Laboratory 1 - Microwave Electronics Laboratory at UCSB

ECE 2C Laboratory 2

Audio Amplifier/Speaker Driver and Microphone Preamplifier In the first week of this lab you will construct a low-power audio amplifier/speaker driver based on the LM386 IC from National Semiconductor. The audio amplifier will be a selfcontained, battery-operated component. In the second part of the lab you will construct a microphone circuit using a compact electret condenser microphone cartridge. These circuit modules are important building blocks of many audio communications systems, and will be used in our ultrasonic transceiver system. There are several things you can learn in this lab if you pay attention. By studying this document and experimenting with the components and circuits in the lab, try to understand the following: • • • • • • •

Electrical characteristics of audio speakers Characteristics of condenser microphones Design of single-supply battery-operated op-amp circuits Use of diode limiters/clamps for input protection Use of bias de-coupling capacitors on ICs Active filters for tone control How to choose DC blocking capacitors

We will discuss the analytical aspects of active filter design in lecture. You will have an opportunity to design an active filter yourself in Lab #3. The objective of the lab is not simply to create a working circuit, it is to learn about circuits! So, as you progress through the lab, try to understand the role of each component, and how the choice of component value may influence the operation of the circuit. Please tinker with component values: that is an especially valuable way to learn. Ask yourself questions such as: Why is this resistor here? Why does it have this resistance value? Why is this blocking capacitor 1µF instead of 0.1µF or 10µF or 100µF? Why was this particular op-amp chosen? It is only when you can answer such questions that you will truly understand the labs and progress towards designing your own circuits. In the process you may even find a better solution; in fact, the present laboratory experiments in this course have several slight modifications from the original that were inspired by student feedback.

ECE 2C

Circuits, Devices, and Systems

Lab 2

Table of Contents Pre-Lab Preparation ............................................................................................. 3 Required Equipment List .................................................................................................. 3 Schedule for Lab #2 .......................................................................................................... 3 Parts List............................................................................................................................ 4 Schematics for Lab #2....................................................................................................... 5 Background Information ...................................................................................... 8 Audio Speakers ................................................................................................................. 8 Microphones...................................................................................................................... 9 First Week: Battery-Operated Audio Amplifier................................................. 11 Some comments on our choice of ICs............................................................................. 11 Step 1 – Speaker Equivalent Circuit ............................................................................... 12 Step 2 – Assemble Amplifier on Breadboard.................................................................. 13 Step 3 – Add Summing network ..................................................................................... 15 Step 3.5 – **Optional** Tone-Control Circuit............................................................... 16 Step 4 – Hardwire the Amplifier ..................................................................................... 17 Possible Improvements ................................................................................................... 18 Second Week: Microphone Pre-Amp/Filter ...................................................... 20 Step 5 – Microphone Pre-amplifier................................................................................. 21 Step 6 – Putting it All Together ...................................................................................... 22 Possible Improvements ................................................................................................... 23 Lab 2 Record ....................................................................................................... 26

© Robert York, UCSB 2006

2

ECE 2C

Circuits, Devices, and Systems

Lab 2

Pre-Lab Preparation Read through the lab experiment and data sheets to familiarize yourself with the components and assembly sequence. Bring your soldering iron, tools, and power supply from lab #1 (the Lab #1 power supply will be used with the microphone pre-amp circuit). Optional: simulate the circuit response in advance using MultiSim.

Required Equipment List Power supply from Lab #1 Solderless breadboard and Wire jumper kit Soldering iron and solder Bench Function generator and oscilloscope

Schedule for Lab #2 To stay on schedule, you must do the following: Week #1:

Complete Steps 1-4 (Audio amplifier)

Week #2:

Complete Steps 5-6 (Microphone circuit)

The audio amplifier project is more difficult and time-consuming than the microphone preamp, so part of week #2 may be used to finish the audio amp. Steps 1-3 (breadboarding and testing) can and should be done in lab. Step 4 (soldering/hardwiring) can and should be done outside lab. Step 4 is very time consuming, so you must try to get through steps #1-3 quickly and efficiently. Also: double check your kit against the parts list that follows. Due to the large number of kits that must be made up, mistakes can happen. If you are missing parts, contact your TA or the ECE Shop to remedy the situation quickly.

© Robert York, UCSB 2006

3

ECE 2C

Circuits, Devices, and Systems

Lab 2

Parts List Laboratory #2

Audio Amplifier and Microphone PreAmp

Microphone Preamp

Audio Amplifier

Qty Description

Circuit

1 1 1 2 1 2 8 2 1 1 1 1 2 4 1 2 1 2 1 4 1 1 1 1 1 1 2 6" 6"

