Opamp Amplifiler

Texas Instruments - TI.
com
Follow Us |
Table of Contents
Introduction
Chapter 1: Operational amplifiers
When is good enough good enough?
Just use a 100 resistor: Understanding a rule of thumb for oscillating amplifiers
Where did all that racket come from?
Is your amplifier offset way out of whack?
A good holiday-season project
Don't be fooled by your amplifier's bandwidth
Sallen-Key lowpass-filter stopband limitations
Choosing antialiasing-filter amplifiers
Closer to real-world analog filters
Simply an adjustable low-pass filter
Photo-sensing circuits: The eyes of the electronic world are watching
Transimpedance-amplifier stability is key in light-sensing applications
Transimpedance-amplifier noise issues
Transimpedance-amplifier application: The pulse oximeter
Remote photo sensing
Single-supply amplifier outputs don't swing rail to rail
The non-negotiable single-supply operational amplifier
What does "rail to rail" input operation really mean?
Rail-to-rail input amplifier application solutions
Voltage- and current-feedback amps are almost the same
Chapter 2: INAs and PGAs
Understanding CMR and instrumentation amplifiers
Common-mode range can bite hard
Instrumentation amplifier input-circuit strategies
Wringing out thermistor nonlinearities
Author's biography
Introduction
The Best of Baker's Best Amplifiers
Introduction, by Bonnie Baker
Amplifiers have been around for a long time. These beasts were just coming alive in the 1930s 1940s, at the hands of
several individuals including Harold S. Black, George Philbrick, and H. W. Bode. Given the long time span between these
beginning years and the present, you would think that we have finally come to terms with understanding this five-terminal
http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:43:25 AM]

Best of Baker's Best - Amplifiers eBook - TI.com
Follow Us |
Table of Contents
Introduction
Author's biography
Introduction
Introduction by Bonnie Baker
Amplifiers have been around for a long time. These beasts were just coming alive in the 1930s - 1940s, at the hands of
several individuals including Harold S. Black, George Philbrick and H. W. Bode. Since the advent of the solid state
operational amplifier (op amp), enabled by the invention of silicon transistors and the integrated circuit (IC), the amplifier has
increasingly become a staple for signal conditioning in most electronic systems. Given the long time span between these
device. Furthermore, isn't there something else out there that has successfully obsoleted these devices?
Some people I have worked with are surprised when I tell them that the op amp is not dying out, but rather they are dynamic and permanent
fixtures in the electronics world. This is particularly true as long as the real-world remains in the analog domain. Amplifiers, including
instrumentation amplifiers (INA) and programmable gain amplifiers (PGA), are here to stay as standalone devices and as integrated
functions inside complex ICs. And, to take it a step further, there is still a lot to learn about amplifiers and their appropriate placement in your
application circuits.
The following excerpts from the EDN Baker's Best column represent the best of the amplifier short articles. They have been compiled for
your convenience. Please enjoy your read through and let me know what you think!
1 Back to top

If you are having difficulty making product-selection decisions in a consumer circuit, such as the temperature-sensor circuit in Figure 1, you
can quickly solve this problem by choosing the absolute best performing parts for each socket. Is this statement true or false? Using this
type of logic may give you a confident feeling that your circuit will work correctly the first time. However, following such logic goes only so far
when you try to justify the cost-versus-performance factors of the products you are using.
In Figure 1, note that a 12-bit converter is at the end of the signal chain. So, are the highest performance analog products in front of the
ADC appropriate? How do you determine which products are good enough for your system? Avoiding production-floor notifications or field
failures may be your definition of "good enough."
Instead of choosing the best products, you can use the RSS (root-sum-square) algebraic approach. One criterion is to keep the signal within
the dynamic range of the full-scale range of the ADC. The product characteristics that influence the extent of the dynamic range are the
system's cumulative offset and gain errors.
As an example, assume that the maximum offset error of IC1 and IC2 is 0.5 mV. The offset error of the ADC is 1 LSB or 1.22 mV. (The
full-scale range of the ADC is 5V.) The gain error of both the sensor cell and the IC1 amplifier configuration depends on the 1% maximum
resistor tolerances as well as on a maximum sensor-resistor tolerance of 2%. The ADC's contributed gain error is 0.098% or equals 4.9 mV
maximum at full scale.
To determine the dynamic-range limitations of the circuit, if you combine all of these terms, you would calculate the combined RSS value of
offset and gain, bringing these errors to the ADC's input. With the RSS formula, you take the square root of the sum of the squares of
several terms that are statistically independent. You cannot use an RSS formula with entities that have correlated variations that are not
statistically independent.
For instance, the worst-case sensor-resistive offset error would be 94 mV10V/V. The contribution of the amplifier-gain stage, IC1 , is 500
V10, the filter-stage (IC2 ) offset error is 500 V, and the ADC (IC3 ) offset error is 1.22 mV. The cumulative possible offset error at the
ADC's input is =940 mV. This calculation illustrates that the sensor cell contributes the most error with little
impact from the amplifiers or the ADC. Using the same logic, you would use the RSS formula to determine gain-error contribution that limits
the dynamic range from the four stages in this circuit.
So, during your first consumer-product-selection attempt, you can use RSS calculations. These calculations can assist you in making logical
and economical product decisions. Once you take this first step, make sure you use the same evaluation technique in your manufacturing
process to quantify the effects of the processessuch as solder reflowthat you impose on these devices and the end-of-life effects due to
environmental exposures.
References
1. Sandler, Steven M, "A Comparison of Tolerance Analysis Methods," AEi Systems LLC, 1998.
2 Back to top
So there I was, a new hire for a leading-edge analog-electronics company. As it was my first job, I was wide-eyed and excitedconfronting
new problems left and right. One problem I ran into involved the stability of an amplifier circuit. In this application, a buffer-amplifier circuit,
with a capacitive load, sang like a bird. Because I had a huge community of experts around me, I ran around and asked for advice. The
words of wisdom that came down to me were, "Oh, put a 100 load resistor between the amplifier output and the load capacitor." When I
asked why, the engineer said, "Just do it. Trust me; it will work."
So, I built a new circuit as suggested, and, lo and behold, the circuit still oscillated. This new circuit was still singing, but it was producing a
new frequency. I returned to the engineer who gave me the first bit of great advice. His recommendation: "Change the 100 resistor to a
500 resistor," still with no explanation. It solved my problem, but, given my work load, I did not return to his suggestion for several years.
Now, it has come back to haunt me. I need to know what is really going on!
What I did not understand then but understand now is that a capacitor and a resistor that hang on the output of an amplifier change the
amplifier's open-loop-gain curve. The combination of the load capacitor, CL ; the load resistor, RL ; and the amplifier's open-loop resistance,
RO, introduces a pole to the open-loop-gain curve, and CL and RL then introduce a zero to the open-loop-gain curve (Figure 1). Creating
this pole and zero does not disrupt the amplifier's stability as long as they cancel each other out before the open-loop-gain curve crosses the
closed-loop-gain curve. If the open-loop-gain and closed-loop-gain curves cross with a 40-dB/decade closure rate, the amplifier circuit will
be marginally unstable or, worse yet, will oscillate.
You can find the pole and zero locations in this circuit in the following equations:
What have I learned from this situation? It pays to understand why an engineer's rule of thumb works. If you comprehend the general
guidelines, you will be OK. If you are not on top of the explanation, however, it will come back to bite you.
References
1. Oljaca, Miro, and Bonnie Baker, "Start with the right op amp when driving SAR ADCs," EDN, Oct 16, 2008.
3 Back to top

Since arriving in the market, CMOS single-supply amplifiers have been beneficial to single-supply-system designers worldwide. Key players
affecting the THD+N (total-harmonic-distortion-plus-noise) characteristics of dual-supply amplifiers are input noise and output-stage
crossover distortion. The THD+N performance of a single-supply amplifier also originates in the amplifier's input and output stages.
However, the input stage's impact on THD+N complicates the nature of this specification with single-supply amplifiers.
Several types of single-supply amplifier topologies can accept input signals across the power supplies. In the complementary-differential-
input-stage topology, when the amplifier's inputs are near the negative rail, the PMOS transistors are on, and the NMOS transistors are off
(Figure 1). When the amplifier's inputs are closer to the positive rail, the NMOS transistors are on, and the PMOS transistors are off.
This design topology has significant variations in the amplifier's offset voltage across the common-mode input range. In the input region
near ground, the PMOS transistor's offset error is dominant. In the region near the positive power supply, the NMOS-transistor pair
dominates the offset error. Both pairs are on as the amplifier's inputs pass between these two regions. The end result is that the input offset
voltage changes between the stages. When the PMOS and NMOS transistors are both on, the common-mode voltage region is
approximately 400 mV. This crossover-distortion phenomenon affects the amplifier's THD. If you configure the complementary-input
amplifier in a noninverting configuration, the input crossover distortion can affect the amplifier's THD+N performance. For instance, in
Figure2 the THD+N is 0.0006% if you avoid using the input transition. If the THD+N tests include the amplifier's input crossover distortion,
the THD+N is 0.004%. You can avoid this type of amplifier crossover distortion by usingg an inverting configuration.
Another major THD+N contributor can be the operational amplifier's output stage. The output stage of a single-supply amplifier usually has
an AB topology. As the output signal sweeps from rail to rail, the output stage displays a crossover distortion similar to the input-stage
crossover distortion, in that the output stage switches from transistor to transistor. Generally, a higher level of quiescent current through the
output stage reduces the amplifier's THD. The amplifier's input noise is another contributor to the THD+N specification. A high level of input
noise, high closed-loop gains, or both can increase the amplifier's overall THD+N level.
To optimize a complementary-input-single-supply amplifier's THD+N performance, place the amplifier in an inverting-gain configuration and
keep the closed-loop gain low. If the system requires the amplifier to be configured as a noninverting buffer, an amplifier with a single-
differential input stage and charge pump is a more appropriate choice.
References
"OPA350, OPA2350, OPA4350 High-Speed, Single-Supply, Rail-to-Rail Operational Amplifiers, MicroAmplifier Series," Texas
1.
Instruments, January 2005.
"OPA363, OPA2363, OPA364, OPA2364, OPA4364, 1.8V, 7MHz, 90dB CMRR, Single-Supply, Rail-to-Rail I/O Operational Amplifier,"
2.
Texas Instruments, February 2003.
4 Back to top

