ELEC 242 Lab

Experiment 4.3

Motor Amplifier

Equipment

Components

The previous Experiment examined the inverting op-amp configurtion. We'll now look at the voltage follower, an example of the other major class of op-amp circuits: the non-inverting configurtion. More significantly, we will boost the output drive capability with an emitter follower buffer so that we can drive low resistance loads, especially the DC motor, without distortion.

We will use this circuit in several subsequent labs, so devote some care to building it. In particular, make it reasonably compact and place it on the breadboard where it won't interfere with building other circuits. Here is a suggested layout for the final version of the circuit (click for a larger view):

However you decide to lay out your circuit, use short wires. Your circuit should be 2 dimensional, not 3 dimensional.

Part 1: Voltage Follower



Step 1:

 With the Ohms function of the DMM, measure the armature resistance $R_A$ of the motor. Because of the commutator, this will vary with the shaft angle, so rotate the shaft and take the average of several consistent readings.

Remark:

 To simplify testing we will use a resistor as the load for our sequence of circuits until the final step. Last week we courted disaster (and burnt fingers) by exceeding the power ratings of our transistor and load resistor. This week we will be a little more careful, both by using components with higher ratings, and by calculating levels to be sure that we stay within them. All the resistors in our parts kit are 1/4 W, but we can make a load with higher power rating by connecting several of these in series or parallel.

Step 2:

 Make a 50 ohm "power resistor" by connecting the 5 10 ohm resistors in your kit in series. Connect one end of this composite resistor to ground.

Question 2:

 Show that this composite resistor will dissipate 1.25 W without exeeding the power rating of any of the individual resistances.

Step 3:

 Set the function generator to produce a 100 Hz, 1 V p-p triangle wave.

Step 4:

 Connect the function generator output to the load resistor. This should have the same effect as in Experiment 3.3, i.e. the function generator output voltage should be reduced by half due to the loading.

Step 5:

 Wire the following circuit
\includegraphics[scale=0.650000]{ckt5.1.ps}


Step 6:

 Connect $v_{in}$ to the function generator output and $v_{out}$ to the load resistor.

Step 7:

 Observe $v_{out}$ with the scope. It should have an amplitude of 1 V p-p, i.e. there should be no attenuation due to loading.

Step 8:

 Increase $v_{in}$ until $v_{out}$ begins to clip. Note the value of $v_{out}$ at which clipping occurs.

Step 9:

 Remove the load resistor and continue to increase $v_{in}$ . Does any clipping occur at the maximum output available from the function generator?

Remark:

 From the spec sheet, $R_{out}$ of the op-amp (open-loop) is about 75$\Omega$ . Yet we can drive a 50$\Omega$ load with no significant change in amplitude. In addition to reducing gain, negative feedback also reduces output resistance. So our inverting amplifier circuit, with feedback, has a very low output resistance, as long as everyting is linear. What's happening when trying to drive a low load resistance to large output amplitudes is that something nonlinear is happening: current limiting. The op-amp contains circuitry to prevent itself from being destroyed by trying to deliver too much power. When the output current reaches a limiting value (the short-circuit current) the op-amp stops increasing the output voltage. This causes distortion of the output waveform, but keeps you from having to replace the op-amp.

Question 3:

 Based on your measurements of the clipping level with a 50$\Omega$ load, what is the short-circuit current limit of your op-amp? How does this compare with the value in the spec sheet?

Part 2: Voltage Follower with Emitter Follower Output

Unfortunately, the current the op amp is capable of delivering is far too low to drive our motor with the torque we will require from it. We can fix this problem by using an emitter follower as a current amplifier to increase the amount of current we can deliver to a load.

Since we're going to be delivering a lot of power to the motor, and since our motor amplifier is not 100% efficient, it's going to be dissipating a lot of power. To keep from burning it (or our fingers) up, or melting the breadboard, we're going to have to get rid of the resulting heat more efficiently.

The way we will get this increased heat dissipation is by adding heat sinks to the transistors. These are simply pieces of metal (aluminum) having high thermal conductivity and large surface area which carry the heat to a greater volume of ambient air than the small area of the transistor case can. We will also use heat sink compound to improve the heat transfer from the transistor to the heat sink.


