Top Banner
Tasks 3 and 4: Bipolar Junction Transistor Amplifiers – Design, Build, Test Background: The common-emitter BJT amplifier is a good example circuit for introducing a key skill-set for this and other electronics laboratory courses: making simulation software (e.g., PSPICE, Multi-Sim, etc.) work together with measurement and signal acquisition software (e.g., LabVIEW) to carry a project through (possibly repeated) stages of design, build and test. The simulation software is the primary design aid, giving you predictions about how a design should operate. It can be used to predict the signals at various nodes in a circuit as functions of time. The measurement software serves as the engine for testing, allowing you to compare the actual signals at given nodes with your predictions. I. Designing with a PSPICE Electronics textbooks usually show several variations on single-stage, common-emitter, bipolar junction transistor (BJT) amplifier circuits. To analyze these circuits using the pencil and paper techniques illustrated in these texts, you need to assume that the BJT is biased in the active region (base-emitter junction forward biased, and base-collector junction reverse-biased). When you want to design and build a BJT amplifier, you face a slightly different problem, which can be broken into four parts: 1) how do you select a DC operating point (Q-point)? 2) what resistor values do you use in the circuit in order to bias the BJT at the desired operating point? 3) how do those resistor values affect the key performance specifications of the amplifier; namely, gain, input impedance, and output impedance? 4) can you adjust the resistor values in the circuit to optimize these performance parameters toward the values you want, without adversely affecting the DC operating point too much? It may look like you could go through this design process by doing each part in sequence, but life is not that simple. For example, changes you make in part 4) to get a higher gain may have a bad effect on the position and stability of the DC operating point back in part 1). The discussion to follow shows how you can use circuit analysis software like PSPICE to handle most, but not all, of the calculations you need to do during the amplifier design process. 1. Selecting an operating point If your amplifier were truly operating in the small-signal regime, where you placed the operating point wouldn’t be very important, as long as it remained in the active region. However, if you want the available peak-to-peak output signal to be as large as possible, you must pick the operating point very carefully. An easy-to-remember approximation is that your maximum peak-to-peak output amplitude is always somewhat less than the magnitude of the supply voltage, V cc . This output comes from the collector-to-emitter voltage swinging between the approximate limits of zero and V cc . Therefore an operating point with the collector-to-emitter voltage in the neighborhood of might be a good initial design target. As you saw in Task 2, you can use PSPICE to study the output characteristic curves of the transistor you plan to use in you circuit. Make a Schematic with your candidate transistor in a circuit like the one shown below. Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm 1 of 12 4/23/2008 12:56 AM
12
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: BJT design build test

Tasks 3 and 4: Bipolar Junction Transistor Amplifiers – Design, Build, Test Background: The common-emitter BJT amplifier is a good example circuit for introducing a key skill-set for this and otherelectronics laboratory courses: making simulation software (e.g., PSPICE, Multi-Sim, etc.) work together withmeasurement and signal acquisition software (e.g., LabVIEW) to carry a project through (possibly repeated) stages ofdesign, build and test. The simulation software is the primary design aid, giving you predictions about how a designshould operate. It can be used to predict the signals at various nodes in a circuit as functions of time. Themeasurement software serves as the engine for testing, allowing you to compare the actual signals at given nodes withyour predictions.

I. Designing with a PSPICE Electronics textbooks usually show several variations on single-stage, common-emitter, bipolar junctiontransistor (BJT) amplifier circuits. To analyze these circuits using the pencil and paper techniques illustrated in thesetexts, you need to assume that the BJT is biased in the active region (base-emitter junction forward biased, andbase-collector junction reverse-biased). When you want to design and build a BJT amplifier, you face a slightly different problem, which can be brokeninto four parts:

1) how do you select a DC operating point (Q-point)?2) what resistor values do you use in the circuit in order to bias the BJT at the desired operating point?3) how do those resistor values affect the key performance specifications of the amplifier; namely, gain, input

impedance, and output impedance?4) can you adjust the resistor values in the circuit to optimize these performance parameters toward the values

you want, without adversely affecting the DC operating point too much? It may look like you could go through this design process by doing each part in sequence, but life is not thatsimple. For example, changes you make in part 4) to get a higher gain may have a bad effect on the position andstability of the DC operating point back in part 1). The discussion to follow shows how you can use circuit analysis software like PSPICE to handle most, but not all, ofthe calculations you need to do during the amplifier design process. 1. Selecting an operating point If your amplifier were truly operating in the small-signal regime, where you placed the operating point wouldn’tbe very important, as long as it remained in the active region. However, if you want the available peak-to-peak outputsignal to be as large as possible, you must pick the operating point very carefully. An easy-to-rememberapproximation is that your maximum peak-to-peak output amplitude is always somewhat less than the magnitude ofthe supply voltage, Vcc. This output comes from the collector-to-emitter voltage swinging between the approximatelimits of zero and Vcc. Therefore an operating point with the collector-to-emitter voltage in the neighborhood of

might be a good initial design target. As you saw in Task 2, you can use PSPICE to study the output characteristic curves of the transistor you planto use in you circuit. Make a Schematic with your candidate transistor in a circuit like the one shown below.