Have you ever spent a great deal of time selecting the perfect operational amplifier for your circuit, only to find that the offset voltage is
wrong at the manufacturer's bench-specified input? What if you find that it is more than 10 times higher than specification in your application
circuit? Do you send the chip in for failure analysis or just toss the chip out and have another look at your list of amplifiers? As an
alternative, I suggest that you try to explain the offset error by re-examining your amplifier's specifications.
If you are using your amplifier as the key component in a transimpedance amplifier, an analog filter, a sample-and-hold circuit, an integrator,
a capacitance transducer, or any other circuit with high-impedance components around your amplifier, you might find that the amplifier's
input-bias current creates an offset-voltage error through the resistors in your circuit.
In the bipolar-amplifier days, the term "input-bias current" was an accurate descriptor, and it still is. A bipolar amplifier's input-bias current is
the same as the base current of the NPN or PNP transistors at the input of the amplifier. The magnitude of the bipolar amplifier's input-bias
current ranges from a few nanoamperes for low-power devices to hundreds of nanoamperes for higher-power devices.
The term "input-bias current" loses its meaning when you look at JFET or CMOS input amplifiers. With these types of amplifiers, the current
sinking or sourcing from the amplifier's input pins is actually the leakage current from the input-ESD (electrostatic-discharge) cells (Figure
1). A more accurate descriptor for this current error is "input-leakage current." The magnitude of leakage current with JFET or CMOS
amplifiers is less than 1 pA at 25C. This specification is independent of the common-mode voltage and the magnitude-amplifier power.
Almost all amplifiers have ESD cells for protection from an ESD event, but you will never see ESD-leakage current in bipolar amplifiers. The
input-bias current swamps out the picoampere leakage current from the ESD cells.
Input-bias and input-leakage current can change over temperature. However, depending on the operational-amplifier design, the bipolar
input-bias current can be fairly stable. The JFET and CMOS input amplifiers may not be, however. Because the leakage current is from the
reverse-biased ESD diodes, the leakage current increases approximately two times per 10C change.
In ensuring that the input-leakage current remains low with JFET and CMOS amplifiers, you must understand the impact of your PCB
(printed-circuit board) on the picoampere levels of current. For instance, a small amount of dust, oil, or water molecules can increase
leakage current and masquerade as input-bias current. The good news is that, if you exercise special care, you can build a PCB that will
adhere to a 1-pA performance specification.
The most effective way you can reduce or minimize the effects of input-bias or input-leakage current is to check your circuit configurations.
As you examine your circuits, look at the voltage characteristics of each node and make sure that you understand the impact of all of the
current paths in your circuit.
5 Back to top

Which of the bulbs on a Christmas tree is the brightest? If you had the time and desire to answer this question, you could use a single
photodiode to determine the brightness of your bulbs. Finding that brightest bulb among the many and the background light would be a
laborious task, however, unless you expanded your design task to using two photodiodes. The two photodiodes let you find a light's position
by monitoring the difference between their output signals.
If you use three op amps in a differential-photodiode-amp configuration, you will see greater accuracy in proximity and difference (Figure
1).
The configurations of A 1 and A 2 act as traditional current-to-voltage converters or transimpedance amps. A 3 and R2 form a difference amp,
subtracting the output voltages of A 1 and A 2 . In this circuit, the incident light on the photodiode causes current to flow through the diode
from cathode to anode. Because the inverting input of A 1 and A 2 has high impedance, the photodiode's currents flow through the R1
feedback resistors. The voltage at the inverting input of the amp tracks the voltage at the amp's noninverting input.
Consequently, the amp's outputs change in voltage along with the IR drop across the R1 resistors. The output voltages of A 1 and A 2 contain
both difference and common-mode signals. A 3 rejects the common-mode signal and delivers the differential-voltage signal to the circuit
output at V OUT .
The key performance parameters for A 1 and A 2 are input capacitance, bias current, offset, noise, and temperature drift. The goal is to select
amps in which these parameters are as low as possible. A 1 and A 2 require low-inputcurrent CMOS or FET op amps.
You can implement a differential amp discretely or with an off-the-shelf product. As long as the resistors surrounding A 3 are equal, the dc-
transfer function of this circuit is 1V/V.
If the resistors around A 3 are not equal, a noticeable gain error can occur between the two input signals. You can easily compensate for this
type of error by replacing any of the four resistors with a potentiometer. More important, however, this type of mismatch can introduce
nonlinearities in the system when the common-mode voltage of the two inputs changes. You define the common-mode voltage of the input
signals as (VA1OUT +V A2OUT )/2. Ideally, the differential amp rejects common-mode-voltage changes. The calculated CMR (common-mode-
rejection) error due to resistor mismatches is 100(1+R 2 /R1 )/(% of mismatch error).
An equal illumination on the two photodiodes makes the output voltage 0V. D1 and D2 respond linearly to illumination intensity, which
makes the magnitude of the output voltage a direct measurement of the difference between the direct light impinging on D1 and D2 .
A single photodiode provides some measure of a light's intensity through the magnitude of the diode's output signal. However, background-
light conditions can influence this magnitude, requiring calibrated and impractical measurement conditions. Adding a matched photodiode
and monitoring the difference between the two diode outputs removes the equal offsets the two diodes produce. Background light adds only
an offset to the photodiode's outputs, and the differential amp removes this effect.
References
1. Photodiode Monitoring with Op Amps, sboa035. Texas Instruments, January 1995.
2. Grame, Jerald. Photodiode Amplifiers: Op Amp Solutions. McGraw-Hill, ISBN: 0-07-024247, 2006.
3. Precision Unity Gain Differential Amplifier, SBOS145, Texas Instruments, August 1993.
6 Back to top

As we design our SAR-converter analog circuits, we may be tempted to match the voltage-feedback amplifier's data-sheet bandwidth to the
bandwidth of our analog signal source. We keep the amplifier bandwidths as low as possible, because faster amplifiers on the board
potentially can produce layout headaches. These layout headaches come from putting higher-speed amplifiers in the circuit that are
producing fast rise and fall times. The faster signals have the potential to produce EMI signals that other devices or traces on the board can
receive, resulting in unexpected noise.
Given this scenario, it makes sense to keep the bandwidth of the circuit's amplifiers as low as possible. With a maximum signal frequency
input from dc to 20 kHz, one would think that the required amplifiers would have very low unity-gain bandwidths or a gain-bandwidth product
(GBWP).
On the contrary, you need higher-speed amplifiers in the circuit for two basic reasons. The primary reason is to accommodate the lost
bandwidth in amplifier circuits that have a closed-loop noise gain greater than one. The other reason is to make sure the bits in your system
at the end of the signal path can reliably convert the signal throughout the entire system's frequency range.
Figure 1 shows an example of how the amplifier's closed-loop gain affects the system's overall bandwidth. The GBWP of this amplifier is 20
MHz. The amplifier's closed-loop gain is 10V/V, or 20 dB. In our circuit, we already require an amplifier that has a bandwidth that is 10 times
higher than our input signal. You can see this by comparing the intersect of the closed-loop gain (f 3 dB ) with the amplifier GBWP frequency
(f GBWP ).
Figure 1: The system bandwidth becomes lower due to the closed-loop (ACL) intercept
point with the amplifier's open-loop gain (AOL ) and the need to accommodate bit accuracy
across the entire frequency range at f 1 .
The amplifier characteristics shown in Figure 1 appear to be a perfect fit, but we are looking for the correct amplifier for a 16-bit system.
One sticky point with this circuit is that the amplifier's closed-loop gain is not equal to 20 dB all the way up to 2 MHz. In fact, the closed-loop
gain at 2 MHz is 17 dB. This is down 3 dB, which is approximately 70.7% lower than the closed-loop gain, or a 29.3% increase is needed to
get up to that curve.
Of course, the circuit's closed-loop gain does not instantaneously change from 20 dB to 17 dB at 2 MHz. Instead, it gradually gets closer to
the closed-loop gain curve starting about a decade before 2 MHz. A 20-MHz bandwidth for an amplifier should be good enough, but let's do
the math.
This intersection has a simple, first-order attenuation. The correct formula to determine the attenuation back from the 3-dB point toward 0 Hz
is f 1 =f 3 dB where A CL1 is the target closed-loop gain, A CL_DC is the closed-loop gain at dc, f 1 is the target
frequency, and f 3 dB is the corner 3-dB frequency of this amplifier system. If you make A CL1 equal to 9.9988752, which produces a
0.0112% error at the full-scale frequency, the bandwidth of the closed-loop system is approximately 100 times lower than f 3 dB .
With our amplifier circuit, we are not so lucky. We have a 20-kHz signal that we need to increase by 10 times. At the end of the signal chain,
we have a 16-bit converter. We find that a 20-kHz amplifier does not work for our circuit. With a signal gain of 10V/V, or 20 dB, the amplifier
bandwidth needs to be at least 10 times higher than the signal. We also find that the amplifier bandwidth needs to be at least 100 times
higher to maintain ADC integrity. This situation places our amplifier unity-gain bandwidth at 20 MHz. So much for lower- speed amplifiers.
References
1. High-Speed, Single-Supply, Rail-to-Rail OPERATIONAL AMPLIFIERS MicroAmplifier Series, Texas Instruments, March 1999.
7 Back to top

When you design an analog, lowpass, antialiasing filter, you would expect its gain amplitude to continuously decrease beyond the filter's
cutoff frequency. For the most part, this assumption is a safe one, but it's not necessarily true with the classic Sallen-Key lowpass-filter
design. The Sallen-Key filter attenuates any signals in the frequency range above the cutoff frequency to a point, and then the response
turns around and starts to increase in gain with frequency.
Figure 1 illustrates the behavior of three Sallen-Key lowpass filters using signal-supply amplifiers. In the top three curves, the diagram
captures the open-loop gain of each amplifier as the response crosses 0 dB. During this test, the configuration for all three amplifiers is a dc
noise gain of 1000V/V, or 60 dB. In the diagram, the bandwidth of op amps A, B, and C are 38 MHz, 2 MHz, and 300 kHz, respectively.
A second set of three curves in Figure 1 shows the frequency response of second-order, Sallen-Key lowpass filters for each amplifier. The
data indicates that the lowpass filters are performing as you would expect for a little more than a decade after the cutoff frequency of 1 kHz.
Although the approximation method does not impact or correct this unexpected behavior, these filters use a Butterworth design. After the
cutoff frequency, all three of the filter's responses show a slope of -40 dB/decade. You would expect this response from a second-order
lowpass filter. Then, at some point, the filter gain begins to increase at a rate of 20 dB/decade. The difference in the frequency response at
the point at which the three amplifiers change to a positive slope depends on the amplifier's output impedance. As the amplifier's open-loop
gain decreases, its closed-loop output resistance increases. Eventually, each filter's response flattens at the 0-dB crossing frequency of the
op amp's open-loop gain. It is no coincidence that the "flattening" of the filter response occurs at this crossing. As the frequency increases
beyond this point, the amplifier's gain is less than 0 dB.
If you use a Sallen-Key lowpass filter, some characterization is in order. You can reduce the impact of the upward trend in the filter's
response by following the offending active filter with a passive RC lowpass filter. The caveat to this action is that the following filter may
interfere with the phase response of your intended filter, which may cause additional ringing in the time domain. Further, this action also
creates a stage whose output is not low-impedance.
Alternative filters can solve the problem without adding an RC filter. When an inverting filter is an acceptable alternative, you can use a
multiple-feedback circuit, which does not display this reversal in the gain response at higher frequencies and does not swing the input
stage's common-mode voltage.
8 Back to top

