ENGR 2031 – Digital Design Laboratory Lab 3 Background What we will cover Oscilloscope Measurements: Period, Frequency, Duty Cycle, Rise and fall times Propagation Delay Lab 3 Oscilloscope Horizontal axis of the oscilloscope screen is time and the vertical axis of the oscilloscope screen is voltage. Modern oscilloscopes can take many measurements automatically, but make sure you can explain how oscilloscope measurements are made manually. Figure 1a, Oscilloscope Measurements Figure 1b, Oscilloscope Oscilloscope can sweep the same waveform over and over or capture single instance of waveform (transient). Measurements The period T of a signal can be obtained directly from the oscilloscope screen. Figure 2, Measurements, Period, Frequency, Duty Cycle The frequency f of signal is just the inverse of the period, f 1 T . The positive duty cycle of a square pulse is the percentage of time a signal is high and is found from the time high divided by the period. The negative duty cycle of a square pulse is percentage of time a signal is low and is found from the time low divided by the period. Figure 3, Measurements, Rise and Fall Times The rise time is the time it takes for a signal to get from 10% to 90% of its value. The fall time is the time it takes for a signal to get from 90% to 10% of its value. Propagation Delay Propagation delay is the time from when the input of a device changes to when output of the device changes (50% to 50%). Figure 4, HC and HCU Transition Times and Propagation Delay Times, Combinational Logic, From Texas Instruments Data Sheet The propagation delay of a circuit is the time from when the input of the circuit changes to when the output of the circuit changes. We typically are interested in the worst case propagation delay; the delay along the longest combinational delay path from input to output. Consider determining the worst case propagation delay for the circuit of Figure 5 that realizes the MSOP Y A B C D A C D Using the data sheet tables (like in Figure 6) and assuming Vcc = 4.5 and the operating temperature range – 55 to 125 degrees Celsius, we obtain the worst case propagation delays for each device in the circuit. 74HC00 (2-input NAND): 27ns 74HC04 (inverter): 26ns 74HC20 (4-input NAND): 30ns The maximum propagation delay (input to output) is for A to Y (A – NOT – NAND4 – bOR4 – Y), Maximum Propagation Delay = 26ns + 30ns + 30ns = 86ns Figure 5, Circuit for Propagation Delay Example Figure 6, Switching Specs for 74HC04, From Texas Instruments Data Sheet Input Combinations for Measuring Propagation Delay The input changes corresponding to the maximum delay path can be measured by setting all the values except the input corresponding to the maximum delay path to set values, then changing the input corresponding to the maximum delay path, and measuring how long it takes from the input changing to the output changing. To determine what the values other than the input corresponding to the maximum delay path we need to identify two rows in the truth table with different outputs and whose inputs differ in only one bit position, that of the signal corresponding to the maximum delay path. It is not as simply as just identify two rows that differ in the one bit position because if the input has more than one path to the input we need to ensure the rows correspond to the maximum delay path. The truth table for the circuit of Figure 5 is given in Figure 7. The rows corresponding to the maximum delay path are shaded. The process for identifying these rows is as follows: First apply propagating values at all inputs of gates except the A input (A is the signal to be propagated in this example). These values are shown in Figure 8a. Propagating values pass a signal or its complement to the output of the device and controlling values fix the value of the output of the device. For NAND and bOR gates, 0s on any input fixes the output at 1 and 1s on all the inputs except one propagates the complement of the other input line (1s are propagating values for NAND and bOR gates and 0s are controlling values). A B C D Y 0 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0 1 1 1 0 1 0 0 1 0 1 0 1 0 0 1 1 0 0 0 1 1 1 0 1 0 0 0 1 1 0 0 1 0 1 0 1 0 0 1 0 1 1 0 1 1 0 0 1 1 1 0 1 1 1 1 1 0 1 1 1 1 1 1 Figure 7, Truth Table for Maximum Delay Path Example Figure 8a, Propagating Values for V Input After applying the propagating values to propagate A, two of the input values for the maximum delay path are known. C and D must both be 1 to propagate A through the NAND4. Next, determine any remaining input values by assigning the appropriate propagating or controlling value to get the correct signal values set previously. With C = 1 and D = 1, the inputs of the second NAND2 are both 0s which will give a 1 output which matches the propagating value set previously. To get a 1 as the output of the first NAND2, we need at least one of its inputs to be a 0 and since we cannot set A, the propagating signal to 0, B must equal 0. The result of this is given in Figure 8b. The input values other than A corresponding to the maximum delay path are B = 0, C = 1, and D = 1. This corresponds to rows three (0011) and eleven (1011) in the truth table. Figure 8b, Propagating and Controlling Values to Propagate V to Output Along Maximum Delay Path Lab 3 Watch oscilloscope video (handouts page of web site). The video covers an oscilloscope that is similar to the one we will use so there will be some things that will be done slightly differently but the major concepts are the same. Read the oscope pointers and pages 1-39 of the oscilloscope manual before lab. Both are posted on the handouts page of the web site. You will need the circuit constructed for lab 2 along with the annotated schematic for lab 3. You will work with the same lab partner as you did for lab 2. You will need the data sheets used in Lab 2 to look up the propagation delays for Lab 3. These can be found in the section called AC Characteristics or Switching Characteristics. “Max” values are worst-case values, and you should always design for the worst case. For lab step 4 and 5, use male to female jumper wires (long green jumper wires) to connect to the oscilloscope probe to the DE2 header pins and rather than directly connecting the probe to the header pins. This will keep one from inadvertently shorting the DE2 header pins and will make it easier to connect the oscilloscope probe to the header pins. For lab step 7, measuring period and positive duty cycle. Make sure to adjust the horizontal and vertical scale so that 1.5 to 2 periods of the signal are shown on the screen. For lab step 14, measuring rise time, fall time, and propagation delays. Make sure that you first verify what the steady state high and low values for the signal are. You can then zoom in as much as needed to obtain an accurate measurement of the time.