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

1 of 12 4/23/2008 12:56 AM

Page 2: BJT design build test

Fig. 1. A Schematic for displaying transistor output characteristics. In Fig. 1, I1 is a simple current source (ISRC in PSPICE) that sets the base current. Similarly, V1 is a simple voltagesource (VSRC in PSPICE) that sets the collector-to-emitter voltage, since the emitter is at ground. Now, let’s use what we know about the active region of BJT’s to determine ranges for base current I1 andcollector-to-emitter voltage V1. The two conditions for active region operation are:

a) The base-emitter junction must be forward biased. This means positive current flows from base to emitter, andthat the base is about +0.7 volts above ground potential. From the way positive current for I1 is defined in theschematic, values of I1 need to be negative in order to bias the BJT in the active region.

b) The base-collector junction must be reverse biased. Thus, the base must be at a potential more negative thanthat of the collector. This implies positive values for V1.

c) For typical transistors in the active region, base currents are small (of the order of microamps). The next step is to use the DC Sweep and Parametric Analyses in PSPICE to plot collector current, Ic, versuscollector-to-emitter voltage V(c) in Fig. 1, with base current as a parameter. Set up the DC Sweep analysis to get alinear sweep of the voltage source, V1, over the range 0 to 10V. Set the Parametric analysis to put the current source, I1, through a Value List consisting of the currents 0, -10u, -20u, -30u, which correspond to base currents of 0, 10, 20,and 30A. If you plot the collector current, IC(Q1) in Probe, the result should look like this:

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

2 of 12 4/23/2008 12:56 AM

Page 3: BJT design build test

Fig. 2. Result of DC Sweep and Parametric analysis of a BJT. You can estimate the for this transistor from Fig. 2. The value is approximately 100. Electronics textbooks oftenemphasize the importance of designing amplifier circuits whose specifications are independent of , because ishighly variable with temperature and the individual transistor chosen. A simple way to see this variability in withPSPICE is with the Temperature Analysis. Check the Temperature Analysis button on the PSPICE Analysis menu and select a very different temperaturefrom the default value of 27 C, and plot IC(Q1) again. The general shapes of the curves will still be similar to Fig. 2,but for 150 C, has increased to about 150. This Temperature Analysis in PSPICE is always available to you as asimple way to check how stable your operating points and your performance specifications are with respect totemperature (or with respect to variations in ), even when the circuit you are analyzing is much more complicatedthan the one in Fig. 1. An example will be given later. Figure 2 also provides you a way to pick a desired operating point for the same transistor in an amplifiercircuit. Suppose your available DC power supply is 10 volts, and you want to avoid distortion in the amplifier output. This suggests an operating point for Vce of 5 volts. Further, suppose that again to avoid distortion, you will try to keepthe base current between the limits of 10A and 30A. This suggests an operating point base current of 20A. Youmust now try to choose resistor values in a given amplifier circuit that will yield this operating point. Since resistorscome in standard values only, and can vary by +ten percent from those values, this process will be approximate. 2. Choosing resistor values for biasing at the desired operating point There are several ways to bias a BJT amplifier circuit. In the example to follow, we will use feedback biasing. When you build your amplifier in the lab, you will be using voltage divider biasing. The amplifier circuit looks like this:

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

3 of 12 4/23/2008 12:56 AM

Page 4: BJT design build test

Fig. 3. Common-emitter amplifier circuit using feedback biasing. Because the capacitors in this circuit are open at DC, only two resistors, Rb and Rc, affect the biasing. If you use thecombination of a DC Sweep and a Parametric Analysis in PSPICE, you can get plots of the collector voltage, V(c), andthe base current, IB(Q1), versus Rb with Rc as a parameter. Set up the DC sweep to let the Global Parameter,RBPAR, vary from 0 to 1.2 Meg. Set up the Parametric analysis to vary the Global Parameter, RCPAR, through aValue List consisting of 300, 500, 1k, 3k, 5k, and 7k. Then, if you create two plots in Probe, one for base current andone for collector voltage, the results should look like Fig. 4., below:

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

4 of 12 4/23/2008 12:56 AM

Page 5: BJT design build test

Fig. 4. Probe plot to search for desired operating point. In Fig. 4, a cursor has been placed at approximately 175K for Rb. Note that the dark blue curves, which correspondto Rc = 1k, intersect the cursor near the desired operating point, V(c)=5v and IB(Q1) = 20A. Next, you can check the bias stability at the operating point by setting the default values of RBPAR andRCPAR to 175k and 1k respectively, and then doing a DC sweep with temperature as the swept variable. PlottingV(c) and IB(Q1) versus temperature leads to the following (quite acceptable) result.