When you digitize an analog signal, you use a lowpass filter to prevent aliasing errors from out-of-band noise. Doing so attenuates
superimposed, high-frequency noise on the analog signal before it reaches the ADC. If the noise on the input signal is more than half the
sampling frequency of the converter, the magnitude of that noise stays the same, but the frequency changes as it aliases back onto your
signal of interest. You cannot use a digital filter to reduce in-band noise after digitizing the signal.
Selecting the correct operational amplifier for an active lowpass-filter circuit can appear overwhelming as you read an amplifier's data sheet
and view all of the specifications.
Before selecting the amplifier, however, you need to determine the filter-cutoff frequency, f CUT (or -3-dB frequency). You can use filter-
design programs to determine the filter's capacitor and resistor values.
Next, you should initially consider only two important specifications when selecting an amplifier for your active lowpass filter: gain-bandwidth
product and slew rate. In Figure 1, for Q i of less than 1, the gain-bandwidth product of the amplifier, f AMP, must be at least
100gainf CUT k i , where Q i is the quality factor of the ith partial filter and ki is the ratio of the partial-filter-corner frequency to the overall-
filter-corner frequency. For Q i greater than 1, f AMP=100gain(f CUT /a i ) , where ai is the ith coefficient in the partial-
filter-transfer function. The gain-bandwidth product of the operational amplifiers appears in the specification table of the respective product-
data sheet.
You should also evaluate the effects of amplifier slew rate. Doing so ensures that your filter does not create signal distortions due to slew
limitations. The slew rate depends on internal IC currents and capacitances. When you send large signals through the amplifier, internal
currents charge these internal capacitors. The speed of this charging process depends on the amplifier's internal resistances, capacitances,
and currents. To ensure that your active filter does not enter into a slew condition, you need to select an amplifier such that the slew rate
(V OUT P-Pf CUT ), where V OUT P-P is the expected peak-to-peak output-voltage swing below your filter's cutoff frequency.
The most common topologies for active, second-order, lowpass filters are the noninverting Sallen-Key and the inverting multiple feedback. If
you need a higher order filter, you can cascade both of these topologies.
When using the Sallen-Key circuit, input-common-mode-voltage range (VCMR) and input bias current (I B) can also affect you. In this
configuration, V CMR limits the range of your input signal. Additionally, the input bias current conducts through the external source
resistance. The voltage drop that the input-bias-current error causes appears as an additional input-offset voltage. Also, be aware that this
circuit has high-frequency feedthrough.
References
Bishop, J, B Trump, and RM Stitt, "FilterPro MFB and Sallen-Key Low-Pass Filter Design Program," Application Note (SBFA001A),
1.
Texas Instruments , November 2001.
2. Mancini, Ron, Op Amps for Everyone, ISBN-0-7506-7701-5, Elsevier-Newnes, April 2003.
9 Back to top

There are two basic filter topologies that everyone recommends when they start to design their signal chain: multiple-feedback (MFB) or
Sallen-Key topologies. But what are the differences and why would you choose one over the other?
The MFB topology (sometimes called infinite gain or Rauch) is often preferred because it has low sensitivity to amplifier variations, such as
the open-loop gain or input range (Figure 1). The MFB topology creates a second-order filter. A MFB single-stage (second-order filter)
provides two poles while inverting the signal. In designs where an even number of stages is required, such as a fourth- or eighth-order filter,
the output polarity is the same as the input signal. A sixth-order filter consisting of three MFB stages inverts the signal three times, resulting
in an inverted output after the third stage. This inversion may or may not be a concern in the application circuit, particularly if you have
differential input/output stages.
Figure 1: MFB second-order single-supply low-pass filter. The gain of this circuit is equal
to R2 /R1 , where V CM =V +/2.
In Figure 1, the gain circuit is negative and equal to the resistor ratio of R2 and R1 . This arrangement allows a variety of negative gains.
Generally, the MFB topology allows low sensitivity to component tolerances. The value required for C1 in an MFB design can be quite low
while the designer is trying to use reasonable resistor values. The caveat is that low capacitor values can result in significant errors due to
parasitic capacitance and the input capacitance of the amplifier.
There are instances where the Sallen-Key topology is a better choice. A single, second-order Sallen-Key stage also provides two poles but
is a non-inverting circuit. This may be preferable over a MFB single stage, but this is not the only potential advantage. As a rule of thumb,
the Sallen-Key topology is better if:
1. gain accuracy is important, and

2. a unity-gain filter is needed, and
3. a low Q is required (for example, Q < 3)
At unity-gain, the Sallen-Key topology has excellent gain accuracy. This is because the op amp is in a unity-gain buffer configuration with a
high open-loop gain. Consequently, the gain is independent of the resistors in the circuit. With the MFB topology, the R2 /R1 resistor ratio and
the resistor errors determine the gain. The unity-gain Sallen-Key topology also requires fewer componentstwo resistors versus three for
the MFB (Figure 2).
Figure 2: A Sallen-Key second-order, dual-supply, low-pass unity gain filter. The gain of
this circuit is equal to 1V/V.
Your circuit also may require a low Q, the quality factor. The relationship between Q and the damping factor is Q = 1/2 , where is the
damping factor. The higher the Q, the more easily the circuit oscillates, particularly at the 3 dB corner frequency.
Figure 3: Sallen-Key second-order, single-supply filter. The gain of this circuit is 1+R 2 /R1 .
V CM =V +/2.
The Sallen-Key topology may be preferable for low-Q, high-frequency filter sections.
Selecting the correct amplifier, for these circuits, presents an interesting challenge. It is not simply a matter of picking an amplifier with a
gain-bandwidth-product that is 100 times higher than the filter's corner frequency. The amplifier bandwidth calculation involves variations of
the amplifier's bandwidth and variations of the passive components over time, to mention a few. Sometimes amplifier manufacturers, such
as Texas Instruments' WEBENCH Filter Designer Web software, provide a list of the appropriate amplifiers for the related filter. This program
provides the actual amplifier suggestion along with options, if you would rather use an amplifier other than the suggested device. Filter
Designer also provides SPICE simulations of the chosen filter's sine wave response, step response, and closed-loop ac response.
The WEBENCH Filter Designer program seems to have broken through some filter design barriers. All I can say is filter design has come a
long way, baby.
References
1. WEBENCH Filter Designer program, Texas Instruments .
10 Back to top

A low-pass filter is the most common filter found in data acquisition systems. Typically this type of filter is used to reduce analog-to-digital
converter (ADC) aliasing errors and noise outside the signal bandwidth. A signal path requires this type of dedicated filter to match the
signal's requirements. If the circuit has a front-end multiplexer, it is possible to have a variety of signals that reach the ADC where each
signal source has its own set of filter requirements. Consequently, a variety of different filters and corner frequency requirements may be
required in the circuit prior to the multiplexer. These filters use independent operational amplifiers (op amps) in combination with fixed
resistors and capacitors.
An alternative filter design solution is to have one programmable filter after the multiplexer (Figure 1). The obvious advantage is reduction in
chip count, from multiple op amps to one single amplifier. Subsequently, the cost is lower when a single filter serves many analog inputs.
You can use a dual digital potentiometer, two capacitors, and a single amplifier to configure a low-pass, second-order Butterworth response
with a programmable corner frequency range of 1:100. Table 1 summarizes the digital potentiometer programmed settings.
Figure 1: A second-order, analog filter using a dual digital potentiometer, two capacitors,
and one operational amplifier reduces chip count.
With this circuit, it is possible to program second-order Bessel or Chebyshev filters with a programmable corner frequency range of 1:100.
Additionally, you can realize a combination of Butterworth, Bessel, and Chebyshev filters with the same circuit using a 1:10 corner frequency
range.
Figure 1 shows the details of a single-supply, unity gain, second-order programmable low-pass Sallen-key filter. The OPA314 is a single-
supply, rail-to-rail op amp. The filter implementation requires two resistors and two capacitors. The dual TPL0102-100, a 100 k 8-bit digital
potentiometer, replaces the two resistors in this circuit. The capacitors are hard-wired in. A reprogramming of the dual digital potentiometer
in Figure 1 changes the filter's frequency cut-off and approximation (Butterworth, Bessel, vs. Chebyshev) of this second-order, low-pass
filter.
You can calculate the appropriate resistance and capacitance with a little research, a sharp pencil, and a good eraser. An alternative to this
tedious design exercise is to determine the capacitor and resistor values using TI's WEBENCH Filter Designer software.
In the input screen, enable the low-pass filter button and type in a filter bandwidth with Ao = 1 V/V, fc = 100, fs = 1000, and Asb = -35 dB.
This will produce a second-order Butterworth filter. In this view, you can also select the supply requirements of Single Supply = +5 V. Then
press the green button, "Start Filter Design."
The next page displays a list of filter response options that meet your requirements. Select a second-order Butterworth from the list and click
"Open Design."
The Filter Designer Design Summary view (page 3) allows the user to adjust the capacitor seed value to the desired value of C1 and C2 on
the left side of the screen, under the "Filter Topology Specifications" section. Change to Capacitor seed value to 15 e-9 or 15 nF, then press
the "update" button. When this capacitor seed is set, the software changes the resistors in the circuit to appropriate values.
To produce the remaining filters in Table 1, return to page one and set the conditions for the next filter. As you do this make sure that fs is
ten times higher than fc.
Table 1 shows the digital potentiometer program setting for Butterworth filter with corner
frequencies ranging from 100 to 10 kHz (C1 = 33 nF, C2 = 15 nF).
In this design, you have learned to quickly produce an adjustable analog filter. This is one technique to design a programmable anti-aliasing
filter. Can you think of another way to do this? Send in your ideas and we can discuss them!
11 Back to top

Silicon photo sensors have been in electronic circuits since the inception of the era of silicon electronics. More than likely, scientists quickly
discovered the photo-sensing characteristics of silicon in the lab, as they worked from the daylight hours into the evening. To this day, IC
designers regularly cover their wafers under test to shield out extraneous light. Although the light sensitivity of silicon is an undesirable by-
product of the silicon, system designers have exploited this transfer of light into electrical energy in various systems. Consequently, a wide
variety of applications use silicon to sense the intensity and characteristics of light.
In these systems, a silicon sensor converts light into charge or an electrical current. These silicon sensors are the "eyes" in the electronics
world that users can employ to analyze blood, search noninvasively for tumors, detect smoke, position equipment, or perform
chromatography, to name a few applications. Basically, system designers understand how to convert light into a current, but the real
challenge is determining how to convert the low-level currents from the photo sensor into a useful electrical representation. To further
exacerbate the difficulty of the design, the required accuracy in these applications continues to increase.
The traditional design topology of the transimpedance amplifier captures this low-level signal in a hybrid approach that starts with an
amplifier and a high-value resistor in the feedback loop. The circuit design uses resistance to provide a real-time, linear representation of
the light source. This circuit places the photodiode across the amplifier's inverting input and ground of the operational amplifier. A resistor
with a value of 100 k to 10 M connects the inverting input of the amplifier to the output. You then connect the noninverting input to
ground (Figure 1). Light excitement on the photo sensor generates charge. The only path of escape for this charge is through the high-
value resistor in the amplifier's feedback loop.
The simplistic approach in Figure 1 is not without its design challenges. The operational amplifier must have relatively low-picoampere
input-bias currents and low input capacitance. An appropriate amplifier for this circuit would have a FET- or a CMOS-input stage with low-
voltage noise and microvolt-offset specifications. In the end, the designer optimizes the stability, bandwidth, low-noise performance, and
layout of this transimpedance-amplifier circuit.
The final design method is not always intuitively obvious. The photo sensor, operational amplifier, amplifier-feedback element, and these
parts' parasitics combine to create quite a rat's nest of formulas for consideration. The signal after the transimpedance amplifier requires a
multipole analog filter. In this manner, combining the input and filtering stages separates the signal of interest from the noise floor. A
sampling ADC digitizes the signal after the analog filter.
Photo-sensing circuits have changed over the years. The first approach was purely analog, using the transimpedance amplifier and
following it with a lowpass filter. From the classic transimpedance amplifier, the switched integrator has gained favor. The switched integrator
was the first step toward bringing the digital portion of the circuit closer to the signal source. The migration of the photo-sensing-application
product has moved on to totally integrated systems, such as the charged digitizing ADC.
12 Back to top