Audio Speaker, 8-Ohm, 2 Watt, 0.3-9KHz LM386N-3 Low-Voltage Audio Power Amplifer LM358 Low-Power Dual Op-Amp 8-pin low-profile IC socket 10 Ohm 1/2W resistor 2.2k 1/4W resistor 10k 1/4W resistor 100k 1/4W resistor 10k trimpot 20k trimpot 470 uF 16V electrolytic capacitor (PC lead) 100uF 16V electrolytic capacitor (PC lead) 10uF 16V electrolytic capacitor (PC lead) 1uF capacitor 0.033uF capacitor (CK05 low-voltage ceramic ) 0.022uF capacitor (CK05 low-voltage ceramic ) 0.1uF capacitor (CK05 low-voltage ceramic ) 1N4148 small-signal Silicon diode 4.5" x 5.67" vectorboard Rubber feet Stereo Audio jack 3.5mm male-to-male stereo patch cord 9V battery leads 9V battery holder (adhesive backed) 9V battery Doubled-sided adhesive tape for mounting speaker flea clips #22 stranded wire (black) #22 stranded wire (red)

1 1 1 2 2 1 1 1 1 2 2

Electret Condenser Microphone Cartridge LF353N Dual Wide-band JFET Op-amp 8-pin low-profile IC socket 2.2kOhm 1/4W 10kOhm 1/4W 33kOhm 1/4W 10k trimpot 1uF 25V electrolytic (PC lead) 10uF 25V electrolytic (PC lead) 100uF 25V electrolytic (PC lead) 0.001uF capacitor (CKO5 low-volt. Ceramic)

© Robert York, UCSB 2006

4

U2 U1 R10 R13,R14 R1-5,R9,R12,R15 R6-7 R8 R11 C9 C10 C4,C6 C1,C2,C3,C5 C7 C11,C12 C8

MIC U1 R3,R5 R1,R2 R4 R6 C1 C3 C6,C7 C2

ECE 2C

Circuits, Devices, and Systems

Schematics for Lab #2

Schematic for the audio amplifier.

© Robert York, UCSB 2006

5

Lab 2

ECE 2C

Circuits, Devices, and Systems

Schematic for the audio amplifier with optional tone-control circuit.

© Robert York, UCSB 2006

6

Lab 2

ECE 2C

Circuits, Devices, and Systems

Schematic for the microphone preamplifier.

© Robert York, UCSB 2006

7

Lab 2

ECE 2C

Circuits, Devices, and Systems

Lab 2

Background Information Audio Speakers Audio speakers convert electrical signals into mechanical motion. The most common speakers for high-quality audio are constructed as shown in the figure below∗. The electric signal is passed through a coil of wire (the “voice” coil), which is suspended a strong magnetic field provided by a permanent magnet. A time-varying current in the coil leads to a mechanical deflection relative to the magnet. The coil is attached to a lightweight conical membrane (usually made from a heavy-grade paper) that couples the mechanical motion of the coil to the surrounding air molecules.

Band of operation

(a)

(b)

(a) Cross section and (b) Impedance curve for a typical 8Ω dynamic speaker.

If you look through catalogs of audio-speakers, you will find many different sizes and price ranges. For example, check out http://www.tb-speaker.com/tbp.htm. High-quality audio speakers can be quite expensive, and are designed for a flat, omni-directional frequency response in the specified operating range. See the example below. Small inexpensive speakers, such as the ones used in this lab, have poor low-frequency response and limited power-handling capacity. Speakers are commonly specified by their frequency response, impedance level, and power-handling capacity. Typical speaker impedances are “8Ω” or “4Ω” or “16Ω”. This is often a source of confusion because it suggests the speaker is modeled by a constant resistance of this value. In reality there is a significant reactive component of impedance and hence a strong variation of impedance with frequency, as shown in the figure above. The impedance can also depend strongly on the surroundings. For example, a speaker

∗

D. B. Weems, Designing, Building & Testing your Own Speaker System, Tab Books: Blue Ridge Summit, PA, 1984 © Robert York, UCSB 2006

8

ECE 2C

Circuits, Devices, and Systems

Lab 2

measured in isolation (the “free-air” response) will have a different impedance than one mounted in a wooden enclosure. The AC impedance often displays at least one resonance, relating to the size and stiffness (“compliance”) of the cone. Electrically this mechanical resonance can be modeled by a parallel RLC circuit. The operating Rc Lc frequency range for the speaker begins at or above this resonance, and in this range the speaker impedance is well modeled by just a resistance in series with an inductance. The Lm Cm Rm inductance comes from the voice coil itself. The resistance term is largely the DC resistance of the voice coil, with a small additional contribution which represents energy conversion to mechanical motion. Equivalent circuit for an audio speaker The combination of the two resistances is usually close to the specified impedance of the speaker (e.g. 8Ω) in the middle of the operating range. A pretty good equivalent circuit model for any speaker is shown above and can be created by measuring or finding the following 5 parameters: ƒ ƒ ƒ ƒ ƒ

DC Resistance of the voice coil, Rc Free-air resonant frequency, fs Input resistance Rt at f=fs (the impedance is purely resistive at resonance) Bandwidth ∆f of the resonance (between the points where |Z|=Rt/2). The Q-factor is then Qms=∆f / fs Inductance Lc of the voice coil (measure at a high frequency, f>>fs)

The free-air mechanical resonance is modeled by a parallel RLC circuit. Since the Qfactor of a parallel resonator is Qms = ω RC , and the resonant frequency is f s = 1/ 2π LC , the equivalent circuit parameters can be found as Rm = Rt − Rc

Cm =

Qms 2π f s Rm

Lm =

1 (2π f s ) 2 Cm

One last point about speakers: for a single-speaker application, the polarity of the speaker is irrelevant. However, if multiple speakers are used, it is important to connect them together with the correct polarity so that the voice coils and cones will move in phase with each other. For this reason, speaker terminal are often designated with a “+” or “-“ symbol.