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

5 of 12 4/23/2008 12:56 AM

Page 6: BJT design build test

Fig. 5. Bias temperature stability evaluation of the feedback-biased amplifier using PSPICE. A temperature swing from 0-100 deg C doesn’t affect the bias point too adversely. The DC part of this design is now complete. We have some (tentative) choices for both resistor values thatyield an operating point that lies in the middle of the active region. 3. Determine amplifier gain, input impedance and output impedance for the chosen resistor values from the DCpart of the design. The quantities you calculate to determine the performance of your amplifier comprise a two-port circuit model like theone in Figure 6.

Fig. 6. The basic amplifier model. The input impedance is Zin, the output impedance is Zout, and the gain into an open circuit (load

resistance RL not connected) is A.

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

6 of 12 4/23/2008 12:56 AM

Page 7: BJT design build test

The specifications you need for an amplifier to perform adequately in a given application depend on both the signalsource and the load at the output;

The input impedance, Zin, must be large enough so that the fraction of signal from the source that appears atthe input terminals is large enough to produce the required output. (You can see that the left side of Fig 6. is avoltage divider circuit.)

1.

The output impedance, Zout, must be small enough so that the voltage divider on the output side of Fig. 6 putsenough voltage across the load to produce the required output.

2.

For the values of output and input impedance obtained, the gain into open circuit, A, must be large enough toproduce the required output.

3.

Obviously, there is some trade-off among these parameters in any given application.

The textbook method to determine gain, input impedance and output impedance involves drawing a small-signal model of the circuit. The details of this method depend on which small-signal model for the transistor you use. As you will see here, you can cause PSPICE to do these calculations for you numerically. However, drawing a small-signal model is still a good idea because of the qualitative insights it provides.

The next step in the amplifier design process is to see what values of gain, input- and output impedance are

available from the resistor values chosen to bias the transistor in the middle of the active region. If you run an ACSweep in PSPICE, with the resistances set by the DC part of your design, the following gain and input impedancecurves result:

Fig. 7. Gain and input impedance from the DC part of the design. Shown below in Fig. 8 is the PSPICE schematic used to obtain Fig. 7. The two resistors, Rb and Rc determine all threeamplifier parameters in the list above, and have been made into parameters so we can vary them in a parametricanalysis. You will need to vary the load resistor Rload in order to calculate the output impedance of your amplifierdesign, so it has also been made into a parameter in Fig. 8.

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

7 of 12 4/23/2008 12:56 AM

Page 8: BJT design build test

Fig. 8. Amplifier schematic with parameter values from DC part of design. Comparing Figs. 7 and 8, you can see how curves of A and Zin were obtained:

RL is set to some arbitrarily large value (here 1Meg) to get the gain into open circuit.1.RB and RC are set to their values from the DC part of the design.2.I(C1), the current through the capacitor C1, is identified as the input current to the amplifier by inspection ofFig. 8.

3.

There is no way for you to tell PSPICE to plot the output impedance directly, but from Fig. 6, you can see that if RLwere set equal to Zout, then the resulting V(out)/V(in) (with the output connected to the load) would be half of itsvalue into an open circuit. This means you can use a parametric analysis in terms of RL to find Zout.

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

8 of 12 4/23/2008 12:56 AM

Page 9: BJT design build test

Fig. 9. Gain vs frequency for different load resistance. Figure 9 shows the results of a PSPICE Parametric analysis of gain versus frequency, for different values of RL:

Green: 1Meg Red: 10k Blue: 1k

Note that the mid-band gain with the 1k load resistance is about 80, which is half of the gain into a load resistance of 1Meg (approximate open circuit). From this, you can estimate the output impedance of this amplifier at approximately 1k. Now that you know all the amplifier parameters for the DC part of the design, you are ready for the next step. 4. Make the necessary modifications for your application, checking results using the Parametric Analysis inPSPICE. For the purposes of this example, suppose your application is amplifying a weak signal source for later measurement. Your final measuring circuit is likely to have a large input impedance, which serves as the load for your amplifier. IfRL is large, the output impedance of your amplifier doesn’t matter very much. However, if your signal source has alarge source impedance, Rs, you might want to try to increase the input impedance above its midband value of 550 inFig. 7.

