| Analog | Analog Experiments | Oscillators | Optics | HTML Test |
|| Getting Started | Preparations | Basic Analog Circuits ||
|| The Basic Inverting Amplifier | Balancing the Input Offset | Summing Amplifier | ±10 Volt Reference | A Non-Inverting Amplifier ||
|Balancing the Input Offset|
In the previous experiment (The Basic Op Amp Inverter), we learned that there are at least three basic limitations on the accuracy of an op amp circuit: the tolerance of the resistors, input offset voltage and current, and output saturation. We can't do anything about output saturation, and we can only deal with resistor tolerances by selecting resistors of the degree of precision required for a specific application. However, in those cases where it is necessary, we can make use of a feature of the basic 741 op amp to balance out the input offsets and thus remove their influence on the output voltage.
Fortunately, in most applications this is not necessary. This permits the construction of dual- and quad-op amp packages, where the offset adjustment connections are not available outside the package. However, there are some cases where the input offset must be balanced as closely as possible. Therefore, we will examine the required method in this experiment.
The recommended circuit for balancing out the input offset is quite simple, as shown here. The offset null pins (1 and 5) give direct access to the 1K emitter resistors in the input stage, and the offset null circuit is simply a 10K potentiometer connected between them, with its slider connected to the negative power supply. This is equivalent to putting a 5K resistance in parallel with each of the 1K resistors inside the IC. The difference is that we can vary the external resistances by adjusting the potentiometer, until the voltage offset becomes zero.
Since we're dealing with the input stage of a high-gain amplifier, the output voltage will be very sensitive to potentiometer changes. Therefore, we'll use a 15-turn trimpot here.
So how do we tell when we have exactly balanced out the offset? After all, the input offset is internal to the IC, and is rated to be no more than 6.0 millivolts (0.006 V) and 0.2 milliamps (0.0002 A). These require very accurate and sensitive measuring equipment, and are inaccessible in any case.
What we can do is select Rin and Rf to amplify an input voltage of zero. We'll make Rin = 1K and Rf = 100K, and connect the input end of Rin to ground. This will give the overall circuit a gain of 100, and assure us that the correct output voltage should nevertheless be 0.000 volt. Any output voltage will be due to offsets in the op amp itself, and we will use the 10K trimpot to balance them out as closely as possible.
To construct and test the offset balancing circuit on your breadboard, you will need the following experimental parts:
Select an area on your breadboard socket that is clear of other circuits. You'll need to mount the 741 IC so that its pins are in line with contacts on the bus strips. Then refer to the image and text below and install the parts as shown.
This project can be placed anywhere on the right side of your breadboard socket. We selected the right end simply for convenience, but you can move it if you like.
Click on the `Start' button below to begin. If at any time you wish to start this procedure over again from the beginning, click the `Restart' button that will replace the `Start' button.
You should have some 0.5" orange jumpers left over from earlier experiments. If not, prepare a 0.5" orange jumper in the manner you have used in the past. Install this jumper as shown in the assembly diagram to the right.
Click on the image of the jumper you just installed to continue.
You should also have some 0.5" blue jumpers left over from previous experiments. Either use one of these or prepare a new one, and install this jumper in the location shown to the right.
Again, click on the image of the jumper you just installed to continue.
Ordinarily, we call for jumpers longer than 0.1" to have insulation on them to help prevent inadvertent contact between wires, and to help identify the function of that jumper later on. However, in this particular case, we need some longer jumpers with no insulation. We'll see why shortly.
Accordingly, prepare a 0.2" bare jumper and install it in the location shown in the assembly diagram.
As before, click on the image of the jumper you just installed to continue.
In a similar manner, prepare a 0.6" bare jumper and install it in the location shown to the right.
Again, click on the image of the jumper you just installed to continue.
Next, prepare a 0.3" bare jumper in the same way, and install it in the location shown in the assembly diagram.
Once more, click on the image of the jumper you just installed to continue.
Locate a 741 op amp IC and install it as shown to the right. Be sure the notch indicating pin 1 is to the left, and that none of the pins bend up under the body of the IC as you install it.
Note that for this particular experiment you must use a 741 IC, not half of a 1458. Only the single 741 op amp has the offset adjustment connections you need to perform the experiment.
Click on the image of the IC you just installed to continue.
Locate a 1K, 1% resistor (color code brown-black-black-brown) and install it in the location shown in the assembly diagram. Note that it is not necessary to form the leads or to clip them short for this experiment; we're demonstrating a concept here, and will not be using this circuit in the future.