A variety of precision applications sense light and convert that information into a useful digital word. At the system's front end, a preamplifier
converts the photodiode's current-output signal to a usable voltage level. Figure 1 shows the front-end circuit of this system, which
comprises a photodiode, an operational amplifier, and a feedback network.
(Figure 1) The transfer function of this system is:
where A OL (j) is the open-loop gain of the amplifier over frequency; is the system-feedback factor, equaling 1/(1+Z F /ZIN); Z IN is the
distributed input impedance, equaling RPD||j(C PD+CCM +CDIFF ); and Z F is the distributed feedback impedance, equaling RF ||j(C RF +CF ).
A good tool for determining stability is a Bode plot. The appropriate Bode plot for this design includes the amplifier's open-loop gain and the
1/ curve. System elements determining the noise-gain frequency response are the photodiode's parasitics and the operational amplifier's
input capacitance, as well as RF , CRF , and CF in the amplifier's feedback loop.
Figure 2 shows the frequency response of the 1/ curve and the amplifier's open-loop-gain response:
f P=1/(2(R PD||RF )(CPD+CCM +CDIFF +CF +CRF )), and f Z +1/(2(R F )(CF +CRF )). The A OL (j) curve intersects the 1/ curve at an interesting
point. The closure rate between the two curves suggests the system's phase margin and, in turn, predicts the stability. For instance, the
closure rate of the two curves is 20 dB/decade. Here, the amplifier contributes an approximately -90 phase shift, and the feedback factor
contributes an approximately 0 phase shift. By adding the 1/ phase shift from the A OL (j) phase shift, the system's phase shift is -90, and
its margin is 90, resulting in a stable system. If the closure rate of these two curves is 40 dB/decade, indicating a phase shift of -180 and a
phase margin of 0, the circuit will oscillate or ring with a step-function input.
One way to correct circuit instability is to add a feedback capacitor, CF , or to change the amplifier to have a different frequency response or
different input capacitance. A conservative calculation that allows variation in amplifier bandwidth, input capacitance, and feedback-resistor
value places the system's pole of 1/ at half the frequency where the two curves intersect:
where f GBW is the gain-bandwidth product of the amplifier. In this design, the system's phase margin is 65, and the step function's
overshoot is 5%.
References
1. Baker, Bonnie, "Photo-sensing circuits: The eyes of the electronic world are watching," EDN, Aug 7, 2008.
13 Back to top

How much noise is too much noise in a photodiode-preamplifier circuit? You can derive the noise performance of a transimpedance
amplifier (Figure 1a) with calculations or by using a Spice simulation. When calculating the noise performance of the circuit, consider six
regions in the frequency spectrum (Figure 1b) and add each region with a root-sum-square equation or the following equation:
The first five regions are equal to the multiple of the areas under the closed-loop-gain and amplifier-noise-density curves. The area under
the noise-density curve in the e1 , flicker-noise (1/f), region is V 1/f:fB-fA =A N , where A N is the amplifier's input-noise-density at
1 Hz and f B is the corner frequency where the flicker noise tapers off. For many CMOS or FET amplifiers, the flicker-noise region usually
ranges from dc to 100 or 1000 Hz. A calculation proves that the contribution to noise in this low-frequency region is relatively low:
where RF is the feedback resistor and RPD is the device's parallel resistance.
In the e2 region, multiply the broadband noise of the amplifier, the closed-loop dc-noise gain (1+R F /RPD), and the square root of the region's
bandwidth. Again, the contributed noise in this region is usually relatively low because of its location in the lower frequency range.
Calculate the noise contribution and the e3 region in the same manner with f P=1/[2(RPD||RF ) (CPD+CCM +CDIFF +CF +CRF )] and
f Z =1/[2(RF ) (CF +CRF )].
where CPD is the device's capacitance and CDIFF is the differential amplifier's capacitance.
The noise in regions e4 and e5 uses the higher-frequency gain of the closed-loop-gain curve with the value of C1 being the parallel
combination of the input capacitors, or [CP-R1||2C CM ||CDIFF ], and C2 is the parallel combination of CF and CRF .
The sixth part of the noise equation, e6 , represents the noise contribution of the feedback resistor. The amplifier does not gain the
contribution of noise from the feedback resistor:
where K is Boltzmann's constant, which is 1.3810-23 ; T is temperature in Kelvin; RF is the feedback resistor in ohms; and BW is the
bandwidth of interest.
When asking how much noise is too much noise in this photodiode-preamp circuit, consider that a 12-bit system operating with a 5V input
range has an LSB of 1.22 mV. The LSB for a 16-bit system with the same input-voltage range is 76.29 V. Both LSBs are peak-to-peak
numbers, and the values in this column are root-mean-square values.
References
2. "Noise Analysis of FET Transimpedance Amplifiers," SBOA060, Texas Instruments, February 1994.
3. Baker, Bonnie, "RMS and peak-to-peak noise trade-off," EDN, May 15, 2008.
14 Back to top

Have you ever been in need of a second opinion when your sanity is at stake? The pulse oximeter may be able to provide that second
opinion if your brain is oxygen-deprived. This condition could affect you if you are a pilot, hiking in the high altitude of a mountain range, or
even undergoing surgery. The pulse oximeter is a noninvasive instrument that monitors SpO 2 (saturation of hemoglobin with oxygen) in
your blood.
You measure the oxygen in the blood by alternating the on-times of a red LED with a 650-mm wavelength and an NIR (near-infrared) LED
with a 940-mm wavelength, taking the ratio between the intensities from a photodiode, and comparing that ratio with an SpO 2 look-up table
in the microcontroller.
The transimpedance amplifier appears in medical and laboratory instrumentation, position and proximity sensors, photographic analyzers,
bar-code scanners, and even smoke detectors. In the medical field, you will primarily find transimpedance amplifiers in the CT (computed-
tomography)-scanner front end and the pulse oximeter. Figure 1 shows a simplified block diagram of a pulse oximeter.
In the circuit in Figure 1, the red LED is on for 50 sec, both LEDs are off for 450 sec, the NIR LED is on for 50 sec, and then both LEDs
are off for 450 sec. The system repeats this cycle continuously. The transimpedance amplifier, A 1 , converts the photodiode current
generated by the LEDs to a voltage at the output. The signal then travels through a bandpass filter and gain stage to the 12-bit ADC. The
signal also travels through a lowpass filter to regulate the driver power to the LEDs. The microcontroller acquires the signals from the 12-bit
ADC, computes the ratio of the red- and NIR-LED signals, and compares the results with a look-up table. The LCD shows a percentage of
oxygenated hemoglobin versus nonoxygenated hemoglobin and your heart rate.
When you choose your device for the pulse-oximeter transimpedance-amplifier circuit, you need to make sure that the amplifier's input-bias
current is very low or in a picoamp region at 25C. The amplifier's input-bias current creates an output-voltage error by conducting through
the high-impedance resistor, RF , in the amplifier's feedback loop. FET- or CMOS-amplifier input devices usually meet this requirement. A
second consideration is that the low-frequency voltage noise of your amplifier must be very low. When you consider the input-voltage noise
of the amplifier, scrutinize the impact of the flicker noise. After the transimpedance amplifier, a bandpass filter eliminates the noise above 5
Hz. Finally, the amplifier's initial offset error and overtemperature should be in the microvolt region if you want to minimize linearity errors. It
may be worthwhile to use an autozero amplifier.
A normal output for the pulse oximeter is approximately 97%2%, ranging from 95 to 100%. The alarms on the pulse oximeter usually
sound when the SpO 2 level drops below 90%. If there is a shortage of oxygen in your system, you may experience poor judgment or loss of
motor function. If a pulse oximeter indicates that your oxygen levels are stable, you may want to explore other diagnostic avenues, or
perhaps you just dance to the beat of a different drummer. Good luck!
References
1. Medical Instruments Applications Guide, Texas Instruments, 2010.
2. Townsend, Neil, MD, "Pulse Oximetry," Medical Electronics, Michaelmas 2001.
15 Back to top

Photodiodes transform a basic physical occurrence, light, into an electrical form, current. Design engineers methodically convert the
photodetector's current to a usable voltage, which makes the manipulation of the photodiode signal manageable. There are many ways to
approach the photosensing-circuit problem, but one issue came to mind as I read the comments in the Talkback section for a recent
column. A reader requested a circuit that reduces the noise impact of a photodiode with a large parasitic capacitance.
A classic photo-sensing system circuit has a photodiode, an operational amplifier, and a feedback resistor/capacitor pair at the front end.
Recalling a circuit from another column, in this circuit, the photodiode, amplifier, and feedback-capacitance elements limit the bandwidth of
the circuit. Yet another column details the stability of this circuit.
When sensing with a photodiode with a large parasitic capacitance or from a remote site, the input of the amplifier has a large capacitance
across its input. The result of this added capacitance increases the circuit's noise gain unless you increase the amplifier's feedback
capacitor. If the feedback capacitor, CF , increases, the bandwidth of the circuit decreases.
To fix this problem, you can use a bootstrap circuit (Figure 1). Photodiodes with a relatively low diode capacitance do not benefit from this
circuit. A unity-gain buffer, A 2 , removes the cable's capacitance and the photodiode's parasitic capacitance from the input of the
transimpedance amplifier, A 1 .
When designing this circuit, you'll find that the type of amplifier you select for A 2 is somewhat easy. The only important performance
specifications are low input capacitance, low noise, a wider bandwidth than A 1 , and low output impedance.
In this design, A 2 's input capacitance is the only capacitance that plays a role in the ac-transfer function of the transimpedance system. The
input capacitance of the buffer replaces the summation of the input capacitance of A 1 , the cable capacitance, and the parasitic capacitance
of the photodiode. A good rule of thumb is to have CA2 <<(CA1 +CCA+CPD), where CA1 and CA2 represent the sum of their input differential
and common-mode capacitance.
With this design, you exchange one noise problem, A 1 , with another, A 2 . The unity-gain buffer removes the noise effect from A 1 . A good
approach is to make the noise of A 2 less than or equal to that of A 1 .
The difference between the input signal and the output signal in this system falls across the cable/diode capacitance. You can keep this
difference low selecting A 2 with wider bandwidth than A 1 and keeping A 2 's output impedance low. A 2 's gain roll-off introduces an upper
limit for the bandwidth improvement, making A 2 's bandwidth much greater than that of A 1 . This circuit requires stability optimization as you
balance CF and the input capacitance of A 2 .
References
1. Baker, Bonnie, "A good holiday-season project," EDN, Dec 15, 2010.
3. Baker, Bonnie, "Transimpedance-amplifier stability is key in light-sensing applications," EDN, Sept 4, 2008.
4. Graeme, Jerald G, Photodiode Amplifiers: Op Amp Solutions, McGraw-Hill, 1996, ISBN 0-07-024247-X.
5. Kurz, Dov, and Avner Cohen, "Bootstrapping reduces amplifier input capacitance," EDN, March 20, 1978.
16 Back to top

Texas Instruments - TI.com
Follow Us |
Table of Contents
Introduction
Author's biography
Introduction
I have worked with people who are surprised when I tell them that the operational amplifier is not dying, but rather they are
permanent fixtures in the electronics world. This is particularly true as long as the real-world remains in the analog domain.
You may say that it will become obsolete soon, but I have a different opinion. Amplifiers, including instrumentation amplifiers (INA) and
programmable gain amplifiers (PGA), are here to stay as standalone devices and as integrated functions inside complex ICs. And, to take it
a step further, there is still a lot to learn about the amplifier and their appropriate placements in your application circuits.
The following excerpts from the EDN Baker's Best column represent the best of the amplifier short articles. They have been complied for
1 Back to top

References
2 Back to top
References
3 Back to top

References
1.
2.
4 Back to top

5 Back to top

1).
output at V OUT .
References
6 Back to top

(GBWP).
(f GBWP ).
the math.
References
7 Back to top


2 Back to top
References
3 Back to top

References
1.
2.
4 Back to top


2 Back to top
References
3 Back to top

References
1.
2.
4 Back to top


Follow Us |
Table of Contents
Introduction
Author's biography
Introduction
1 Back to top

References
2 Back to top
References
3 Back to top

References
1.
2.
4 Back to top

5 Back to top

1).
output at V OUT .
References
6 Back to top

(GBWP).
(f GBWP ).
the math.
References
7 Back to top


5 Back to top

1).
output at V OUT .
References
6 Back to top

(GBWP).
(f GBWP ).
the math.