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

9 of 12 4/23/2008 12:56 AM

Page 10: BJT design build test

Fig. 10. Gain(top) and input impedance (bottom) curves for RB = 100k (green), 175k (red) and 330k (blue). Figure 10 shows the effects of varying the base resistance, RB, in this amplifier circuit. Increasing RB from its originalvalue of 175k (from the DC part of the design) up to 330 k increases the input impedance from 550 up to about1k. However, as the upper plot in Fig. 10 shows, there is no such thing as a free lunch: the midband gain hasdropped from 160 down to about 120. Using a similar parametric analysis with the collector resistance, Rc, you can show that varying the collectorresistance in this circuit has almost no effect at all on the input impedance. Like all design processes, this one is iterative and imposes trade-offs. PSPICE provides you with ways toevaluate the trade-offs and come to a decision on a final design.

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

10 of 12 4/23/2008 12:56 AM

Page 11: BJT design build test

II. Designing an amplifier for later Build and Test in Task 4

Design a voltage-divider biased common emitter amplifier by selecting the appropriate standard values for the threeresistors, Rb1, Rb2, and Rc, in the circuit above, where Vcc = 10v:Use the PSPICE design techniques discussed above to make your design meet the following requirements:

Input Impedance: greater than 500 at midband.Output impedance: less than 1kat midband.Gain into open circuit load: as high as possible at midband consistent with the above requirements.

Required Notebook EntriesTask 3-1: ____Print out a PSPICE schematic of your final design showing the standard resistor values to be used. Task 3-2: ____Print out a simulation plot of input impedance vs. frequency for your design, and estimate its value atmidband.Task 3-3: _____Print out a parametric analysis plot of V(out)/V(in) for a Value List of values for the load resistor,Rload , that will allow you to calculate the output impedance and the gain into an open circuit load.Task 3-4: _____Use the results of Task 3-3 to calculate the midband values of the output impedance and gain into anopen circuit load for your amplifier design.Task 3-5: Use a small-amplitude sine wave with a frequency at midband as an input to your simulated amplifier andrun a Transient Analysis:

Task 3-5.1:______Print out a simulation plot of the input and output voltages vs. time on the same graph(different Y-axes) for a value of the input amplitude small enough so that no distortion is visible in the outputvoltage waveform.Task 3-5-2:______Print out a simulation plot of the input and output voltages vs. time on the same graph(different Y-axes) for a value of the input amplitude just large enough so that distortion is visible in the outputvoltage waveform.

Task 3-6: ______ Using an input amplitude small enough so that no distortion in the output waveform is visible (Task3-5.1), print out a simulation plot of the collector voltage waveform vs. time and use it to calculate the operating point(quiescent or Q-point) value for collector voltage.Task 3-7: ______ Using an input amplitude small enough so that no distortion in the output waveform is visible (Task

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

11 of 12 4/23/2008 12:56 AM

Page 12: BJT design build test

3-5.1), print out a simulation plot of the collector current waveform vs. time and use it to calculate the operating point(quiescent or Q-point) value for collector current.Task 3-8: ______Build the amplifier you designed on your proto-board and save it for future test and measurement. Additional Task Information: General: In these tasks, the term “midband” means a frequency high enough so that the capacitors in thecircuit are nearly short-circuits. This corresponds to frequencies where the gain and input impedance are essentiallyflat with respect to frequency as shown in the above figures like 9 and 10. You will need to do both Transient analysis and AC Sweep analysis in PSPICE to do these design simulations. If youneed review on these, see the PSPICE summary link. Task 3-1: Expect to run many PSPICE parametric analyses using different combinations of standard resistorvalues in order to come up with a good circuit design. You only need to show your final design in your notebook. Task 3-2 thru 3-4: The impedances and gains you are obtaining in these tasks are meaningful only for thesmall-signal model for your amplifier. Be sure that when you make the simulations for these tasks that the inputsine-wave amplitude is small enough so that no distortion in the output is visible. Tasks 3-6 and 3-7: In your studies of the BJT amplifier in lecture, all time-functions for voltage and currentconsist of sinusoidal waveforms riding on DC levels. The DC levels represent the Q-point (or Operating Point, or BiasPoint) values for the voltage or current being measured. Therefore for example, if the DC level of the collectorcurrent is 10 mA in your time-domain simulation, then that is the Q-point for collector current. Good time-domainplots from your simulations should show 3-4 cycles of your waveforms under steady-state conditions. .

Bipolar Junction Transistor Design with PSPICE http://users.ece.utexas.edu/~buckman/ee438Fall04Tasks/Task%203.htm

12 of 12 4/23/2008 12:56 AM