ME w144L: Dynamic Systems & Controls (DSC) Laboratory
Course Log for Summer 2016

Last updated 5/30/16 - NOTE: any content in gray regions are planned/projected topics
Lecture meets on TH 1-2pm in ETC 7.146 | Labs meet Tuesday in ETC 4.160, 10-12 and 1-3 pm
SuMTuWThFSa
29303112
Course Intro & Lab 1 lecture
(ETC 7.146, 1 pm)
3

4
56
7
Lab 1: Intro Lab
(ETC 4.160 at lab time)
8

9
Modeling and simulation of compound pendulum

Pre-Lab 2 - Tutorial
1011
1213
14
Lab 2: Simulation with LabVIEW CD&S

15
16
Two-can system modeling and simlation
Pre-Lab 3
1718
192021
Lab 3: Two-can system lab
22
23
Two-can continues
(updated)
2425
262728
Lab 3: (cont.)
29
30
No lecture - complete Lab 3 LE
Jul 1 2
34

5
Lab assessment - assess hands-on, including simulation
6

7
Vibration,  Accelerometers, and two-story system
8

9
1011

12
Lab 4: Vibration and acceleration measurement
13

14
LV vision meas and analog meter model

1516
1718

19
Lab 5: LV vision measurement and meter modeling
20
21
Frequency response and system identification
2223
2425

26
Lab 6: Freq response  measurement and system ID for meter using vision

27 - Reading day for 9 week classes; Final exams for 9 week

28
Feedback control

29

30
311

2
Lab 7: Feedback control using vision

34
No planned lecture


56
78
9
Hands-on quiz review and practice
10

11 - Last class day for whole session courses1213
*
14*15 - Monday
Lab Final (hands-on, in ETC 4.160)
--

Lab 1: Introduction to DSC Lab (Week 1-2)

A.  Lab Lecture and Information (6/2/2016) - Thursday, 1 pm, in ETC 7.146 - attend for course/lab introduction
B. Pre-Lab 
  • Future labs will have pre-lab questions here that will be due at lab time.  These pre-labs should be completed in your laboratory notebook unless otherwise specified.
  • In this first lab, you will be experiment with a compound pendulum, experiment with using an angular potentiometer as a sensor to measure pendulum angular position, and learn about how LabVIEW programs (virtual instruments, or VIs) are written for data acquisition (DAQ) and signa capture, processing, and storage.  Complete the following pre-lab preparation.  
  1. It should be clear to you why we distinguish between a compound pendulum and a simple pendulum.  The physical model for each will have been briefly discussed in lecture and will covered in detail next week.  Do a quick self-study before lab on these topics and summarize your understanding. (15 min)
  2. Consider the compound pendulum setup we will be using in the lab.  If you were asked to come up with another angular sensing method besides the angular potentiometer, what would you recommend?  You might conduct some online research on ways to measure angular rotation.  Compare this alternative to the potentiometer using any metrics you deem relevant (e.g., cost, power requirements, output signal type, resolution in measuring angle, etc.). (15 min)
  3. It is expected that you have been previously introduced to LabVIEW, but may not have done extensive programming.  You will program in LabVIEW in this course.  It can be helpful to skim and be familiar with content in Chapters 1, 2 and 3 of the Getting Started with LabVIEW (2013) tutorial.  These exercises introduce basic LabVIEW programming concepts and terminology.  You might also watch these online tutorial videos that introduce LabVIEW.  You do not have to watch them all, but you might revisit these after next week's lab to get more guidance and tips on programming in LabVIEW. (30 min) - nothing to submit
C. Lab Work (6/9/15, meet in ETC 4.160 at assigned lab time)
  • TA may briefly review pre-labs 1 & 2
  • Introduce DSC Lab
  • Guided exercises. The TA will provide guided laboratory exercises for the following:
    1. Measure pendulum geometry
    2. Introduce the NI myDAQ device, show how it is connected via USB, demonstrate connections required for measurement, and provide guidance for basic usage.  
    3. Guide you through setting up the pendulum experiment and calibrating the angular potentiometer.
    1. Develop a VI for measuring the potentiometer circuit output voltage
    2. Complete the calibration of the potentiometer (angle in degrees vs. voltage) for use in these experiments
    3. Include code in your VI that saves captured signals to a measurement file
    4. Capture and save at least three different signals with initial conditions for the free response of the pendulum or as specified by the TA.  These data files will be used for the LE and for next week's lab.  One or more of these experiments should be used to estimate the pendulum moment of inertia.
D. Lab Evaluationsubmit a specified by TA
  • LE for this week will be composed of specific notes and or responses to questions that the TA will dictate during the lab.  The lab notebook recording of these notes/responses will be graded. 
----------------------------------------------------------------------------------------------------------------
------------------------All content below this line is still in progress------------------------------
-
---------------------------------------------------------------------------------------------------------------
Lab 2 - Modeling and Simulation of Compound Pendulum Using LabVIEW (Week 3)