Follow Us |
Table of Contents
Introduction
Author's biography
Introduction
1 Back to top

References
2 Back to top
References
3 Back to top

References
1.
2.
4 Back to top

5 Back to top

1).
output at V OUT .
References
6 Back to top

(GBWP).
(f GBWP ).
the math.
References
7 Back to top


References
7 Back to top

cutoff frequency, all three of the filter's responses show a slope of 40 dB/decade. You would expect this response from a second-order
8 Back to top

Before selecting the amplifier, however, you need to determine the filter-cutoff frequency, f CUT (or 3-dB frequency). You can use filter-
data sheet.
References
1.
9 Back to top


8 Back to top

data sheet.
References
1.
9 Back to top


the MFB (Figure 2).
V CM =V +/2.
long way, baby.
References
10 Back to top

range.
filter.
In the input screen, enable the low-pass filter button and type in a filter bandwidth with Ao = 1 V/V, fc = 100, fs = 1000, and Asb = 35 dB.
"Open Design."
11 Back to top

12 Back to top

closure rate of the two curves is 20 dB/decade. Here, the amplifier contributes an approximately 90 phase shift, and the feedback factor
contributes an approximately 0 phase shift. By adding the 1/ phase shift from the A OL (j) phase shift, the system's phase shift is 90,
and its margin is 90, resulting in a stable system. If the closure rate of these two curves is 40 dB/decade, indicating a phase shift of 180
and a phase margin of 0, the circuit will oscillate or ring with a step-function input.
overshoot is 5%.
References
13 Back to top

the noise-density curve in the e1 , flicker-noise (1/f), region is V 1/f:fBfA =A N , where A N is the amplifier's input-noise-density at
f Z =1/[2(RF ) (CF +CRF )].
combination of the input capacitors, or [CPR1 ||2C CM ||CDIFF ], and C2 is the parallel combination of CF and CRF .
where K is Boltzmann's constant, which is 1.381023 ; T is temperature in Kelvin; RF is the feedback resistor in ohms; and BW is the
References
14 Back to top

8 Back to top

data sheet.
References
1.
9 Back to top


the MFB (Figure 2).
V CM =V +/2.
long way, baby.
References
10 Back to top

range.
filter.
"Open Design."
11 Back to top

12 Back to top

overshoot is 5%.
References
13 Back to top

f Z =1/[2(RF ) (CF +CRF )].
References
14 Back to top

8 Back to top

data sheet.
References
1.
9 Back to top


the MFB (Figure 2).
V CM =V +/2.
long way, baby.
References
10 Back to top

range.
filter.
"Open Design."
11 Back to top

12 Back to top

overshoot is 5%.
References
13 Back to top

f Z =1/[2(RF ) (CF +CRF )].
References
14 Back to top

11 Back to top

12 Back to top

overshoot is 5%.
References

8 Back to top

data sheet.
References
1.
9 Back to top


the MFB (Figure 2).
V CM =V +/2.
long way, baby.
References
10 Back to top

range.
filter.
"Open Design."
11 Back to top

12 Back to top

overshoot is 5%.
References
13 Back to top

f Z =1/[2(RF ) (CF +CRF )].
References
14 Back to top

13 Back to top

f Z =1/[2(RF ) (CF +CRF )].
References
14 Back to top

your blood.
References

Follow Us |
Table of Contents
Introduction
Author's biography
Introduction
1 Back to top

References
2 Back to top
References
3 Back to top

References
1.
2.
4 Back to top

5 Back to top

1).
output at V OUT .
References
6 Back to top

(GBWP).
(f GBWP ).
the math.
References
7 Back to top

8 Back to top

data sheet.
References
1.
9 Back to top


the MFB (Figure 2).
V CM =V +/2.
long way, baby.
References
10 Back to top

range.
filter.
"Open Design."
11 Back to top

12 Back to top

overshoot is 5%.
References
13 Back to top

f Z =1/[2(RF ) (CF +CRF )].
References
14 Back to top

your blood.
References
15 Back to top

References
16 Back to top

Single-supply amplifiers do not truly swing rail to rail at the output. Near the rail, the amplifier is nonlinear. For linear operation, the output of
single-supply amplifiers can come within only 50 to 300 mV of each rail (Figure 1).
http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 1:13:55 PM]

your blood.
References
15 Back to top

References
16 Back to top

Single-supply-amplifier, rail-to-rail-output ads can give a false sense of security. Figure 1 shows a typical single-supply amplifier's output
swing as you drive the output to the rails.
The amplifier's linearity starts to degrade long before reaching the output-swing maximums, and the amplifier output never reaches either
rail.
The conditions of the dc-open-loop-gain specification define the amplifier's linear operating output range. The dc open loop gain in decibels
is 20 log(V OUT /V OS ), where V OUT is the output voltage and V OS is the input offset voltage. When you drive the output high, V H is the
maximum voltage level at the output in the dc-open-loop-gain measurement. V OH is the absolute maximum voltage level with respect to
V DD (drain-to-drain voltage) that the output can reach. V L is the minimum voltage level at the output in the dc-open-loop-gain measurement,
and V OL is the absolute minimum voltage level that the output can reach. V H is less than V OH, and V L is greater than V OL .
From a signal-chain perspective, you can see an op amp's output limitations to swinging rail to rail when the op amp is driving an ADC. The
FFT plot in Figure 2a shows the amplifier/ADC-combination response to a 1-kHz signal in a 5V system. The amplifier's typical closed-loop
bandwidth is about 3 MHz with a typical slew rate of 2.3V/sec. The amplifier output voltage swings from 140 mV to 4.66V. In this 5V-supply
system, the headroom between the signal and rails is 140 mV. For this amplifier, the V OL minimum specification is 15 mV above ground.
The V OH maximum specification is V DD 20 mV.
Figure 2a illustrates the nonlinearity-output-stage effects with a single-supply CMOS amplifier by showing distortion at 2, 3, and 4 kHz and
so on. By reducing the amplifier's output signal to 272 mV from each rail, the data looks perfect with only the ADC distortion (Figure 2b).
When using a single-supply amplifier, read the fine print! Some single-supply amps have output-stage charge pumps, allowing the
amplifier's output swing to go to and well beyond the power-supply rails. In every case, read your data sheet and refer to the conditions on
the open-loop-gain test.
17 Back to top

Fundamental analog devices that serve applications such as high-resolution delta-sigma or SAR (successive-approximation-register)
converter systems are feeling the crunch from amplifiers that have difficulty with achieving good rail-to-rail input performance. The simple
rail-to-rail operational amplifier must have a transistor design that spans the power supply with minimal distortion.
The trend toward designing single-supply op amps started in the 1970s with a single differential-input stage that spanned a portion of the
common-mode input range. Later, designers added a second, or complementary, differential-input stage. The two stages shared, with some
distortion, the rail-to-rail input operation across the complete amplifier's rail-to-rail common-mode range (Reference 1). Neither of these
approaches produced an amplifier adequate for the high-precision systems to span the amplifier's full common-mode input range.
Eventually, IC designers borrowed a technology from other devices to solve this problem. They began to use the all-too-common charge
pump to push a single differential-input stage of the amplifier above the positive-power supply (Figure 1). Amplifier designers place the
switching mechanism's frequency above the amplifier's bandwidth and keep the switching noise lower than the amplifier's thermal noise
floor.
The single differential-input stage with a charge pump buys you a 20- to 30-dB increase in the amplifier's CMMR (common-mode-rejection
ratio). This increase has a positive effect on amplifiers in buffer configurations. You can also expect almost a tenfold decrease in the
amplifier's THD (total-harmonic-distortion) performance. So, if you use an amplifier that has a charge pump in its input stage to drive high-
precision SAR or delta-sigma converters, your system's performance will improve.
For example, the THD of an ADC driven by an op amp in a buffer configuration is the root-sum square of distortion contributions of the ADC
and op amp. In this configuration, the system THD is:
where THDOPA =20log(THD OPA% 100) and THDOPA% is the THD specification in the operational amplifier's data sheet in units of
percentage.
Using these equations, if an operational amplifier with a complementary input stage has a THD specification of 0.004%, with an input
voltage of 4V p-p, and the 16-bit SAR ADC has a THD specification of 99 dB, the system THD is 88 dB. Alternatively, if the op amp's
input stage has a charge pump with a THD specification that is 0.0004%, the system THD becomes 98 dB.
Single-supply amplifiers continue to keep pace with high-resolution converters because engineers implement innovative amplifier-circuit
topologies, such as an input stage with a charge pump. The charge pump is a good stopgap; however, engineers continue to demand lower
system power supplies and insist on better signal integrity.
References
1. Baker, Bonnie, "Where did all the racket come from?" EDN, April 23, 2009.
2. "OPA365, OPA2365 2.2V, 50MHz, Low-Noise, Single-Supply Rail-to-Rail Operational Amplifiers," Texas Instruments, June 2006.
3. "OPA333, OPA2333 1.8V, microPower CMOS Operational Amplifiers, Zero-Drift Series," Texas Instruments, March 2006.
18 Back to top

