then perhaps you will be better prepared to listen.

 

Nominal values are not REAL values. The circuit you suggested has no real application. It is specifically used to exaggerate the bias problem. It does demonstrate the problem but it does not show a real result in a circuit.

Okay, you've answered my question.
You are a troll.

 

Instead of arguing with every suggestion that is made, why don't you just conclude that none of this is important, that the text books have no relation to real life, and that the component manufacturers just enjoy wasting time and money characterizing their parts for no good reason. Then stick your head in the sand and ignore the issue. As long as you stay with safe hobbyist circuits that are way away from the bounds where this stuff becomes important, you may well never get bit. When and if you do get bit, then perhaps you will be better prepared to listen.

How much difference does this resistor make in a simulation using a modern FET input op amp? Are the text book written back when the LM741 was the workhorse? Are the books obsolete?

 

How much difference does this resistor make in a simulation using a modern FET input op amp? Are the text book written back when the LM741 was the workhorse? Are the books obsolete?

Yep, the books are all obsolete and the manufacturers continue to just blindly still characterize their parts because it's just the way it's always been done.

Why even worry about simulations using any kind of op-amp since these are just simulations and not real circuits.

None of this makes any difference at all. You were right all along. Now you can declare victory and be happy.

 

Okay, you've answered my question.
You are a troll.

I am just trying to get a grasp on the subject .. instead of just believing and following procedures. When is this resistor needed? When is this resistor not needed? If I don't question what is said how do I learn? Am I a troll just because I don't believe the same stuff you do?

 

Yep, the books are all obsolete and the manufacturers continue to just blindly still characterize their parts because it's just the way it's always been done.

Why even worry about simulations using any kind of op-amp since these are just simulations and not real circuits.

None of this makes any difference at all. You were right all along. Now you can declare victory and be happy.

Is this stuff a religion to you? Stuff to be believed without question? If you run a simulation do you always accept the results to be the truth? Are all parts exactly as the spice model says?

Don't get me wrong. Simulations are a necessary part of the developing reality as parts move away from being breadboardable. YES, bias problems are real. My question is where and when.

 

I am just trying to get a grasp on the subject ........?

No, you are not. The math and concepts have been explained to you. You refuse to accept it.

Is this stuff a religion to you? Stuff to be believed without question?.....

Religion???? Give us a freakin break!! We model circuits. We build circuits, after a while we know how much to trust the models. It is a circle of learning.

............ My question is where and when..............

Always, its called engineering.