A. Lab Lecture and Information (6/11/15) - no physical meeting - review slides/videos below
B. Pre-Lab 2 (Item #2 via email; Items #1 and #3 hardcopy at beginning of your next lab; no abstract)
  1. To prepare for lab, complete the following tutorial introduction to LabVIEW Control Design and Simulation.
  2. NOTE: if you want to install LabVIEW on your own computer, please see these instructions.
  3. The following videos provide a hands-on introduction.  It is necessary to have access to LabVIEW and to have the LabVIEW Control Design and Simulation Module installed as well.
    1. Simple Pendulum Simulation Tutorial (1) - State space form (4:58)
    2. Simple Pendulum Simulation Tutorial (2) - Creating the LV CD&S simulation (13:38)
    3. Simple Pendulum Simulation Tutorial (3) - Setting up simulation parameters (13:23)
    4. Simple Pendulum Simulation Tutorial (4) - Running test cases (5:52)
  4. Submit: The TA will require that the simple pendulum simulation VI you complete in this tutorial is ready for use at the beginning of Lab 2.  No other pre-lab is required.
C. Lab 2 (meet at your scheduled lab time in ETC 4.160)
Note: The number of significant digits on the exported data will depend on the display format for the X or time) and Y (amplitude) axes.  Right-click on the graph and select 'Properties' from the pop-up menu - you should see this menu.  Make sure each axis uses a suitable notation and outputs enough significant digits for your analysis.  For example, using floating point with %0.1f gives only 1 digit, which is not good enough.  Safe to use '%g' or '%#g', the latter removes trailing zeros.
Here is a reference: LabVIEW format specifiers
D. Lab Evaluation 2 - may be adjusted by TA
  1. Submit a summary of results from using your simulation VI to determine the level of linear damping and Coulomb friction required so your model results match the experimental measurements. Report the values for these coefficients that give the best results.  Provide a screen shot of results from plotting the simulated and measured results for the 'best friction  model'.  Make sure to summarize all parameters used, including mass moment of inertia, mass, and length to CG for the compound pendulum. (1 page of screen shot/graph + table of data)
  2. Submit an explanation of how you approached the problem of determining the dominant form of friction in the pendulum system.  What process did you follow that you found successful in getting the best match between the measured and simulated pendulum response?  How did you decide that the model was sufficiently valid? ('how' refers to what criteria you used, what factors you used to make your decisions about validity, etc.)  Are there any shortcomings in the models that you would address if you had time?  (1 page)
  3. OPTIONAL (bonus points at discretion of TA) A pendulum's period of oscillation depends on the amplitude of oscillation.  This means that your measurements of the pendulum angle decaying from, say, 90 degrees down to 10 degrees and lower might show that the period is actually changing.  Analyze your data and determine if it is possible to measure and estimate how period changes with amplitude for this pendulum system.  Submit an explanation of your data analysis and a summary and explanation of your results. (1 to 2 pages)

----------------------------------------------------------------------------------------------------------------

Lab 3 - Two-Can System Modeling, Sensing, and Simulation (Weeks 4 and 5)

NOTICE:  UPDATES MADE FOR WEEK 2
  • This is a two-week lab.  In the first week of this two-week lab, you will should become familiar with the lab setup, get guidance in setting up to collect data, get your simulations and (possibly) DAQ VIs running, take preliminary data, test out your experimental protocol, etc.  You should be prepared to return in the second week and work independently to complete collection of all your data and run demonstrations as described below.
  • There is a pre-lab in the first week, but no LE.  In the second week there is no pre-lab, only the LE.
  • The LE consists of two "in-lab" demonstrations in the second week before you leave lab.  There is also a written part of the LE as usual.

