ELEC 242 Lab

Experiment 8.1

Static Performance

Equipment

Components

Part 1: Calibration

The output of the summing amplifier is
Rearranging this we have that the system is in balance (i.e. no correction is being applied) when Because this is a rather awkward relationship (and because due to variations in component values it isn't very accurate anyway) we have labeled the "Desired Position" slider on the controller in terms of the value of we are trying to achieve. Since we are going to be measuring just how accurately we achieve it, it will be necessary to calibrate more accurately the mapping from the slider value to the voltage actually generated.


Step 1:

 Set up the position control system as it was in Part 2 of Experiment 7.3. Verify that it still works. Be sure to stop the controller program.

Step 2:

 Connect A/D input 3 (pin 4 on the interface board socket strip) to (the output of the summing amplifier).

Step 3:

 Load and run the "Calibrate" Labview program. This will find the actual range of and the actual values of required at each end of the range.

Step 4:

 When the Calibrate program has finished, write down the values for Vdrive hi, Vdrive lo, and Vact lo.

Step 5:

 Load the "Controller 3" Labview program. This is just like the "Controller 2" program, but it allows for calibration, and also displays the error.

Step 6:

 Enter the values for Vdrive hi, Vdrive lo, and Vact lo into the the appropriate fields on the control panel.

Step 7:

 Run the program and verify its operation.

Part 2: Reduction of Hysteresis



Step 1:

 With the power to the motor turned off, put the hook of the spring balance through the hook on the cord wrapped around the drum.

Step 2:

 Slowly pull upward on the spring balance until the drum starts to turn. The highest value of force read on the balance before the drum turns is the static Coulomb friction, the amount of force which must be applied before any motion can take place. Record this value in your lab notebook.

Step 3:

 Turn the power back on and restart the "Controller 3" program.

Step 4:

 Set the Desired Position control to -2.5.

Step 5:

 Manually turn the spool until the Error indicator reads zero.

Step 6:

 Slowly raise the Desired Position until the spool moves. Note the value of the Error and Vdif just before it moves.

Step 7:

 Slowly lower the Desired Position control until the spool moves again. Again, note the values of Error and Vdif just before the motion takes place. The difference between the two values of error is the hysteresis in the mapping of desired to actual position.

Step 8:

 Replace (the 220k feedback resistor in the summing amplifier) with values of 470k, 1M, 2.2M and 4.7M. For each value, repeat the measurement of hysteresis.

Step 9:

 Plot the hysteresis vs. .

Question 1:

 Explain the behavior of as the gain is changed.

Part 3: Disturbance Rejection

Because of the hysteresis caused by friction, it is difficult to measure the disturbance rejection with the system at rest. If we apply a load smaller than the frictional force, nothing happens. When we exceed the threshold, it suddenly lurches forward, as we saw above.

One way around this is to measure the average error with the system in motion. If we do this both going up and going down, the effects of friction should cancel, leaving us with the error caused by the weight of the load.


Step 1:

 Load the "Ramp Response" Labview program. This will run the hook down and back up at a constant rate, while measuring the difference between the commanded position and the actual position.

Step 2:

 Enter the calibration values for Vdrive hi, Vdrive lo, and Vact lo into the the appropriate fields on the control panel.

Step 3:

 We will be measuring this error both as a function of the load and as a function of the loop gain, as determined by . Make two tables, one for average error and one for peak error. Allow for weights of 0, 1, 2, 3, and 4 oz and for values of of 220k, 470k, 1M, 2.2M, and 4.7M.

Step 4:

 With set to 220k and no weights on the hook, run the program. If all is well, the hook will lurch to the top, slowly run down, then slowly run back up. When this is finished, the program will display a plot of the actual trajectory, the error as a function of time, and two numbers: the average error and the maximum error. Record these values in your tables.

Step 5:

 Repeat the measurement with 1, 2, 3, and 4 ounce weights on the hook. Also make printouts of several representative plots for your notebook. With the 4 oz. weight it is likely that nothing will happen. With the 3 oz. weight you may get erratic performance. If so, try several times and record the best measurement.

Step 6:

 Repeat the previous two steps with values of of 470k, 1M, 2.2M, and 4.7M.

Step 7:

 Make a plot showing the average error vs. load for each value of . Make another showing the average error vs. for each value of load.

Question 2:

 What is the expected behavior of the error in position as a function of the weight of the load and of the controller gain A ? Do your measurements support this expectation?