A hot discussion topic with single-supply operational amplifiers is whether they are capable of rail-to-rail input or output operation. Vendors
of single-supply op amps claim their amplifiers have rail-to-rail input capability, but the chip designer has to make some compromises to
achieve this type of performance.
A common single-supply amplifier input topology has parallel PMOS and NMOS differential input stages, combining the advantages of those
stages to achieve actual rail-to-rail input operation (Figure 1). When you bring V IN+ toward the negative rail, the PMOS transistors are
completely on, and the NMOS transistors are completely off. When you bring the input terminals to the positive rail, the NMOS transistors
are in use, and the PMOS transistors are off.
Figure 1: This composite input stage of the op amp uses PMOS and NMOS differential
pairs so the input-voltage range can extend from above the positive rail to below the
negative rail.
Although the precision, low-power OPA344 input stage in Figure 1 has rail-to-rail input operation, there are performance compromises that
the circuit designer must address. The design topology in Figure 1 can have wide variations in offset voltage across the amplifier's
common-mode input range. In the region near ground, the PMOS offset-error portion of the input stage is dominant. In the region near the
positive power supply, the NMOS offset error dominates.
Figure 2: As the amplifier's common-mode voltage changes from ground to the positive
supply, the input stage of the CMOS amplifier completely changes from its PMOS input
pair to its NMOS input pair at -2V below the 3V positive supply rail.
The best way to view the input stage's behavior is to look at the offset voltage versus the common-mode input voltage (Figure 2). The 4.6-
MHz, rail-to-rail input/output LMP7701 CMOS amplifier in Figure 2 exhibits the offset-voltage-error crossover behavior around 1.4V. At
lower common-mode input voltages, the PMOS transistors are in operation, with the NMOS transistors turned off. At approximately 1.1V,
the NMOS transistors start to turn on. As the common-mode input voltage increases, the NMOS section of the circuit finally takes over, with
the PMOS transistors completely off. From approximately 1.1 to 2V, both the PMOS and NMOS transistors are operating.
There are circuit-design tricks at your disposal for minimizing this input-stage crossover effect; you can read about them in "Rail-to-rail input
amplifier application solutions."
Single-supply amplifier manufacturers also claim they have devices that will swing rail-to-rail on the output. With those types of amplifiers,
the output cannot go all the way to the rails, but it can get close.
Next time, I'll talk about the rail-to-rail amp's output stage and its ability to achieve rail-to-rail performance.
References
1. OPA344 data sheet, Texas Instruments.
2. LMP7701 data sheet, Texas Instruments.
19 Back to top

Many single-supply operational amplifiers are capable of rail-to-rail input operation, if they have the proper input circuitry to support this
function. This article is an expansion of the blog post "What does 'rail to rail' input operation really mean?" By examining an additional input-
stage topology and implementing these op amps into some common op-amp circuits, you will see what I mean.
The article discusses a composite input-stage topology for single-supply op amps. You may recall that this topology exhibited an offset-
voltage, crossover distortion as the input common-mode voltage traveled across the full rail-to-rail input voltage range. Switching from one
differential input stage to the other causes this distortion.
Revisiting single-supply input topologies

The CMRR (common-mode rejection ratio) specification describes changes in the amplifier's offset-voltage versus common-mode input
changes. The specification conditions can subtly describe the amplifier's rail-to-rail input topology. A typical common-mode rejection
specification for an amplifier with a composite input stage has two or more CMRR specifications (Table 1).
Table 1: The single-supply op-amp data sheet describes the CMRR test conditions as well
as the actual specified values. These specification tables provide evidence of the
characteristics of the input-stage topology, which in this case is an amplifier with a
composite input stage.
In Table 1, the CMRR is equal to 20*log (V CM /V OS ). Line 1 verifies that the amplifier has true rail-to-rail input capability. Line 2 specifies
the CMRR from ground to 1.8V below the positive power-supply rail as 92 dB (typ). Note that this specification does not describe the range
specified in Line 1. Line 3 specifies the CMRR across the entire input range as 70 dB (typ, V S=5V). This disparity in specification only points
out that the amplifier's input stage has a composite input topology.
There are more ways to tackle this input topology problem with the IC design. The composite input stage is one example. As a second IC-
design strategy, a unique zero-crossover input topology provides superior common-mode performance over the entire input range.
In Figure 1, a regulated charge pump lifts the top of the differential input as well as its biasing current source to 1.8V above the power-
supply voltage, V S. This increased headroom allows for a single differential PMOS input stage to replace the composite differential input
stage. A typical common-mode rejection specification for an amplifier with a charge-pump input stage has one CMRR specification. This
specification encompasses the full amplifier input range.
Figure 1: This single PMOS differential input stage in conjunction with a high-side charge-
pump eliminates the common-mode crossover distortion found in composite input stages.
Figure 2 compares the response of the charge pump to composite amplifier inputs. The difference between these two topologies in this
figure is obvious.
Figure 2: The op-amp offset voltage of an op amp with a charge-pump input remains
constant across changes in the common-mode voltage. The offset voltage of an op amp
with a composite input varies across changes in the common-mode voltage.
The op-amp topology in Figure 1 provides superior common-mode performance over the entire input range. In fact, the input range extends
at least 100 mV beyond both power-supply rails. You may think that the charge pump's output ripple voltage will become a noise problem.
However, the charge-pump design can provide an output ripple voltage that is low enough so as to not produce undesirable noise or
distortion at the output of the op amp.
Application solutions
Circuit designers can use rail-to-rail input op amps in virtually any op-amp configuration. To achieve optimum performance, however, circuit
designers need to consider the behavior of the single-supply op-amp input stage. Let's consider the behavior of the composite and charge-
pump amplifier inputs in a single-supply inverting amplifier, noninverting amplifier, and buffer circuits.
In many applications, the amplifier's common-mode input voltage can remain at a static voltage. This is true for the inverting-amplifier circuit
found in Figure 3. The inverting-amplifier circuit gains the input signal as well as changes the signal polarity.
Figure 3: The inverting-amplifier circuit keeps the amplifier's common-mode voltage at a

constant voltage, V B.
The circuit requires a bias voltage, V B, to keep the output range at V O between the power-supply rails. When V B establishes the common-
mode voltage of the amplifier, the transfer function for this circuit is expressed as
The V B voltage can be anywhere between ground and V S, as long as the combination of the elements in this circuit (VB, R1 , R2 , and V IN)
keeps the output (VO) between V S and ground. Since the V B voltage is static, the amplifier's common-mode voltage remains constant.
When using composite input amplifiers, choose the V B voltage to be below or above the PMOS/NMOS transition region.
Bottom line: inverting-amplifier circuit

Composite input: performs without distortion with V B outside the PMOS/NMOS transition region
Charge-pump input: performs without distortion
The noninverting amplifiers common-mode voltage (Figure 4) is equal to the input voltage, V IN. The noninverting amplifier circuit gains the
input signal with a voltage level-shift implemented with V B.
Figure 4 The noninverting gained amplifier configuration allows the amplifiers common-
mode voltage to change with input signals.
For proper operation, the noninverting amplifier circuit may require a bias voltage, V B, to keep the output range at V O between the power-
supply rails. Here V B establishes the common-mode voltage of the amplifier. The transfer function for the circuit is expressed as
The input common-mode voltage (VIN) can vary between ground and V S, as long the voltage at V O remains between the power-supply rails.
VB can be assigned so that the input signal is not required to go across the composite input stages crossover region. If the input signal
always remains less or more than the composite input-stages transition voltage, the circuit will not create distortion from the composites
input stage, crossover phenomena.
Bottom line: noninverting-amplifier circuit
Composite input: the input signal (VIN) must not travel into the transition region of the PMOS and CMOS differential input stages
Charge-pump input: performs without distortion in reaction to common-mode voltage or V IN changes
With a unity-gain buffer (Figure 5), the input signal (VIN) can traverse from one rail to the other rail. In this circuit, V IN establishes the
amplifiers common-mode voltage. If the circuit designer uses the amplifier in a voltage follower or buffer configuration, the composite
amplifier exhibits some limitations in linearity due to the input-stage topology. If the circuit designer wants a distortion-free output from this
amplifier circuit, an op amp with a composite input topology may limit the input range of the buffer.
Figure 5 The input signal, V IN, of the buffer circuit changes the amplifiers common-mode
voltage. Composite input op amps can create distortion at the output, V O, as V IN changes.
The transfer function of the circuit in Figure 5 is shown here:
The input common-mode voltage (VIN) can vary between power-supply rails. If V IN travels across this entire range, the composite input
amplifier produces distortion at the circuit output, V O. If the input signal always remains less than the composite input stages transition
voltage, the circuit does not create distortion from the composites input stage, crossover phenomena. Alternatively, an amplifier with a
charge-pump input stage does not create this unwanted distortion across the entire input range.
Bottom line: buffer circuit
Composite input: the op amps input range (VIN) must not travel into the transition region between the PMOS and CMOS differential

your blood.
References
15 Back to top

References
16 Back to top

rail.
17 Back to top

floor.
percentage.
References
18 Back to top

negative rail.
References
19 Back to top


figure is obvious.



your blood.
References
15 Back to top

References
16 Back to top

rail.
17 Back to top

floor.
percentage.
References
18 Back to top

negative rail.
References
19 Back to top


figure is obvious.



18 Back to top

negative rail.
References
19 Back to top


rail.
17 Back to top

floor.
percentage.
References
18 Back to top

negative rail.
References
19 Back to top


figure is obvious.


input stages
Charge-pump input: performs without distortion in reaction to common-mode voltage changes
Conclusion
Manufacturers use several input topologies in their designs of rail-to-rail input amplifiers. It is true that circuit designers can use rail-to-rail
input amplifiers in virtually any op-amp configuration. However, if you choose to use a rail-to-rail input amplifier, it pays to understand the
CMRR impact on your circuit. The composite input amplifier produces a crossover distortion, requiring special considerations in some
classes of circuits. The charge-pump input amplifier does not produce this same distortion.
References
1. Edgar Snchez-Sinencio, "Rail-to-Rail Op Amps," Ax-09, TAMU, AMSC.
2. 50 MHz, Low-Distortion, High CMRR, RRI/O, Single-Supply Operational Amplifier (Rev. D).
20 Back to top

Current-feedback amplifiers have a higher slew rate than do voltage-feedback amplifiers. As such, current-feedback amps can better solve
high-speed problems than their voltage-feedback counterparts. The name "current-feedback amp" carries some mystique, but, generally,
the application-circuit configurations for voltage- and current-feedback amps are the same, except for a few key points.
First, the feedback resistor of a current-feedback-amp circuit must stay within a small range of values. Lower value resistors reduce the
current-feedback amp's stability. The feedback resistor's higher values reduce the current-feedback amp's bandwidth. You can find the
prescribed feedback-resistor value in the current-feedback amp's product data sheet. The voltage-feedback-amp's feedback-resistance
value is more forgiving. This amplifier's drive capability limits the resistor's minimum value, and the overall circuit noise limits the maximum
value.
Figure 1 shows a circuit that is appropriate for either a current- or a voltage-feedback amp. If the feedback resistance, RF , equals 2R IN,
where RIN is the input resistance, the closed-loop gain of each channel is 2V/V. At first glance, it is easy to assume that the closed-loop
bandwidth equals the gain-bandwidth product divided by each channel's gain, or |2V/V|. Don't make this assumption!
If you use a voltage- or current-feedback amp with the circuit in Figure 1, the noise gain is:
(1)
where N is the number of input channels. This circuit's bandwidth, with a voltage-feedback amp, equals the gain-bandwidth product divided
by the noise gain. For instance, if you have a voltage-feedback amp with a gain-bandwidth product of 180 MHz and there are three input
channels (N=3) at a gain of 2V/V, the circuit's closed-loop bandwidth is 25.7 MHz. Additional channels reduce the closed-loop bandwidth,
even though the input signals continue to see a gain of 2V/V.
If you use a current-feedback amp with the circuit in Figure 1, the amplifier's closed-loop bandwidth depends less on the closed-loop gain
and the number of input channels. If you design this circuit with such an amp, you would first pick the optimum feedback resistor, per the
manufacturer's specification and the circuit's noise gain. You would then select the appropriate value for RIN. From this point, if you add
channels to the circuit, a small variation in the signal bandwidth and gain peaking in circuit may occur. If that scenario is a concern, go back
and refine your feedback-resistor selection. For both current- and voltage-feedback amps, the noise gain always equals the result of
Equation 1, but you can reduce the feedback-resistor value with the current-feedback-amp circuit and get an increase in circuit bandwidth.
21 Back to top