A.  Lab Lecture and Information (1st week) - updated 6/25/14 based on 1st week progress
B. Pre-Lab 3 - note: hardcopy submission due at your next lab meeting; no abstract required
  1. Two-can system LabVIEW CD&S simulation.  Develop a two-can simulation model in LabVIEW ready for use in lab.  The only parameter inputs required are K1 (bottom can) and K2 (top can), in addition to initial conditions for each can. 
    1. First, here are some tutorial videos to help you build the simulations: a) building a one-can simulation (5:29), and (b) building a two-can simulation (10:57)
    2. Here is a table of can dimensions.  Use Can Set 1 from this table to calculate ideal values for K1 (bottom can) and K2 (top can).  You will need to refer to the slides appendix for details on calculating ideal flow coefficients.  Record your calculations as part of this pre-lab in your working lab notebook.
    3. Submit your LabVIEW simulation VI via Canvas to your lab TA.  The program should be readable and neatly organized. 
C. Lab 3 Work (meeting in ETC 4.160 at assigned time)
  • Pre-lab must  be completed before scheduled lab meeting
    1. It is expected that you will have a working two-can simulation
    2. It is expected that you will have a plan for completing data collection, flow coefficient determination., and pressure sensor calibation (having reviewed lecture videos)  Use of pressure sensors in 2nd week
  • Guided Exercises (Week 1).  The TA will introduce the setup and provide initial guidance.
    1. Laboratory setup familiarization. The TA will review the lab setup, the two cans, available supplies and protocols, use of the pressure sensors, and configuration used for data collection (myDAQ connections).  Here is a list of available equipment supplies:
      1. Two-can set up
      2. Set of volume measuring plastic beakers (5 sizes per set) and syringes
      3. Stop watch (optional); may use personal time piece or phone
      4. Rulers/scales as needed
      5. (more may be added here)
    2. There may be some DAQ setup in first week - follow TA instructions if you have time to begin using the pressure sensors.  Otherwise, this may be done next week.
  • Student Exercises (Week 1).  The TA will assist only as needed
    1. Data Collection for K Determination.  For each can in your two-can setup, run at least three different emptying experiments and demonstrate the LabVIEW VI developed for estimating K values.  Take the mean value of the three trials for your K value for each can.  You should also collect data for the two-can system
    2. You may need to update your two-can simulation model using the parameter data  measured for your specific lab setup. 
    3. Test your model: The TA will provide a test initial volume for Can 1 (upper).  Use your simulation to estimate the time for a peak volume to be reached in Can 2 (lower), which should have a zero initial volume.  Compare your simulation results to those from a single experimental test using same initial condition.
D. Lab Evaluation - no lab evaluation for 1st week.  Review the following (updated 6/25/14)
    1. Typically it is expected that you will complete experiments on the 2 cans to collect data for estimating the K coefficients.
    2. The K coefficients should allow you to run your simulations and do some quick evaluations.  For example, you might use a pre-determined set of initial conditions in your simulation and predict what you will have seen in the lab in a physical test.
    3. The second week of lab will depend on what endpoint you reaches this first week.  If you did not complete the basic experiments for K value determination, then those must be completed in 2nd week.  If you completed those experiments, then you can continue in second week by using pressure sensors and developing a DAQ system.  Your TA will guide and grade accordingly.
    4. Complete any pre-lab preparation for the 2nd week of lab, especially as dictated by the TA.
----------------------------------------------------
A.  Lab Lecture and Information (2nd week)
  • The lecture slides are reviewed in the following videos:
    • If you completed the two-can study in the first week, you may use pressure sensors in the last part of this week's lab.  The following video introduces use of pressure sensors:
      Measurements and sensing
      (8:19)
      NOTE You should be prepared to confirm the calibration of the installed pressure sensor.
B. Pre-Lab 3 (2nd week) - as specified by the TA
B. Lab 3 Work (2nd week) (meeting in ETC 4.160 at assigned time)
  • Student Exercises (Week 2).  The TA will assist/guide as needed
    1. It is assumed that your lab group collected K values for the two-can system in 1st week.  If not, then those experiments should be completed before moving to next steps.
    2. Verify your model. Use experimentally measured K values in your two-can simulation, verify simulation by observing how well it predicts key measures (peak values, times to empty, etc.).
    3. Test your model - filling/spilling trial: The TA will provide a test initial volume for Can 1 (upper).  Use your simulation to estimate the time for a peak volume to be reached in Can 2 (lower), which should have a zero initial volume.  Compare your simulation results to those from a single experimental test using same initial condition.  Important: results from this step required for the LE (see below).
    1. Set up for measurement of water height in cans using pressure sensors.  TA must approve this step.
    2. See OPTIONAL MATERIAL below on pressure sensor usage and familiarity.
    3. Data Collection to verify Pressure Sensor Calibation.  The TA will advise on validity of your setup and recommend any calibration tests as needed.
    4. Collect time traces for one- and two-can experiments.
    5. Use time traces to propose improvements to your model of the two-can system
D. Lab Evaluation 3 - due date as specified by TA - submit your LE using required format in hardcopy form - abstract required
  1. Demonstrate in lab: Run a single experiment to validate whether the initial volume specified for Can 1 achieves the desired height (volume) in Can 2.
  2. Demonstrate in lab: Use your simulation model of the two-can system to determine the initial volume in Can 1 that will maximize the volume in Can 2 without spilling.
  3. Written submission: Submit a written LE that includes the following items
    1. abstract describing lab and results
    2. Compare your values for K_1 and K_2 as measured in the lab to those from the theoretical (ideal) calculations.  Explain any differences and whether they make sense given the assumptions made in the ideal model.
    3. Describe the results and any discrepancies in your experiments on the filling of Can 2.
    4. In a summary/discussion, explain any updates you would make or did make to your initial lab procedure and provide any insight gained on what you think would have improved your results.
    5. Summarize the final model you used and any improvements you were able to make.