Step 1:

 Find the TIP31C (NPN) and TIP32C (PNP) power transistors in your parts kit. The leads are identified below. Note that their order is different from the 2N3904 you used last week.
\includegraphics[scale=1.000000]{scanl8b.ps}


Step 2:

 Get two heat sinks and clips.

Step 3:

 Remove the two tabs from the bottom of each heat sink. The easiest way to do this is to use your pliers and peel them back like the lid of a sardine can.


Step 4:

 Put a dab of heat sink compound on the back of each transistor.

Caution

Try not to get any of this stuff on your clothes. It's almost impossible to get out.


Step 5:

 Clip each transistor to a heat sink. Since the clip may cover up the markings on the transistor, identify each transistor (or heatsink) so you can tell them apart.


Step 6:

 Before plugging the power transistors into the breadboard, use your pliers to twist each lead 90 degrees. This will place the narrow, smooth face of the lead (rather than the wider, rough edge) between the fingers of the breadboard connector, making it easier to insert, and making the breadboard last longer.

Step 7:

 Also get two 1N4148 diodes.

Caution

There are two types of diodes in your parts kit with identical glass packages: the 1N4148 signal diodes and the 1N5230 Zener diodes. They are identified by (very small) printing on the case. Be sure to get the correct diodes.

The diodes are a bit of magic to prevent voltage transients caused by the motor's inductance (which we haven't studied yet) from destroying the transistors.

Step 8:

 Add these components to your circuit as shown below. Note that the emitter terminal is the one with the arrow, regardless of whether it is on the top or the bottom.

\includegraphics[scale=0.650000]{ckt5.3.ps}


Step 9:

 Set the function generator output to 3 V p-p. Connect the function generator input to $v_{in}$ and the load resistor to $v_{out}$ .

Step 10:

 Sketch the output waveform, $v_{out}$ .

Step 11:

 Increase the function generator AMPLITUDE control to its maximum value. Is there any clipping?

Part 3: Dead Zone Reduction



Step 1:

 Move the feedback from around the op-amp to around the combination of the op-amp and the emitter follower buffer. Also add the capacitor as shown. This is another bit of magic, this time to try to control oscillations.
\includegraphics[scale=0.650000]{ckt5.5.ps}


Step 2:

 Connect $v_{in}$ to the function generator and $v_{out}$ to the load resistor, as in the previous part.

Step 3:

 With $v_{in}$ a 3 V p-p triangle wave, sketch $v_{out}$ .

Step 4:

 Increase the function generator AMPLITUDE control to its maximum value. Is there any observable distortion in $v_{out}$ ?

Part 4: Driving the Motor

Now for the moment of truth: Can we use this circuit to drive the motor and achieve an improvement in performance?


Step 1:

 Disconnect the 50 ohm load resistor.

Step 2:

 Wire the tip of J1-6 (pin 8 on the interface board socket strip) to $v_{out}$ . Wire the ring (pin 9) to ground.

Step 3:

 Set the function generator to produce a 10 Hz sine wave.

Step 4:

 Press the strobe test disk onto the shaft of the motor.

Step 5:

 Plug a phone plug patch cord into J1-6 of the interface board. Plug the motor into the 3-pin connector on the other end. Be sure that the red wire on the motor lines up with the white mark on the cable connector.

Step 6:

 Turn on the power and adjust the function generator AMPLITUDE control until the oscillation of the test disk is about 180 degrees.

Step 7:

 Disconnect the ungrounded side of the motor from $v_{out}$ and connect it directly to $v_{in}$ (i.e. the function generator output). What is the approximate angle of oscillation?

Step 8:

 Connect the oscilloscope to the function generator and motor so that you can measure the voltage across the motor.

Step 9:

 Set the function generator to produce a full scale 4 Hz square wave. Sketch the waveform of the voltage that actually appears across the motor.

Step 10:

 Without changing the function generator settings, reconnect the motor to $v_{out}$ . Sketch the waveform that now appears across the motor.

Step 11:

 Don't disassemble this circuit, as we will use it in subsequent Labs.