Microphones A speaker can be used in reverse to create a microphone. In the case the incoming sound wave leads to a mechanical deflection of the cone and voice-coil. According to Faraday’s law, a time-varying current will be induced because the coil is moving in through a magnetic field (produced by the permanent magnet). Although any speaker could be used for a microphone, most speakers are unnecessarily large for this purpose, except in simple intercom applications where it is common to use the same component to perform both the speaker and microphone functions.

© Robert York, UCSB 2006

9

ECE 2C

Circuits, Devices, and Systems

Lab 2

Another type of microphone is the “condenser” microphone, which exploits electrostatic forces instead of magnetic induction. The “cone” in this case is a thin metallic membrane that forms one side of a parallel-plate capacitor (condenser is an old-fashioned term for capacitor). An incoming sound wave causes the membrane to vibrate and hence the capacitance changes. If the capacitor is charged through an external pull-up resistor as shown in the figure below, the time-varying capacitance will induce a time-varying current through the resistor and hence an AC voltage. +V

+V

R

R out

sound

out

sound

(a)

(b)

Two types of condenser (capacitance) microphones. (a) standard condenser mic. (b) electret condenser mic. The electret uses a special capacitance with a fixed charge (an electret) integrated with a FET buffer. The dashed line indicates the internal detail of the microphone cartridge; the resistor is supplied externally by the designer.

An especially popular type of condenser microphone today is the “electret condenser” variety. In this case the time-varying capacitance is used to modulate the gate voltage on a built-in FET, which buffers and amplifies the signal. The capacitance is constructed from an electret material that has a permanent polarization or fixed charge, hence it is not necessary to bias the capacitor as in a conventional condenser mic, but the FET buffer does require biasing. An illustration of the electret microphone and biasing is shown in the figure above. In both cases, the output signal should be AC-coupled to the amplifier through a largevalue DC blocking capacitor. The pull-up resistor R determines the output impedance of the device, and sets the bias voltage on the FET buffer. The FET typically draws about 0.1-0.2 mA, and can operate over a wide range of voltages, so the choice of R is not critical. Construction of an electret microphone cartridge Most microphones generate very low-level signals, on the order of a millivolt or less. Hence we always need to amplify the audio signal substantially. This is the purpose of the circuit implemented in the second part of this lab. If you are interested, two good sources of information on microphones are: http://en.wikipedia.org/wiki/Microphone http://www.acoustics.salford.ac.uk/acoustics_world/id/Microphones/Microphones.htm

© Robert York, UCSB 2006

10

ECE 2C

Circuits, Devices, and Systems

Lab 2

First Week: Battery-Operated Audio Amplifier Follow the instructions below CAREFULLY. Failure to do so could result in damage to the lab equipment, destruction of parts, and possible injury to you and your lab partner. ‰

Each step begins with a check box like the one at the left. When you complete a step, check the associated box. At the end of this document is a LAB RECORD in which you will record key observations and results from the experiment. When you are finished with the lab, you must demonstrate a working circuit and a completed lab record, and get this signed and certified by one of the lab TAs.

Some comments on our choice of ICs Why the LM386? At first glance this project looks simple; all we need is an amplifier to drive an audio speaker, right? Why not just use a simple op-amp gain stage? Well, one immediate challenge is that speakers have a low impedance, typically 8-Ohm. Even for a low-power 0.5W audio amplifier, P = I 2 R / 2 requires that the amplifier must be able to source a peak AC current of around 350 mA with an 8-Ohm load. This is a lot more than a typical opamp is designed to provide. For example, the LM358 and LF353 (two op-amps used in this lab) can source around 20-40mA at most. So we need a device that can handle higher currents. We also want a device that can be operated from a single supply voltage (a battery in this case). It is possible to make a suitable amplifier using an op-amp and a couple of external power transistors, but we went with a simpler solution, and chose a chip that was specifically designed as an audio driver, the LM386.

LM386 internal circuit and 8-pin DIP pin assignments.