OPTIONAL MATERIAL (if pressure sensors are used in week 2)

Pressure Sensor Usage and Familiarity
- This material is provided as reference for this lab.

The TA may provide a review of the pressure sensors, and configuration used for data collection (myDAQ connections).   The two-can setup makes use of two (2) PX409 pressure sensors (mounted, but not connected to DAQ).  To use these sensors, you need to:
  1. Verify Range. The specific pressure sensor used in lab is the Omega PX409-10WG5V.  Refer to the full manufacturer's specifications provided here.  The `10W' designates that this has a full scale pressure range of 10 inches of water, `G' means it reads gauge pressure, and the full scale output range is 0 to 5 V (calibration sheets for all pressure sensors in lab).  Assume that the ratio of output voltage to applied pressure follows a linear relation.  Estimate the peak voltage (i.e., full can) you'd expect to measure for each can in your lab setup, given that these sensors are mounted in the base of the can.  This value needs to fall within the allowable input voltage range for the myDAQ to be used in lab
  2. Understand Power Connections. The manufacturer's specifications for the PX409 indicate that for the 5V output the EXC power must be between 10 and 30 V and supply at least 10 mA.  Study the power supply specifications from the myDAQ manual (see page 43).  For this laboratory study, how much total power is needed by two PX409s, assuming you will use one of the myDAQ power supplies (specify which one)?  Explain your power requirement estimations.  Can you power both PX409s as required using the myDAQ, or would you need to find another power supply?
  3. Understand Signal Connections.  The sensor cable connections for the PX409 sensor are shown in the table below, in the column designated by 5/10V.  Use this diagram to sketch and explain (also on lecture slides) how you would connect the cables COMMON, +OUTPUT, and +EXC to the available myDAQ connections.  Note that `NC' means `no connection'. EXC is excitation.
Basic DAQ setup and calibration for pressure sensors
  1. NOTE: it is possible to damage the pressure sensors: if you make the wrong connections and burn out components, excessively pull cables, etc.  Pre-lab should have been completed to learn about proper cable connections, power requirements, etc.  Practice care in using the laboratory equipment which is shared by all the students in this course.  Your TA must approve your setup before you can make any connections.
  2. The TA will initiate and/or guide you in developing the DAQ VI for this lab.  
  3. You need a calibration VI that allows you to test your pressure sensor calibration while you are running your experiments for determining K coefficients.  
  4. Use your VI to make a quick calibraton of each pressure sensor so you can measure the volume in each can.  Record the calibration factors (in units of volume/voltage).  You can compare to the calibration sheets provided by the manufacturer.  Note that each pressure sensor has a unique serial number.
  5. Consider building a VI that outputs DC values of the pressure sensor (for static testing) but also plots volume vs. time.
  6. Once your have built your VI and initiated your experiments, the TA will only provide help as needed.

-
---------------------------------------------------------------------------------------------------------------
Lab 4 - Vibration modeling and accelerometer measurement (Week 7)
In this is one-week lab, you will focus on learning about vibration modeling and motion measurement using accelerometers.  First, a one-story model will Lab studied which follows closely the 2nd order system model concepts that will have been introduced in the ME 344 lecture.  By configuring the lab setup as a two-story building model, it is possible to experiment with methods that can be used to study higher order systems.  The lab will show how well theoretical eigenvalues calculated from the model can be used to predict measured natural frequencies of the two-story system.