Click on the image of the resistor you just installed to continue.
Locate a second 1K, 1% resistor (color code brown-black-black-brown) and install it as shown in the assembly diagram. Again, it is not necessary to form or cut the leads for this experiment.
Again, click on the image of the resistor you just installed to continue.
Locate a 10K, 15-turn trimpot. Look at the underside of this component. You should find raised ridges at the ends, which will keep the body of the trimpot off the surface of the breadboard socket. This construction is intended to allow the trimpot to clear copper foil traces and irregularities on a printed circuit board, but also allow us to run bare jumpers under its body. There isn't enough space for insulation, however, which is why we specified bare wires in this experiment.
Install this trimpot as shown in the assembly diagram, with the three connecting pins inserted according to the gold squares shown to the right.
Click on the image of the trimpot you just installed to continue.
Finally, locate a 100K, 1% resistor (color code brown-black-black-orange) and install it as shown in the assembly diagram. As with the 1K resistors, you need not form or clip the leads, but you should make sure that none of the resistor leads touch each other or any of the bare jumpers you have already installed.
As always, click on the image of the resistor you just installed to continue.
This completes the construction of your experimental circuit. Check your assembly carefully against the figure to the right, and correct any errors you might find. Then, proceed with the experiment on the next part of this page.
Turn on your voltmeter and set it to measure voltages in the range of ±20 volts, and connect it to monitor the output voltage of the 741 op amp at pin 6 (the top end of the 100K resistor). Turn on power to your experimental circuit, and note the output voltage of this circuit. If the trimpot is still at one end of its range from the last experiment, the output voltage may well be quite high. This is not a problem; simply note it for now.
Adjust the trimpot over its range and note the effect on the output voltage. Then, readjust the trimpot to reduce the output voltage to below ±1.00 volt.
Reduce the range on your voltmeter to measure voltages up to ±2 volts, and then continue to adjust the trimpot to reduce the output voltage to zero.
If your voltmeter has a 200 mV range, switch down to that and use the trimpot to adjust the output voltage to as close to zero as you can. By now the adjustment will be very sensitive to slight changes, so you may have difficulty balancing out the last millivolt. If so, don't worry about it; just do the best you can.
When you have reduced the output voltage as closely as possible to zero, leave the trimpot alone and simply observe the output voltage for a few minutes. What does this observation show you?
Finally, remove the trimpot from the breadboard socket and set it aside. What is the output voltage now? Keeping in mind the op amp gain of 100, what was the effective initial input offset voltage?
When you have made your determinations, turn off the power to your experimental circuit and voltmeter and compare your results with the discussion below.
The 10K trimpot allows considerable control over the input offset errors inherent in the 741 op amp. You should have been able to adjust the output voltage over a range between ±3 and ±4 volts. Thus, you can certainly adjust the trimpot to balance the circuit for a zero volt output.
As you continued to reduce the output voltage and set your voltmeter to more sensitive ranges, you found that adjusting for precisely zero volts at the output was a bit more difficult. At the millivolt range, even a very slight adjustment of the trimpot caused a significant change in output voltage. Nevertheless, you should have been able to set the output to some value less than ±1 mV.
It is possible to get more sensitive adjustments by replacing the 10K trimpot with two fixed 4.7K resistors and a 1K, 15-turn trimpot. This will reduce the overall adjustment range to about 10% of its original range, and allow greater adjustment sensitivity over the reduced range. However, as you discovered when you watched the output voltage for several minutes, this is not helpful in this case. The op amp output voltage did not remain constant, but kept shifting up and down by a millivolt or more.
This behavior is called drift, and the output voltage of the op amp will continue to drift over time, and with changes in temperature. As a result, you can minimize the offset, but you cannot completely eliminate it over a period of time. You'll need to use other techniques to minimize and compensate for drift in a critical op amp circuit.
When you removed the trimpot, the output voltage jumped to an amplified representation of the inherent offset. Our sample 741 jumped to an output of 79.5 mV, indicating that the input offset voltage and current of this particular IC combine to a result that is slightly less than 0.8 mV. This is well within the maximum ratings of this IC, and is small enough that we can ignore it in the remaining experiments covered in these pages.
When you have completed this experiment, make sure power to your experimental circuit and voltmeter is turned off. Remove all of your experimental components from the breadboard socket and put them aside for use in later experiments.
All pages on www.play-hookey.com copyright © 1996, 2000-2015 by
Please address queries and suggestions to: firstname.lastname@example.org