This is an old chip that has been a popular choice for low-power audio applications for many years. There are many other audio amp ICs on the market (LM380 and LM383 are similar but higher-power amps from National), but the LM386 is sufficient for our purposes. It comes in three-flavors, the LM386-1, LM386-2, and LM386-3, which can

© Robert York, UCSB 2006

11

ECE 2C

Circuits, Devices, and Systems

Lab 2

provide 0.3W, 0.5W, and 0.7W respectively, more than adequate for this lab. Another nice thing about the LM386 is that the gain-frequency curve can be shaped with some external feedback components, so it is a very flexible device. If you do a web search on “LM386”, you will find many examples of clever circuits that people have come up with over the years. The one in this lab is a minor modification of a circuit described in the data sheet. Why the LM358? For the remainder of the circuit (a simple unit-gain summing network) we have chosen an LM358 op-amp, which is a low-power device that can be operated from a single voltage supply, therefore appropriate for battery-operated circuits. Low-power op-amps draw very little quiescent current and hence help prolong the life of the battery. The LM358 is a garden-variety op-amp and there are many other possibilities that could be used, but this one is very cheap! We also chose a dual device in case you choose to do the optional tonecontrol circuit. It doesn’t matter which of the two internal op-amps you use for the summing network or the tone-control circuit, they are identical.

Step 1 – Speaker Equivalent Circuit We’ll start by measuring the characteristics of the audio speaker in your kit. This is a small, general-purpose speaker used for intercoms and other inexpensive audio systems. ‰

Using the R-L-C meter in the lab, measure the coil resistance and coil inductance (the meter uses a frequency that is well above the speaker’s mechanical resonance).

‰

Now we’ll use the function-generator at your workstation to drive the speaker. Since the speaker has a small impedance, we must be careful to keep the amplitude of the function generator low so that it doesn’t have to source a lot of current. Adjust your function generator to produce a sinusoidal wave with an AC amplitude of Vg= 200 mV and zero DC offset.

‰

Electrically connect your speaker to the function generator as shown below, using a small resistor R in series with the speaker (something around 50 Ω, or two 100 Ω resistors in parallel, seems to work well).

R

+ Vg

⎛ VL Z = R⎜ ⎜ Vg − VL ⎝

VL

-

⎞ ⎟⎟ ⎠

|Z| ‰

The voltage across the speaker is proportional to the magnitude of its impedance, so by monitoring the voltage we can map out the variation of impedance with frequency. Find the resonant frequency (somewhere in the range of 100-200 Hz), and record the bandwidth. Then compute the equivalent circuit parameters as discussed in the background information.

© Robert York, UCSB 2006

12

ECE 2C

Circuits, Devices, and Systems

Lab 2

Step 2 – Assemble Amplifier on Breadboard The schematic for an LM386-based amplifier is shown below. We will first construct this amplifier using your plastic solderless breadboard and bench power supply. Once you debug the circuit and demonstrate that it works properly, you will “hard-wire” it onto a vectorboard and power it from a 9V battery. The idea of breadboarding the circuit first is not to create more work! It is good practice in general, since it allows you to debug your design more easily, allows you to experiment with component values, and more importantly will help preserve your battery!

Schematic of LM386 Amplifier.

‰

First study the schematic carefully, and locate all the necessary parts in your kit. Find the LM386 chip and compare it with the figure below and with the schematic to correlate pin numbers with the circuit connections.

‰

Configure your breadboard with connections to the power supply and appropriate wiring to the power busses.

‰

Add the LM386 IC and make the power (pin 6) and ground (pin 4) connections as shown in the schematic. Add the 100uF and 10uF bias/bypass capacitors.

‰

Using the schematic above, add in the resistor and capacitor elements as shown. Note the role of each component: o 1µF capacitor: DC blocking, AC couples the input signal o 10k trimpot: voltage-divider for volume control o 10µF capacitor: Internal AC bypass (see data sheet). o 10kΩ resistor and 33nF capacitor between pins 1-5: bass-boost feedback circuit (see data sheet), helps compensate for the poor low-frequency response of our speaker © Robert York, UCSB 2006

13

ECE 2C

Circuits, Devices, and Systems

Lab 2

o 470µF capacitor: DC blocking (why is this so large compared with other blocking/bypass capacitors?) o 10Ω resistor and 0.1µF capacitor: a “snubber” circuit for high-frequency stabilization, prevents potential oscillation due to inductive loading. ‰

Add in the back-to-back diodes at the input terminal of the device. These diodes clamp the input voltage on pin 3 at +/- 0.7 V, to insure that excessive voltage is never applied to this pin, which could damage the circuit.

‰

Add the speaker. You will need to first solder wire jumper leads to the speaker terminals.

‰

You are now ready to test the circuit. First, connect your bench function generator to the oscilloscope and adjust for a 1kHz sinewave with a 0.2 V amplitude, and verify on the oscilloscope. Then apply this signal to your amplifier along with the +9V bias. If all goes well, you should hear a tone. Adjust the volume control as necessary.

Function Generator

Power Supply

Speaker +9 V Amplifier Circuit

Testing the circuit

‰

With the volume adjusted to a reasonable level, sweep the frequency to determine the lowest and highest audible tones and record.

‰

Observe the output waveform at pin 5 on the oscilloscope for a 0.5 V amplitude input signal at 1kHz. Note that the output is level shifted to approximately half the power supply voltage. Can you see distortion as the volume is increased? Why is this happening?