A. Lab Lecture and Information - online only
  • Vibration modeling.  The following lecture videos refer to these lecture slides (pdf):
    1. Vibration modeling of a mass-spring-damper system (20:27 min)
    2. Description of  laboratory setups and laboratory objectives (7:08 min)
    3. Practical estimation of model parameters (mass, stiffness, damping) (7:45 min)  | Den Hartog Appendix
    4. Experimental estimation of damping in an underdamped system from measured signals.  This video describes the log decrement method, which will be used in lab to estimate damping. (9:24 min)
  • Accelerometer concepts.  You will be using an accelerometer in this lab to measure motion of the one- and two-story system.  If you are not familiar with accelerometers, this lecture provides a review of the workings of accelerometers, describing them as base-excited vibrational systems.  Here are lecture slides (pdf) and an online video discussion (~24 min).

    NOTE: The total time on the above videos is about one hour, equal to the regularly schedule weekly class lecture for this course.  Reviwing the  two-story system model will require a little more time.  At minimum, review the pdf notes from the video  which summarize the model.  The TA will provide some review during lab, but any preparation on your part will help you to complete the work.

    Two-story system:  Depending on how the first part of this lab progresses, there will be time to experiment with the the lab setup as a two-story system.  The following describes the model and how to calculate eigenvalues in LabVIEW. 
  • Two-story system model. The two-story system is a 4th order system, and the model is derived in detail in the video provided below.  This is a good example of systems you may be studying in your ME 344 course.  Review these notes from the video (pdf) to see the type of modeling and analysis described step by step in the video: 
     Two-story system model - model of the two-story system is derived in this online video (32:23 min).
  • Eigenvalues of two-story system. One experiment you might run with the two-story system is to simply deflect and release into vibration.  The resulting motion captured by the accelerometer(s) will show that the response is composed of two natural frequencies, rather than just one as before.  You can measure these frequencies in the lab.  The TA will provide guidance on how to estimate these frequencies from the system eigenvalues using the mass, spring, and damping parameters you estimated in lab. This bonus video reviews how to use the two-story model results for this purpose:
  • Using LabVIEW to calculate eignevalues (pdf slides).. These slides describe two different ways to use LabVIEW to calculate eigenvalues from a state-space model.
  • A review of eigenvalues and their relation to system natural freuencies is provided here:
    Eigenvalues and natural frequencies (video).
    You will be asked to calculate the eigenvalues for the two-story system in lab and to relate those values to measured frequencies.  This video explains how eigenvalues are found from the two-story model (7:24 min).
B. Pre-Lab - submit pre-lab as specified by your TA
  1. Model of single-story vibrating system.  The lecture slides (and video above) describe a lab model of a single-story system as a second order system.  In lab, you will be measuring the acceleration of this system after it is released from an initial deflection.  The frequency of the ensuing vibration can be measured from the acceleration signal.  This pre-lab asks you to estimate the vibration frequency before going to lab.  This requires that you estimate the stiffness and mass of the system components using the description provided in the lecture slides and this material and geometry data for raw materials.  Don't forget there are screws, nuts, and washers (8 on each story; about 5 grams each screw/nut/washer).
    Submit a summary of calculations and values for mass, stiffness, and undamped natural frequency for a one-story system based on these values (in SI units).
  2. Peak acceleration expected. Assume the largest initial condition you can give to a mass-spring damper system with the values of m and k estimated in problem 1 is about 0.04 meters. Estimate  the peak acceleration the system would experience during the ensuing motion after being released from rest with the given initial condition.
    Submit an estimate of the peak acceleration for the case described.
  3. Review of accelerometer basics. Here is a brief explanation of the general specifications for the CXL04LP3 model accelerometer which will be used in this lab.  
    Submit Explain what is meant by sensitivity, input range, bandwidth, and zero g output.  Summarize the indicated values for these specifications in the datasheet provided for the CXL04LP3.