The three-op-amp instrumentation amplifier in Figure 1 is seemingly a simple configuration in that it uses a basic, decades-old operational
amplifier to gain a differential input signal. The op amp's input offset-voltage error is easy to understand. The definition of an op amp's open-
loop gain has not changed. The simple idea of an op amp's CMR (common-mode rejection) has been around since the beginning of op-amp
time. So what is the hang-up?
Equation 1 yields the common CMR for a single op amp and instrumentation amplifier:
Equation 1
where G is the system gain, V CM is the changing common-mode voltage that you apply equally to the system's input terminals with
respect to ground, and V OUT is the change in the system's output voltage with respect to the changing V CM values.
With CMR, the inner workings of the op amp are straightforward; the change of offset voltage is the only concern. Two factors influence an
instrumentation amplifier's CMR. The first and most dominant factor is the balance of the resistor ratios across A 3 . For instance, if R1 equals
R3 and R2 equals R4 , the CMR of the three-op-amp instrumentation amplifier is ideally infinite.
At a real-world level, however, the relationship of R1 , R2 , R3 , and R4 to the instrumentation amplifier's CMRspecifically, matching the R 1 -
to-R2 ratio to the R3 -to-R4 ratiois critical. These four resistors combine with A 3 to subtract and gain the signals from the outputs of A 1 and
A 2 . A mismatch be-tween the resistor ratios creates an error at the output of A 3 . Equation 2 gives the contribution to the CMR error with
respect to the relationship of these resistors:
Equation 2
For instance, if R1 , R2 , R3 , and R4 are approximately the same value and the ratio of R3 to R4 is 1.001 of R1 /R2 , this 0.1% mismatch will
cause a degradation of the instrumentation amplifier's CMR from ideal to a 66-dB level. At a gain of one, CMR A3 is equivalent to the CMR
of the entire instrumentation amplifier.
As Equation 1 states, the instrumentation amplifier's CMR increases as the system's gain increasesa nice feature. Equation 1 might
motivate an instrumentation-amplifier designer to ensure that there is plenty of gain available, but A 1 and A 2 's open-loop gain error places a
limit on this strategy. An amplifier's open-loop gain is 20log(V OUT /V OS ), where V OS is the offset voltage. As the gain of A 1 and A 2
increases, the offset errors from the amplifier's open-loop gain also increase. The changes in output swing of A 1 and A 2 typically span the
supply rails. At higher instrumentation-amplifier gains, the open-loop gain error of the op amps dominates. These errors degrade the CMR
of the instrumentation amplifier at higher gains. Consequently, the instrumentation amplifier's CMR performance values tend to reach a
maximum value at higher gains.
So, from the CMR perspective, instrumentation amplifiers are systems in which various parts contribute to the CMR error at different system
gains. This situation is not so mysterious when you think about the inside of this device. As you separate the parts, the picture becomes
clear.
References
1. Baker, Bonnie, "CMRR not created equal for all INAs," Electronics World, December 2008.
2. Kitchin, Charles, and Lew Counts, "The right way to use instrumentation amplifiers," EDN, Sept 15, 2005, pg 69.
3. Amplifiers and Linear, Texas Instruments.
22 Back to top

My last column provided a glimpse inside the CMR (common-mode rejection) of a three-op-amp INA (instrumentation amplifier) and
revealed the main contributors to total CMR error (see "Understanding CMR and instrumentation amplifiers," EDN, Nov 26, 2009, pg 14).
This story goes deeper, however, if you look at the common-mode range of the same device. Of all the performance characteristics of an
INA, the most misunderstood is the common-mode-range requirement. So how do designers calculate the INA's common-mode range?
Consider the exposure to an input/gain-overload condition on the INA as a possibility.
Three basic kinds of nodes in the INA can cause input/gain-overload problems (Figure 1). Pay attention to the voltage levels of the INA's
V IN+ and V IN input pins, the V OA1 and V OA2 output levels of A 1 and A 2 , and the V OUT output-swing capability of A 3 . As you work with
these concepts, you may notice that an input signal into an INA can produce an incorrect output signal that is nevertheless within the
device's normal output range.
The applied input voltages to the INA are equivalent to the common-mode input voltage plus or minus the differential input signal. The input
stages of A 1 and A 2 limit the range of these two input voltages. The maximum and minimum input-voltage limits vary from device to device.
The input voltages at V IN+ and V IN and the gain of A 1 and A 2 cause the internal output voltages to increase or decrease. Note that the A 1
and A 2 stages do not gain the input common-mode voltage, V CM . An input/gain-overload condition can occur if A 1 's output voltage, A 2 's
output voltage, or both violate the internal output-swing restrictions. This condition is impossible to directly measure. You must be aware of
the limitations of V OA1 and V OA2 and then calculate whether your design is at risk using the equations in Figure 1.
An example of this type of input/gain-overload condition occurs when the in-range voltages on the INA's inputs drive A 1 , A 2 , or both to their
positive or negative output-swing limit. In this condition, A 3 measures an erroneous difference voltage. This erroneous voltage plus the
voltage reference to the INA is incorrect and may be inside the output range of A 3 .
The final place to look for an input/gain-overload condition is at the output, V OUT , of the INA. The A 3 output restrictions are similar to those
of any other amplifier. For instance, in a single-supply environment, the output swing never spans all the way to the rails.
The common-mode behavior of the three-op-amp INA may surprise you if you ignore the nuances of the internal A 1 and A 2 output stages.
All of the internal amplifier's output voltages relate to the linear common-mode input range of the complete INA. Most product data sheets
provide illustrations of the effects of input/gain-overload conditions. You can now use those graphs to your advantage.
23 Back to top

Many industrial and medical applications use instrumentation amplifiers to condition small signals in the presence of large common-mode
voltages. INAs have complete closed-loop operational amplifiers with feedback components included. Under normal conditions, an INA is
easy to use, as long as you pay attention to the input stage and the output of the first stage. The quick and inexpensive solutions described
here can help you avoid the INA circuit's input-stage pitfalls.
In a classic, three-op-amp INA (Figure 1), the input stage has two op amps in an adjustable-gain configuration and provides high input
impedance on both the inverting (VIN) and noninverting (VIN+) inputs. The output stage has four matched resistors around a single op amp.
When the circuit designer exercises proper precautions with the input pins, this configuration rejects external common-mode voltage and
noise. The output stage has a reference voltage that level-shifts the output with respect to ground. The level shift is convenient in single-
supply applications. As with many INAs, you can program the gain found in Figure 1 with a single resistor, RG.
Depending on the INA's silicon process, V IN+ and V IN connect to a bipolar transistor base, FET gate, or CMOS gate. All inputs to the INA
require a current-return path to ground and a bias-voltage reference. Without the current-return path and the bias-voltage reference, the
INA input stage saturates or floats to an undesirable voltage. Either condition creates an invalid output voltage.
The floating thermocouple circuit in Figure 2a does not provide a current path to ground or a bias-voltage reference for the INA's input pins.
Thus, the input-current leakages are not dissipated, and the two inputs can float to any undefined voltage. That situation, in turn, causes the
INA's output to change to an invalid output voltage, usually in the middle of the INA's output range. The invalid voltage can appear to be
legitimate, making it difficult to detect the correctness of the INA input implementation. Figure 2b illustrates a correct thermocouple
connection to an INA; both inputs have a path to groundin this case, through a 10-k resistor, biased to a voltage within the INA's input
range, or ground.
Circuit designers often misapply a thermocouple or even a two-wire microphone to the INA's input circuit; the problem arises when the INA
inputs are connected without proper consideration to current paths or biasing. Heed those considerations for the INA's input stage, and you
can be confident that the INA's output voltage is representative of the thermocouple's voltage.
But waitthere may be a problem with the output values of A 1 and A 2 (Figure 1). How would you solve this problem? Comment below. I'd
really like to hear from you!
References
Kugelstadt, Thomas, "Getting themost out of your instrumentationamplifier design," Analog Applications Journal, Texas Instruments, 4Q
1.
2005.
2. Trump, Bruce, "Instrumentation amplifiersavoiding a common pitfall," TI E2E Community, March 27, 2012.
3. Trump, Bruce, "Instrumentation amplifiersthey're not op amps butwhat are they?" TI E2E Community, April 2, 2012. .
24 Back to top

Thermistor-temperature-sensing devices present a design challenge if you intend to use such a device over its entire temperature range.
Typically, the thermistor is a high-impedance, resistive device that eases one of the interface issues as you convert the thermistor
resistance to voltage. The more difficult interface challenge is to capture the nonlinear behavior of the thermistor in a digital representation
with a linear ADC.
The term "thermistor" originates from the descriptor "thermally sensitive resistor." The two basic types of thermistors are negative- and
positive-temperature-coefficient devices. The negative-temperature-coef ficient thermistor best suits precision temperature measurements.
You can determine the surrounding thermistor temperature by using the Steinhart-Hart equation: T=1/(A 0 +A 1 (lnR T )+A 3 (lnR T 3 )). In this
equation, T is the temperature in degrees Kelvin; RT is the thermistor resistance at temperature T; and A 0 , A 1 , and A 3 are constants that
the thermistor manufacturer provides.
The thermistor-resistance change over temperature is nonlinear, as the Steinhart-Hart equation describes. When measuring temperature,
drive a reference current through the thermistor to create an equivalent voltage. This equivalent voltage has a nonlinear response. You can
try to compensate for the thermistor's nonlinear response with a look-up table in your microcontroller. Even though you can run this type of
algorithm in your microcontroller firmware, you need a high-resolution converter to capture data during temperature extremes.
Alternatively, you can use hardware-linearization techniques before digitization and a lower resolution ADC. One technique is to place a
resistor, RSER , in series with the thermistor, RTHERM , and a voltage reference or the power supply (Figure 1). Setting the PGA
(programmable-gain amplifier) at a gain of 1V/V, a 10-bit ADC in this circuit can sense a limited temperature range (approximately 25C).
In Figure 1, note that resolution is lost at high temperatures. Increasing the PGA's gain at these temperatures brings the output signal of the
PGA back into a range at which the ADC can reliably provide conversions that identify the thermistor temperature.
The microcontroller firmware's temperature-sensing algorithm reads the 10-bit-ADC digital value and passes it to a PGA hysteresis-software
routine. The PGA hysteresis routine checks the PGA gain setting and compares the ADC digital value with the trip points that Figure 1
indicates. If the ADC output is beyond a trip-point value, the microcontroller sets the PGA gain to the next higher or lower gain setting. If
necessary, the microcontroller can again acquire a new ADC value. The PGA gain and ADC value then pass to a microcontroller piecewise
linear-interpolation routine.
Obtaining data from a nonlinear thermistor sometimes can seem like an impossible task. You can combine a series resistor, a
microcontroller, a 10-bit ADC, and a PGA to overcome the measurement difficulties of a nonlinear thermistor across a temperature range
greater than 25C.
References
1. Baker, Bonnie C, "Advances in measuring with nonlinear sensors," Sensors magazine, April 1, 2005.
2. "Introduction to NTCs: NTC Thermistors," BC Components data sheet, March 27, 2001.
25 Back to top

1).
output at V OUT .
References
26 Back to top
About the author

Bonnie Baker is a Senior Applications Engineer with the WEBENCH team for Texas Instruments and has been involved with analog and
digital designs and systems for over 25 years. In addition to her fascination with circuit design, Bonnie has a drive to share her knowledge
and experience. She has written hundreds of articles, design and application notes, conference papers, and authored a book: "A Baker's
Dozen: Real Analog Solutions for Digital Designers." Check out her latest blog, On Board with Bonnie, where she navigates the ins and
outs of signal chain designs. You can reach Bonnie at [email protected].
27 Back to top
Content provided courtesy of EDN.