‰

Remove the LM386 temporarily and apply a +/- 1V sinusoid at 1kHz to the amplifier input and observe the waveform at pin 3 under full-volume conditions. You should observe the clamping or limiting action of the input diodes. Record this waveform in the LAB RECORD.

‰

Reconnect the IC and speaker and insert the bench multimeter in series with the power supply as shown above to measure the DC current flow into the amplifier. Record the current under the operating conditions described in the LAB RECORD.

© Robert York, UCSB 2006

14

ECE 2C

Circuits, Devices, and Systems

Lab 2

Step 3 – Add Summing network Now add the summing network. This is used to combine several audio inputs into a common output signal. For example, if we want to hook our “mono” amplifier to a “stereo” signal, such as produced by an MP3 player or CD player, then we need to add the left and right stereo channels together so that the amplifier can amplify both simultaneously and deliver them to the speaker. Stereo Input Jack C1

R1

3

10kΩ

R4

C2

R2

10kΩ

1uF-POL

10kΩ

1uF-POL 1

2

C3

Aux Input

4

R3 10kΩ

1uF-POL

LM358AN

2

U1A

1

3

R5 10kΩ

8

VCC

VCC

Sets ref level at Vcc/2

9V

9V

R6 100kΩ C4 10uF-POL

R7 100kΩ

Single-ended unity-gain summing network and LM358 pinout

‰

Add the summing circuit as shown above. Just put in the AUX channel for now, don’t worry about the stereo jack and associated RC elements. There are two important points to understand here, both of which are related to the use of a single (battery) supply. First, in the textbook dual-supply op-amp summing circuit, the positive (non-inverting) input is usually grounded. The voltage at this pin defines the “reference” voltage for the input signals, which is the desired average or DC output level in an audio system. We usually choose this level to maximize the possible AC voltage swing, hence it should be midway between the two supply voltages. For a dual (bipolar) supply, this would be ground (0 V), but for our 9V battery supply, it is half the supply voltage, or 4.5 Volts. This is accomplished using the two 100k resistors in a voltage divider (we choose 100k resistors here to minimize the current draw from the battery). The 10µF bypass capacitor helps prevent this voltage level from fluctuating during operation, maintaining a constant potential and hence a good AC ground. Secondly, since the input and outputs of this summing network will be capacitively coupled, the data sheet recommends adding an external resistive DC current path to ground at the output of the op-amp, to increase the bias current through the transistors in the opamp’s output stage during operation. This is what R5 is for. You’ll have to wait until ECE 137AB to fully appreciate the details. Again, we’re choosing a large value resistor here to minimize the DC current draw from the battery.

© Robert York, UCSB 2006

15

ECE 2C

Lab 2

Once everything is put together, test your circuit as in the previous step, applying the signal from the function generator to the input ports. When you are confident that your circuit works, proceed to the next step.

Step 3.5 – **Optional** Tone-Control Circuit The LM358 is a dual op-amp, so we have an extra op-amp to play with. Why not use it? This is not a required element of your project, but is a common feature in simple audio systems to improve/adjust the sound quality. There are many possibilities for tone-control circuits. The basic idea is to selectively boost or “cut” (attenuate) signals depending on their frequency. This is usually done using an op-amp gain stage with frequency-selective feedback. The circuit below is a simple example that basically starts with a unity-gain inverting amplifier configuration, and adds series RC networks in parallel with feedback resistors. A potentiometer controls the tapping point for the feedback. When the wiper is at its midpoint, both sides of the feedback network are identical, and the system has unity gain for all frequencies (assuming identical component values). When the wiper is at the top, there is a low-frequency gain of around 3 (10 dB), and high-frequency attenuation of about 1/3 (-10dB). When the wiper is at the bottom, the action reverses so that the low-frequencies are cut and the highfrequencies are boosted.

input

output

Single-knob tone control circuit (adapted from http://members.tripod.com/roymal/ReverbTone.htm ). This makes use of the second op-amp in the LM358. You can use the same Vref as for the summing circuit. 20 15 10 5 0 -5 -10 -15 -20 10 100 1,000 10,000 100,000 Frequency, Hz

Voltage Gain, dB

‰

Circuits, Devices, and Systems

Bode plots for the tone-control circuit at opposite extremes of the potentiometer setting.

© Robert York, UCSB 2006

16

ECE 2C

Circuits, Devices, and Systems

Lab 2

Step 4 – Hardwire the Amplifier Now we are ready to put everything together, and add the stereo jacks and battery supply. Before we go further, let’s consider the standard 3.5mm audio connectors. They look like the picture below, and are probably familiar to everyone.

Gnd Ch 2 Ch 1

Standard 3.5mm stereo audio plug

The stereo “plug” (sometimes called “male” instead of “plug”) has three separate cylindrical connectors, each separated by a thin plastic insulator. The largest of these is the ground connection. The other two carry the left and right channels of the stereo signal. The 3.5mm designation refers to the diameter of the cylindrical conductors. There are other diameters on the market, but 3.5mm is the standard for audio.

Pin 1

3.5mm stereo audio jack used in this lab.