C. Lab Work (meeting in ETC 4.160 at assigned time)
  • Pre-Lab should be submitted as specified by TA.
  • Guided Exercises.  The TA will introduce the setup and provide initial guidance.
    1. Accelerometer usage. The TA will guide you in making connections and in testing operation of an accelerometer, explain basic operation, and help with understanding of zero G outputs, etc.  The TA will explain how to verify the operation of the accelerometer using a quick '2 g' test.
    2. Vibration of single-story system. The TA may provide additional discussion on the simple vibration model using motion of an individual story of a two-story lab setup.  You may be asked to test the top and bottom stories individually.
    3. Finding system damping using logarithmic decrement.  The TA will provide guidance on learning how to use the log decrement method.  The log decrement method introduced in a lecture video (above) can be used to estimate the damping ratio for an underdamped system.  We provide here a subVI for LabVIEW that will analyze a measured signal and estimate damping using this approach.  
      • Here is the LogDecrement sub VI - there are 2 additional inputs that allow you to set the width and threshold parameters - these will be discussed by TAs in lab
      • To test the LogDecrement code, you need an unforced response signal.  Add code to this LabVIEW VI that calculates the underdamped displacement response for a mass-spring-damper system with a known zeta value, feeding the time and x data to the log decrement code to estimate 'zeta' (you are using the theoretical model to simulate the type of response you might measure during a physical experiment).   Consider a system that is underdamped with values of damping ratio of 0.01, 0.05, and 0.1, and calculate the corresponding value of the damping, b.  Use x(0) = 0.02 m, and xdot(0) = 0 (release from rest).  Experiment with estimating 'zeta' using different numbers of data points calculated in the response (as this can affect the estimate, as can noise).  Confirm that the log decrement code estimate the zeta value you specified in the model.
    4. Discuss a lab procedure for Student Exercises.
  • Student Exercises.
    1. Estimate effective damping ratio uising log decrement.  Run experiments to estimate the effective system damping ratio ('zeta') for each story of the two-story system lab model.  You should use the logarithmic decrement method.
    2. Estmate the damped and undamped natural frequencies.  This should be done for each story.
    3. Compare measured system parameters to those estimated in pre-lab.  Use your measured natural frequency to estimate stiffness values and compare to values estimated in pre-lab work.  Come up with other ways to improve your stiffness values if necessary.
    4. Damping coefficient.  Estimate the effective damping coefficient, b, for each story.
    5. Unforced response prediction.  From a model of a simple mass-spring-damper system, the unforced displacement response of the single-story can be calculated using an analytical soution if the system parameters and initial conditions are known. Use this solution to predct the acceleration response and compare to that measured using an accelerometer by plotting both together over time.  Consider the case where the single-story is given a known initial deflection and released from rest.

      Two-story exercises (to be completed only if specified by TA):
    6. Unforced response of two-story system.  Set up the two-story system.  With guidance from the TA, set up a vibration state and measure the dominant frequencies in the acceleration measured by one accelerometer (mounted on top story).The TA will discuss using built-in LabVIEW functions to extract frequencies from the measured signals (either FFT or Extract Tone).
    7. Compare measured versus predicted eignevalues.  Use your system parameter values to calculate the system eignevalues from which you can estimate the system natural frequencies.  Compare these to values you have measured from acceleration signals in step 6.
      • The TA may provide sufficient background for this step, or
      • You can review the extra videos in the lecture material (above) to complete this as part of the LE.
D.  Lab Evaluation - due as specified by the TA (may be completed in lab)
  1. Submit or present a summary of your parameter values for the two-story lab system (mass, stiffness, damping; estimated natural frequencies).  Explain why there may be differences.  This should include comparison of measured and theoretical natural frequencies for one-story system.
  2. Submit or present results from comparing measured and predicted acceleration for a test where one story is deflected with a known initial deflection value and released from rest.  These should be plots of acceleration over time.
  3. Submit or present (as required by TA) a comparison of measured and theoretical natural frequencies for two-story system.

Lab 5 - LabVIEW Vision Measurement & Analog Meter System Modeling (Week 8)
This is a one-week lab that introduces an experimental analog meter setup that will be studied using LabVIEW Vision.  This lab is required before moving on to system identification and feedback control labs using vision and the analog meter setup.