Copyright 2014 EDN. All rights reserved.

value.
(1)
21 Back to top
input stages
Conclusion
References
20 Back to top

value.
(1)
21 Back to top

Equation 1
Equation 2
clear.
References
22 Back to top

23 Back to top

range, or ground.
References
1.
2005.
24 Back to top

with a linear ADC.
greater than 25C.
References
25 Back to top

1).
output at V OUT .
References
26 Back to top
About the author

27 Back to top


input stages
Conclusion
References
20 Back to top

value.
(1)
21 Back to top

Equation 1
Equation 2
clear.
References
22 Back to top

23 Back to top

range, or ground.
References
1.
2005.
24 Back to top

with a linear ADC.
greater than 25C.
References
25 Back to top

1).
output at V OUT .
References
26 Back to top
About the author

27 Back to top


input stages
Conclusion
References
20 Back to top

value.
(1)
21 Back to top

Equation 1
Equation 2
clear.
References
22 Back to top

23 Back to top

range, or ground.
References
1.
2005.
24 Back to top

with a linear ADC.
greater than 25C.
References
25 Back to top

1).
output at V OUT .
References
26 Back to top
About the author

27 Back to top


input stages
Conclusion
References
20 Back to top

value.
(1)
21 Back to top

Equation 1
Equation 2
clear.
References
22 Back to top

23 Back to top

range, or ground.
References
1.
2005.
24 Back to top

with a linear ADC.
greater than 25C.
References
25 Back to top

1).
output at V OUT .
References
26 Back to top
About the author

27 Back to top


greater than 25C.
References
25 Back to top

1).
output at V OUT .
References
26 Back to top
About the author

27 Back to top


Best of Baker's Best - Amplifiers eBook - TI.com
rail.
The V OH maximum specification is V DD -20 mV.
17 Back to top

floor.
where THDOPA =20log(THD OPA-%100) and THDOPA-% is the THD specification in the operational amplifier's data sheet in units of
percentage.
voltage of 4V p-p, and the 16-bit SAR ADC has a THD specification of -99 dB, the system THD is -88 dB. Alternatively, if the op amp's input
stage has a charge pump with a THD specification that is 0.0004%, the system THD becomes -98 dB.
References
18 Back to top

negative rail.
References
19 Back to top


figure is obvious.


input stages
Conclusion
References
20 Back to top

value.
where RIN is the input resistance, the closed-loop gain of each channel is -2V/V. At first glance, it is easy to assume that the closed-loop
bandwidth equals the gain-bandwidth product divided by each channel's gain, or |-2V/V|. Don't make this assumption!
channels (N=3) at a gain of -2V/V, the circuit's closed-loop bandwidth is 25.7 MHz. Additional channels reduce the closed-loop bandwidth,
even though the input signals continue to see a gain of -2V/V.
21 Back to top

Equation 1
Equation 2
clear.
References
22 Back to top

V IN+ and V IN- input pins, the V OA1 and V OA2 output levels of A 1 and A 2 , and the V OUT output-swing capability of A 3 . As you work with
The input voltages at V IN+ and V IN- and the gain of A 1 and A 2 cause the internal output voltages to increase or decrease. Note that the A 1
23 Back to top

range, or ground.
References
1.
2005.
24 Back to top

with a linear ADC.
greater than 25C.
References
25 Back to top

1).
output at V OUT .
References
26 Back to top
About the author

and experience. She has written hundreds of articles, design and application notes, conference papers, articles, including a monthly column
in EDN magazine and on edn.com called "Bakers Best." Additionally, Bonnie has authored a book: "A Baker's Dozen: Real Analog
Solutions for Digital Designers." Check out her latest blog, On Board with Bonnie, where she navigates the ins and outs of signal chain
designs. where she navigates the ins and outs of signal chain designs. You can reach Bonnie at [email protected].
27 Back to top


IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, enhancements, improvements and other
changes to its semiconductor products and services per JESD46, latest issue, and to discontinue any product or service per JESD48, latest
issue. Buyers should obtain the latest relevant information before placing orders and should verify that such information is current and
complete. All semiconductor products (also referred to herein as components) are sold subject to TIs terms and conditions of sale
supplied at the time of order acknowledgment.
TI warrants performance of its components to the specifications applicable at the time of sale, in accordance with the warranty in TIs terms
and conditions of sale of semiconductor products. Testing and other quality control techniques are used to the extent TI deems necessary
to support this warranty. Except where mandated by applicable law, testing of all parameters of each component is not necessarily
performed.
TI assumes no liability for applications assistance or the design of Buyers products. Buyers are responsible for their products and
applications using TI components. To minimize the risks associated with Buyers products and applications, Buyers should provide
adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, or
other intellectual property right relating to any combination, machine, or process in which TI components or services are used. Information
published by TI regarding third-party products or services does not constitute a license to use such products or services or a warranty or
endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of the
third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of significant portions of TI information in TI data books or data sheets is permissible only if reproduction is without alteration
and is accompanied by all associated warranties, conditions, limitations, and notices. TI is not responsible or liable for such altered
documentation. Information of third parties may be subject to additional restrictions.
Resale of TI components or services with statements different from or beyond the parameters stated by TI for that component or service
voids all express and any implied warranties for the associated TI component or service and is an unfair and deceptive business practice.
TI is not responsible or liable for any such statements.
Buyer acknowledges and agrees that it is solely responsible for compliance with all legal, regulatory and safety-related requirements
concerning its products, and any use of TI components in its applications, notwithstanding any applications-related information or support
that may be provided by TI. Buyer represents and agrees that it has all the necessary expertise to create and implement safeguards which
anticipate dangerous consequences of failures, monitor failures and their consequences, lessen the likelihood of failures that might cause
harm and take appropriate remedial actions. Buyer will fully indemnify TI and its representatives against any damages arising out of the use
of any TI components in safety-critical applications.
In some cases, TI components may be promoted specifically to facilitate safety-related applications. With such components, TIs goal is to
help enable customers to design and create their own end-product solutions that meet applicable functional safety standards and
requirements. Nonetheless, such components are subject to these terms.
No TI components are authorized for use in FDA Class III (or similar life-critical medical equipment) unless authorized officers of the parties
have executed a special agreement specifically governing such use.
Only those TI components which TI has specifically designated as military grade or enhanced plastic are designed and intended for use in
military/aerospace applications or environments. Buyer acknowledges and agrees that any military or aerospace use of TI components
which have not been so designated is solely at the Buyer's risk, and that Buyer is solely responsible for compliance with all legal and
regulatory requirements in connection with such use.
TI has specifically designated certain components as meeting ISO/TS16949 requirements, mainly for automotive use. In any case of use of
non-designated products, TI will not be responsible for any failure to meet ISO/TS16949.
Products Applications
Audio www.ti.com/audio Automotive and Transportation www.ti.com/automotive
Amplifiers amplifier.ti.com Communications and Telecom www.ti.com/communications
Data Converters dataconverter.ti.com Computers and Peripherals www.ti.com/computers
DLP Products www.dlp.com Consumer Electronics www.ti.com/consumer-apps
DSP dsp.ti.com Energy and Lighting www.ti.com/energy
Clocks and Timers www.ti.com/clocks Industrial www.ti.com/industrial
Interface interface.ti.com Medical www.ti.com/medical
Logic logic.ti.com Security www.ti.com/security
Power Mgmt power.ti.com Space, Avionics and Defense www.ti.com/space-avionics-defense
Microcontrollers microcontroller.ti.com Video and Imaging www.ti.com/video
RFID www.ti-rfid.com
OMAP Applications Processors www.ti.com/omap TI E2E Community e2e.ti.com
Wireless Connectivity www.ti.com/wirelessconnectivity
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright 2016, Texas Instruments Incorporated

Opamp Amplifiler

Uploaded by

Copyright:

Available Formats

Opamp Amplifiler

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Opamp Amplifiler

Uploaded by

Copyright:

Available Formats

Texas Instruments - TI.

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:43:25 AM]

Chapter 1: Operational amplifiers

When is good enough good enough?

Where did all that racket come from?

Is your amplifier offset way out of whack?

A good holiday-season project

Don't be fooled by your amplifier's bandwidth

Sallen-Key lowpass-filter stopband limitations

Choosing antialiasing-filter amplifiers

Closer to real-world analog filters

1. gain accuracy is important, and

Simply an adjustable low-pass filter

Photo-sensing circuits: The eyes of the electronic world are watching

Transimpedance-amplifier stability is key in light-sensing applications

(Figure 1) The transfer function of this system is:

Transimpedance-amplifier noise issues

Transimpedance-amplifier application: The pulse oximeter

Remote photo sensing

Single-supply amplifier outputs don't swing rail to rail

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/21/14 9:27:15 AM]

Chapter 1: Operational amplifiers

When is good enough good enough?

Where did all that racket come from?

Is your amplifier offset way out of whack?

A good holiday-season project

Don't be fooled by your amplifier's bandwidth

Sallen-Key lowpass-filter stopband limitations

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:58:49 AM]

Where did all that racket come from?

Is your amplifier offset way out of whack?

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:48:43 AM]

Where did all that racket come from?

Is your amplifier offset way out of whack?

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:48:43 AM]

Chapter 1: Operational amplifiers

When is good enough good enough?

Where did all that racket come from?

Is your amplifier offset way out of whack?

A good holiday-season project

Don't be fooled by your amplifier's bandwidth

Sallen-Key lowpass-filter stopband limitations

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:58:49 AM]

A good holiday-season project

Don't be fooled by your amplifier's bandwidth

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:48:43 AM]

Chapter 1: Operational amplifiers

When is good enough good enough?

Where did all that racket come from?

Is your amplifier offset way out of whack?

A good holiday-season project

Don't be fooled by your amplifier's bandwidth

Sallen-Key lowpass-filter stopband limitations

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:58:49 AM]

Sallen-Key lowpass-filter stopband limitations

Choosing antialiasing-filter amplifiers

Closer to real-world analog filters

http://www.ti.com/ww/en/analog/Amplifiers-eBook/[1/14/14 10:48:43 AM]

Choosing antialiasing-filter amplifiers