In your kit, you should have been provided a stereo “jack” (or “female”) as shown above. There are a large variety of different jack configurations, depending mostly on how they will be mounted (i.e. Surface mount, PC board, panel mounted, etc.). Some jacks also have more than three pins which are connected in a way that gives an open or short connection when a plug has been inserted or removed from the jack. The one we will use is just a simple 3-terminal PCB stereo jack. On the backside you will see each pin marked by a number, corresponding to the diagram above. Okay, lets build the circuit!

© Robert York, UCSB 2006

17

ECE 2C

Circuits, Devices, and Systems

Lab 2

‰

Transfer the amplifier design to the vectorboard and “hardwire” everything into place. Use the IC socket provided in the parts kit so that if anything goes wrong with the LM386 or LM358, you can replace it without having to resolder the circuit.

‰

Add the 9V battery leads supplied in your kits. If you were given a battery holder, install that as well.

‰

Attach the stereo jack to the edge of the vector board. Connect pins 2 and 5 to the summing network, and connect pin 1 to ground.

‰

Attach the remaining “aux” input on the summing network to a flea clip.

‰

Lastly, attach the speaker to the vectorboard. If the speakers have a mounting bracket, you will be given some screws/nuts to attach the speaker to pre-drilled mounting holes, otherwise you will be given some double-sided tape to secure the speaker.

‰

Test your circuit using the bench power supply as in Step 2. When it is working, add the battery and test again under battery operation. Answer all the remaining questions in the lab record.

‰

Using the patch-cord in your kits, you can now test your amplifier by hooking it up to the headphone jack on a portable CD player, MP3 player, or computer.

Congratulations: you now have a portable audio amplifier!

Possible Improvements The circuit you’ve just built is a nice building block for many projects where it is desirable to have a simple, inexpensive and versatile audio output. The obvious deficiencies with this system are: 1) it is monophonic (not stereo); 2) it has relatively low output power; 3) it is not very efficient in terms of battery life; and 4) the small speaker size limits the lowfrequency response. These are all areas that can be easily improved upon if you are interested. For a true stereo output we would need separate amplifiers for the left and right channels, not to mention separate left/right speakers. Nowadays you can often pick up a cheap set of speakers designed for use with computer systems that often have surprisingly good performance, much better than the little paper-cone speaker we chose for this lab. In terms of power and efficiency, there are now several single-chip Class-D audio poweramps on the market that can generate much high powers with very little power consumption. As one of many such examples, the National Instruments LM4663 chip is a self-contained 2W stereo amplifier chip that can directly drive two external speakers and a set of headphones, powered off a single 5V supply, with >83% efficiency, and it costs around $1.50! Clearly if you really needed to build an audio power amp for a critical engineering application, a single-chip solution like the LM4663 is probably going to be much more attractive in terms of size and cost than piecing together a system from several discrete parts, as we did in this lab. On the other hand, you most certainly learned more

© Robert York, UCSB 2006

18

ECE 2C

Circuits, Devices, and Systems

Lab 2

about the circuit building it from scratch then you would have by simply hooking up the LM4663 circuit below!

Another possibility for future work is to create a more sophisticated tone control, like a multi-band graphic equalizer. Most work on similar principles as the single-knob tone control circuit we used here. A web search will turn up many possibilities, and you can also come up with your own ideas. Lastly, you may wish to put your circuit in some kind of enclosure to protect it, with shaft & knob potentiometers for the volume and tone controls, and perhaps a jack for headphones (note that there are headphone jacks available that have an extra set of contacts for disabling the speaker connection when a headphone plug is inserted). Another nice addition would be a power switch with an LED power indicator. We didn’t put an LED in the project to reduce the quiescent current draw and thus keep the battery alive longer.

© Robert York, UCSB 2006

19

ECE 2C

Circuits, Devices, and Systems

Lab 2

Second Week: Microphone Pre-Amp/Filter This week we will implement the microphone preamp system. This will later be used to drive the AM modulator as part of the ultrasonic transmitter, so instead of hardwiring this circuit, just build it in a corner of your breadboard and leave it there until we get to Lab 4. We will test it using your audio amplifier. Why the LM353? Almost any generic op-amp can be used for a microphone pre-amplifier. The signal levels are very low, and all we need is a simple gain stage with a high input impedance. If we were to operate the microphone circuit from a battery, the LM358 would be fine, but this microphone circuit will be used in our AM modulator circuit of Lab #4, which will be powered from a bipolar +/-12V supply, so we went with a dual-supply version. Like the LM358, the main virtue here is cost and availability!

LF353 pinout, and example power connections and bias decoupling capacitors.

This is a good place to offer a tip about using op-amps in analog or mixed-signal circuits: it is considered good practice to always include bias-decoupling capacitors for each IC in the circuit. The figure above shows the bias arrangement for the LF353 on a breadboard. To understand why these are useful, consider the fact that most circuits are processing time-varying or rapidly switching signals, and thus the DC current draw from the powersupply rails is not a constant. Instantaneous current surges in one part of the circuit can cause the supply voltage or current to vary in other parts of the circuit. Thus the biasdecoupling capacitors at each IC serve as a local charge reservoir to help maintain a steady supply current during operation. Obviously the bigger the capacitance the better, but 0.1uF is a very common choice. Unless your design is tightly constrained in terms of cost and parts count, it is advisable to add these decoupling capacitors in all your op-amp circuits.