A. Lab Lecture and Information - online only
B. Pre-Lab note: submit answers to ALL questions per instructions by TA

Pre-lab 1 & 2 are discussed within the lecture slides above and reviewed in video 1.
  1. Sizing meter resistance. If you wanted the analog meter used in this lab to have full scale response with a 10 volt input, what series resistance would you use?  Submit your analysis.
  2. Finding angle from three measured points. Given you can measure three different points from the image, as shown in this image, express an algorithm that will allow you to find needle length, a, the distance, c, and the angle, q.  Submit a description of your analysis and algorithm design..
  3. Lab procedure.  Review the Lab Work exercises and submit a draft of lab procedures for completing the work.  The TA will review these procedures and provide guidance/feedback, but you will be expected to complete the lab work using your own approach with minimal guidance.
C. Lab Work (meeting in ETC 4.160 at assigned time)
  • Submit Pre-Lab as specified by TA - 2 questions within the slides
  • Guided Exercises.  The TA will introduce the setup and provide initial guidance.
    1. Laboratory setup familiarization (guided by TA): 
      • Review concepts as needed for using LabVIEW to conduct image analysis and image capture
      • SubVIs from TA(s): vision_setup.vi | vision_acquire.vi | vision_cleanup.vi
      • Conduct tutorials on LabVIEW image acquisition and processing
      • Guide examples on basic image analysis and image capture experiments
      • The analog out feature on the myDAQ will be needed to drive the meter
    2. Using the analog meter
      • Gain familiarity with analog meter 
      • Circuit diagram for analog meter box | TAs will explain operation - see slides
        • Both switches in 'up' position (1) places resistors in series with meter (calibrated mode)
        • Both switches in 'down' position (2)  places the potentiometer and 5.1K resistor in series (allows to change properties of the system)
      • Testing a VI for controlling the meter position (analog output)
    3. Review lab procedure for Student Exercises. The TA will give a brief review of the Student Exercises.
  • Student Exercises.
    1. Use the vision instrument to characterize the meter system and model the static gain relating angular position to input voltage
    2. Demonstrate open-loop control of angular position - i.e., specify an angle to be achieved and the VI should use an open loop control function to specify a drive voltage
D.  Lab Evaluationall LE work is through demonstration during lab
  1. Demonstrate a vision-based motion sensing system that can measure angular position of the meter needle.  You will need to show a working VI during lab.
  2. Demonstrate use of working VI to measure static relation between angular position and input voltage.  
  3. Demonstrate open-loop control of meter needle angular position per instructions by TA

Lab 6 - Frequency Response Measurement of Analog Meter System (with Vision) (Week 9)
In this  one-week lab you will learn about frequency response measurement and system idenfification using the analog meter system.  The vision measurement system used in the previous lab will be used to provide laboratory measurements that can be used to build a model and guide system parameter determination.  Understanding the forced response of a system as frequency of the input is changed is a fundamental concept in dynamic systems.

A. Lab Lecture and Information - online only
  • Lecture slides (pdf) | no lecture for this lab
  • It is assumed that you were introduced to transfer functions and frequency response functions in ME 344.  This lab will make use of this background, and you should draw on your course notes as needed to support your work for this lab.
B. Pre-Lab note: for Summer 2015, no pre-lab required

C. Lab Work
(meeting in ETC 4.160 at assigned time)
  • Guided Exercises.  The TA will provide support in getting started with this week's lab.
    1. Laboratory setup familiarization (guided by TA): 
      • The analog meter and vision measurement system used last week will be used this week.
      • The TA will provide help in making sure your setup is functional.
    2. The TA may provide his/her own perspective on measuring frequency response functions (FRF), which are the amplitude and phase functions that can be derived theorectically from a system transfer function.  This lab is concerned with experimental measurement of these relations.
    3. The TA will provide input on setting up a FRF measurement experiment for measuring analog meter needle position (angle) and input voltage amplitude.  These measurements are made at various frequency values, and when these tests are done using a sine wave input it is sometimes called a swept sine test.  
    4. The TA may recommend ways to measure the signal amplitudes and phase relationships.  Yo may consider using the "Extract Single Tone Information" VI.  This VI can be found under the menu Signal Processing->Waveform Measurements.  See the help guide on that VI.
  • Student Exercises. The TA will provide assistance only as needed.
    1. Make measurements of the amplitude and phase (between angle and input voltage) of the analog meter needle angular position (theta) as the voltage input frequency is varied.  In a swept sine test, it can sometimes be necessary to vary the voltage amplitude, especially to make sure you have a large enough output signal to measure.  Make sure to also measure the voltage amplitude.  This testing will require repeated measurements at each frequency, measuring multiple cycles of sinusoidal data.  It can be convenient and illustrative to organize this data into a table with headings: frequency (Hz), voltage amplitude (V), needle angle amplitude (rad), phase (deg)
    2. Generate experimental FRFs (amplitude ratio and phase) using your data.
    3. Compare the DC gain (value close to zero frequency) to the static gain you measured last week.  These values should be the same.  
    4. It is expected that the system will follow either a 1st or 2nd order model.  You should be familiar with the transfer functions and FRFs for these models from ME 344.  Compare your measured FRF with these two models and use this comparison to "identify" your system.  Make estimates of the key system parameters.
D.  Lab Evaluation - due as specified by the TA (may be completed in lab) - IN PROGRESS
  1. Submit a summary of your data collection procedure 
  2. Submit graphical results comparing your measured frequency response functions along with 1st and 2nd order models.  These theoretical models will need to have proposed model parameters: time constant for 1st order, damping ratio and natural frequency for 2nd order.  You models should also incorporate the static gain value.  Summarize your final model decision; i.e., is it a 1st or 2nd order system, report on the model parameters for your final model, and provide rationale for your decision.

