Lab 2 - ECE 420L

Authored by: Justin Le

February 6, 2015

Email: lej6@unlv.nevada.edu

 

Pre-Lab

 

Review the video lecture and notes on scope probes.

Vary the parameters in the simulation from the lecture to ensure understanding of the circuit.

Review the operation and analysis of simple RC circuits and Bode plots.

 

 

Experiment 1

 

Shown in order are waveforms measured by a 10:1 probe undercompensated, overcompensated, and compensated correctly.
 


Figure 1a.


Figure 1b.


Figure 1c.
 
On the oscilloscope used for this experiment, the probe type is set on the “Channel” menu by selecting the attenuation factor.
 
The schematic of a 10:1 scope probe is shown in Figure 1d. Its input resistance is shown to be 9 M ohm to maintain the 10:1 voltage divider with the scope, whose input resistance is 1 Mohm. Similarly, the input capacitance of the probe is chosen to be 12 pF, or one-ninth of the combined parallel capacitance of the cable and the scope.
 
The calculation in Figure 1e shows that the input voltage of the scope is indeed one-tenth of the voltage at the probe tip.
 


Figure 1d.


Figure 1e.
 
 

Experiment 2

 
To measure the capacitance of a length of cable, the cable can be used as the capacitor in a series RC circuit and the rise time of its output measured. The time at which its output reaches 0.5 of the input should be approximately 0.7 * RC.
 
For this experiment, an input square wave of 1 V and a series resistance of 1 M ohm were used. The halfway rise time was measured to be 104 micros, as shown in Figure 2a. The calculation in Figure 2b shows that the capacitance of the cable is 149 pF, which approximates the capacitance of 128 pF obtained by a capacitance meter.
 


Figure 2a.

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2b.
 
  

Experiment 3

 

To demonstrate the difference between compensated and uncompensated probes, a  0 to 1 V pulse at 1 MHz was applied to a voltage divider consisting of two 100k resistors. Figures 3a and 3b show the output of the voltage divider when measured by a compensated scope probe and an uncompensated cable, respectively.
 
The capacitance provided by the compensated probe greatly reduces the total capacitance seen at the scope input. Thus, the probe output increases quickly enough to be observed on the scope, even at a high frequency, as shown in Figure 3a. In contrast, the uncompensated cable causes a relatively large capacitance to appear at the scope input. Thus, changes in the cable’s output are negligible on the scope, as shown in Figure 3b. The cable’s time constant is much greater than that of the compensated probe.
 


Figure 3a.


Figure 3b.

 

  

A Final Remark
 
In order to directly measure a circuit on a PCB using the uncompensated cable, a test point must be implemented to prevent the loading effects seen in Experiment 3. The test point would consist of a large resistance in parallel with a small capacitance and connected in series with the cable. The test point imitates the circuitry inside the tip of a compensated probe, as seen in Figure 1d. The resistance and capacitance are thus chosen as noted in Experiment 1 to maintain a desired ratio between the cable’s input and output voltages.

 
 

Figures

 

For Experiment 1:

    abc:  Laboratory results.

    d:    Schematic.

    e:    Calculation.

For Experiment 2:

    a:    Laboratory result.

    b:    Calculation.

For Experiment 3:

    ab:   Laboratory results.

 

 

Click to view all labs.