© Robert York, UCSB 2006

20

ECE 2C

Circuits, Devices, and Systems

Lab 2

Step 5 – Microphone Pre-amplifier ‰

Configure your solderless breadboard with +/- 12 Volt and ground connections using the bench power supply. Add the large 100uF supply capacitors (C6 and C7) onto your breadboard as shown in the schematic.

‰

Begin construction by inserting the LF353 dual op-amp IC on the breadboard. Add power connections (pins 4 and 8) and bias decoupling capacitors as shown earlier.

‰

Add the components for the first stage of amplification. Note we are AC-coupling the microphone to the amplifier, and this sets a low-frequency cutoff (calculate it!). Leave out C2 for now (the capacitor in the feedback path). We will talk about this later. You should have built the following circuit, minus the microphone cartridge:

‰

Now test the amplifier circuit without the microphone. Using your bench function generator, apply a 1V sinusoid and sweep the frequency from 10Hz to 100kHz to locate the lower corner frequencies of response and record this in your lab record.

‰

Now temporarily add in capacitor C2. This introduces another pole in the system to reduce the high-frequency gain (calculate it!). Sweep the frequency to determine the resulting upper cutoff frequency, and record this in the lab record along with the voltage gain in the passband (around 1 kHz). Remove C2 when you are finished.

‰

Now let’s add the microphone into the circuit. In our circuit we’re biasing the mic through a resistor divider. This is not common, but we need to bias the mic below its maximum rated bias (10V), and since the mic buffer current draw is somewhat unpredictable, it is safer to use a divider as shown. The actual bias voltage will be somewhere in the range of 46V depending on how much current the mic will draw.

‰

The electret condenser microphone cartridge is shown below. Depending on which unit was selected this year, it may or not have mounting pins pre-installed. If it doesn’t, you will need to add some wires as shown so that it can be easily inserted into your breadboard. Take care to note which is the ground pin: that is the pin that is electrically connected to the © Robert York, UCSB 2006

21

ECE 2C

Circuits, Devices, and Systems

Lab 2

outer metal casing. The microphone will not work if hooked up incorrectly because of the internal JFET buffer

Ground pin

Electret condenser microphone cartridge. If yours does not have mounting pins installed, solder some wire jumpers as shown in the lower figure. Take care not to solder the two connections together! ‰

Now test the microphone circuit by observing the output signal on the oscilloscope. Try whistling into the microphone⎯you should be able to produce a pretty nice sinewave. Record observations in the Lab RECORD. Note that when you stop talking into the microphone there is significant electrical noise in the system, appearing as a random voltage fluctuation at the output. Try to quantify the amplitude. Such noise is always present to a certain degree in any electronic circuit, and represents a lower limit on the sensitivity or fidelity of the signal transmission unless steps are taken to minimize it.

‰

Add in the remaining circuit elements according to the schematic. This last step just adds a simple inverting stage that can amplify or attenuate the signal. This will be helpful later to adjust the microphone signal level for a desired amplitude to drive the AM modulator.

Step 6 – Putting it All Together The last step is to hook up all both circuit modules to make a complete audio system that detects/receives an acoustic signal, electronically amplifies it and transmits it over a cable, and converts it back to an acoustic signal. This is obviously the basis of most telephone, intercom, and public address systems. ‰

Connect the microphone circuit output to the unused “aux” input on your amplifier circuit. Demonstrate the working circuit to the TA and answer the remaining questions in the LAB RECORD. For this step, C2 should be removed from the circuit, and the level-adjust potentiometer should be set to the middle of its range. It also helps to keep the microphone and audio amp circuits as far apart as possible, try using a long cable here:

© Robert York, UCSB 2006

22

ECE 2C

Circuits, Devices, and Systems

Bench Power Supply

‰

Use a long cable if possible

Speaker Batteryoperated Amplifier Circuit

Microphone Circuit (on breadboard)

+/-12 V

Lab 2

Since the electret condenser microphone is omni-directional, it may prove difficult to operate the system under full volume conditions. One issue that can exacerbate these unwanted feedback effects is that some cheap microphone cartridges have a little more sensitivity at the high frequencies, such as shown here. This is where an extra pole in the response may help. Try adding C2 into your circuit and see if that improves the system’s ability to operate under high-volume conditions with a voice signal. If you built the tonecontrol circuit, that can also help in a similar fashion. A potential problem

Typical sensitivity response for inexpensive electret microphones ‰

As a last step, remove C2 and use the potentiometer on your microphone circuit to adjust the output level to around 100mV max. amplitude under normal speaking conditions.

Congratulations: you now have a useful microphone circuit!