Lab 7 - Feedback Control of Analog Meter  (Week 10)
This is the third and last in a series of labs using LabVIEW and vision.  You will now use the analog meter system as a plant to be controlled using your vision-based measurement of analog meter needle angular position as feedback.  This lab will give you an opportunity to study the feedback control concepts introduced in your ME 344 lecture and apply them to a system.  You will learn how to implement feedback control and practice tuning a PID controller.  You will also study feedback control through simulation and compare to your lab results.

NOTE: This is the last lab in the course.  Next week you will practice hands-on methods to prepare for the hands-on exam.

A. Lab Lecture and Information - online only
  1. Introduction (7:25 min) and slides (pdf) - this video explains the basic goals of the feedback control lab next week
  2. Building a model of the analog meter (15:32 min) and slides (pdf) - Last week you will have built an experimental model using frequency domain testing.  This video and slides reviews how you'd derive the model from scratch and it is put into transfer function form.  This form can then be used in a simulation as explained in Pre-Lab question 1.
  3. Review of feedback control concepts and terminology (20:20 min) and slides (pdf) - It is expected that you have been introduced to introductory feedback control concepts in your ME 344 lecture.  This video reviews these concepts as well, and prepares you for Pre-Lab question 1 where you will build a feedback control simulation in LabVIEW using your analog meter model.
B. Pre-Lab note: submit via Canvas per TA instructions; this pre-lab has help in form of video/slides
  1. Build a model simulation.  Use a LabVIEW simulation of the analog meter model in TF form.  Apply a step voltage input and compare the static gain and the dynamic model.  For static gain, use the value you measured in lab.  Use the values of damping ratio and natural frequency estimated from your frequency domain testing in last week's lab.
    Simulation of the analog meter in LabVIEW using a transfer function form (10:40 min) and slides (pdf)
    Here is a starter VI:  PreLab_SimTFModel_gaps.vi
  2. Simulate feedback. Use the model of the meter movement to simulate a model of closed-loop feedback control using the Control and Simulation Module.  The idea is to simulate the system operating with various types of reference inputs.  Start with zero and introduce step input changes to test the response.   Experiment with the PID gains.  Try to determine  the PID gains that will give minimal overshoot.
    Simulation of closed-loop control of the analog meter using LabVIEW (16:02 min) and slides (pdf)
C. Lab Work (meeting in ETC 4.160 at assigned time)
  • Submit Pre-Lab prior to lab meering
  • Guided Exercises.  The TA will introduce the setup and provide initial guidance.
    1. Review lab procedure for Student Exercises. The TA will give a brief review of the Student Exercises.
  • Student Exercises.
    1. Implement feedback control for analog meter position.  Test your feedback control using a constant reference angle.  Demonstrate control at various reference angles and use these tests to tune the controller.  Make a note of how effectively your simulations enabled you to narrow in to a proper range of values for proportional and integral gain.
    2. Implement tracking of a signal from the signal generator VI and demonstrate that you can:
      1. Stabilize the needle position at a fixed angle
      2. Introduce tracking of a dynamic reference signal.  First use a sinusoidal signal with oscillation about a fixed value.  Demostrate tracking over  various values of forcing frequency.  Based on your frequency domain testing,  you should know the range over which you can get good tracking.
      3. Switch the tracking signal from a sinusoidal signal to a square wave, and repeat the test in 'b'.
    3. The TA may also evaluate the ability of your controller to: a) regulate position when there are changes in system parameters, and/or b) given external disturbances.
D.  Lab Evaluation - due as specified by the TA (may be completed in lab); abstract required

Submit a summary of your feedback control implementation.  Report on how well your simulations (from pre-lab work) supported your lab work; i.e., were you able to implement and tune the feedback controller in a reasonable amount of time, and did the results from simulations match the lab results using your model parameters from frequency domain system identification?  You should compare the controller gain values and the simulation and lab results.  The TA may have clarification or additional points for you to discuss.


Hands-On Practice/Review (Week 11)

This week is devoted to review/discussions during scheduled lab times this week. 

Hands-On Quiz (Week 12)  - alternate times may be scheduled by the TA



  Return to Syllabus | Back to Calendar