Possible Improvements In contrast to the audio amplifier, potential improvements to the microphone circuit are more subtle, and largely involve the microphone itself, not the electrical circuit. The little omnidirectional electret component that we choose is very convenient because it is small, and cheap, but do we really want an omni-directional mic in our system? Probably not, since people are usually speaking directly into the mic, hence the omni-directionality just creates problems by picking up unwanted noise from the surroundings, or undesired feedback from the distant speaker. There are some electrets on the market that have a © Robert York, UCSB 2006

23

ECE 2C

Circuits, Devices, and Systems

Lab 2

more unidirectional “cardoid” reception pattern, but these are quite expensive. One fairly simple way to make our mic directional is to put a little cone around it, which can be fashioned out of some heavy paper as shown below.

Condenser microphones are very sensitive devices, and will generate a signal in response to any nearby air movement. In a “windy” environment, such signals can easily swamp out the desired audio signal. So another useful improvement is to put some kind of a “windscreen” in front of the microphone. The windscreen is, in effect, a mechanical highpass filter, and helps improve the dynamic range of the microphone. Even with the windscreen, microphones can still be sensitive to bursts of air associated with the actual voice signal, such as “p” sounds. These effects can be reduced with a so-called “popfilter”, which is a large windscreen some distance away from the microphone. You will often see such things on microphones used in professional recording studios.

Windscreen

Pop filter

Another potential improvement to minimize noise pickup is to use a balanced amplifier configuration instead of the single-ended system we used. One example is described in http://www.eclectic-web.co.uk/index.php?jump=mike/electret_a.htm .

© Robert York, UCSB 2006

24

ECE 2C

Circuits, Devices, and Systems

Lab 2

A typical vocal signal is comprised mostly of frequencies in the 300-3000 Hz range. Therefore, any frequencies outside of this range are unnecessary in our system and hence represent unwanted noise. Thus one improvement might be to include a bandpass “speech filter” in the system to pass only frequencies in the vocal range. One possible 2nd-order active bandpass speech filter schematic is shown below, comprised of cascaded low-pass and high-pass filters. This is not the only way to make a bandpass filter, but it is an easy one to understand. In fact we will use similar filters in the design of our demodulator section in Lab #3: R7 & R1 Can be adjusted to shape the passband response R8 47k

R4 47k

R7 33k B

+12V

+12V C1 In 0.068uF

C4 .068uF

+ R5 10k

1/2 LF353

R1 33k

R2 10k

R3 10k

+

C2 .0047uF

-12V

1/2 LF353

Out

-12V

C3 0.0047uF

R6 10k

C2 & C3 (or R2 and R3) determine upper cutoff

C1 & C4 (or R5 and R6) determine lower cutoff

Speech Filter schematic (bias decoupling capacitors not shown)

© Robert York, UCSB 2006

A

25

ECE 2C

Circuits, Devices, and Systems

Lab 2

Lab 2 Record Group #_________________

Names: ______________________________ ______________________________

Step 1 – Equivalent Circuit modeling for the Speaker Record the input resistance at 1kHz: ____________ Ohm Record the coil inductance at 1kHz: ____________ H Record the resonant frequency: ____________ Hz Record the | Z | / 2 bandwidth: ____________ Hz Calculate the Q-factor of the resonance

Q: ____________

Calculate the equivalent circuit parameters for the parallel resonance: Rm: ____________ F

Cm: ____________ F

Lm: ____________ F

Steps 2-3 – Audio Amplifier on Breadboard What is the low-frequency cutoff of the amplifier: ____________ Hz

Record lowest frequency audible tone: ____________ Hz Record Highest frequency audible tone: __________ kHz

© Robert York, UCSB 2006

26

ECE 2C

Circuits, Devices, and Systems

Lab 2

Are the limits above related to limits on your hearing or are they due to the limited frequency response of the speaker itself?

Why is the output waveform clipped for large input signals?

Record input waveform at pin 3 with a ±1V sinewave at 1kHz and the speaker disconnected:

Average current drawn by audio amplifier under full-volume conditions with a ±0.025 V 1kHz sinewave input: ________________ mA Average current drawn by audio amplifier under full-volume conditions with a ±0.25 V 1kHz sinewave input: ________________ mA Average current drawn by audio amplifier when playing music audio under normal listening conditions :

________________ mA

From your answer above, estimate the battery life for this circuit assuming a 500 mA-hr battery rating:

________________ hrs

© Robert York, UCSB 2006

27

ECE 2C

Circuits, Devices, and Systems

Lab 2

Step 5 – Microphone preamp circuit Low frequency 3dB cutoff:

Predicted: __________ Hz

High-frequency cutoff (w/C2): Predicted: _________ kHz Measured Mid-band Gain:

Measured: __________ Hz Measured: __________ kHz

____________

Typical output signal amplitude for a whistling tone in mind-band at 6” away from mic: Approximate output noise amplitude:

____________ V

____________ V

Step 6 – Mini Intercom System Do you experience difficulty in operating the system under full-volume conditions? If so, Why?

What steps helped improve the operation of the system, if any?

TA Certification: _______________________

© Robert York, UCSB 2006

28

Date: ______________