Controls/Spring 2012/Test 2

This page is the review sheet for Test II for ECE 382 and ME 344 in Spring 2012; post any questions or requests for clarification to the discussion portion of this page (see tab above).

Previous Tests

Previous ECE 382 tests are available at Dr. G's Big Box of Random. The closed books parts of Spring 2009 and 2010 for ECE 382 are a good start, but note that they do not cover PD, lead, PI, or lag control in the same way that you have done in the homework and thus represent an incomplete measure of the test.

IMPORTANT NOTE!

Test II this year is wholly closed-book.

Equation Sheet

There will be an equation sheet for the test. It will be the same as the equations sheet for Test I. If people would like equations other than that, post them on Piazza for consideration.

Test II Spring 2012 Coverage

• Given a block diagram, be able to simplify it into either unity feedback form (single Geq) or the four-block paradigm of input transducer, output transducer, plant, and controller. The latter version is needed when finding actuator error signals (i.e. comparing Force to Force).
• Given a system, be able to determine the system type, appropriate static error constants, and steady state errors for step, ramp, and parabolic inputs. Know how to adjust when these are not unit step, ramp, or parabolic inputs.
• Given a system with gain control, be able to determine the minimum possible error by using a Routh Array to determine the range of stability and the static error constants and steady state errors calculated as functions of gain.
• Given a system with gain control, be able to sketch the root locus including the real axis portions, calculations of the asymptotes center and directions, \(j\omega\) axis crossings and K values for them, and break points. For the latter, you will be given solutions to \(dK/d\sigma=0\) and will need to know what those solutions mean. You will not need to calculate or apply angles of departure or arrival.
• Given a specific set of transient design criteria which pinpoint a desired set of system poles, be able to locate those poles. For example, if you are given a system and a particular OS% and Ts.
  • Be able to determine the location of the PD zero to make this happen and the gain for the compensator in that case.
  • Be able to discuss when you may want to use a lead controller and place the zero somewhere other than the PD location; also be able to determine the location of the lead pole and the compensator gain in this case.
• Given a specific desired type, static error constant, or steady-state error for a system, be able to design a PI or lag controller to make that happen. Also be able to discuss the main considerations for where to place the poles and zeros for these kinds of compensators.
• Be able to "put it all together" to design a compensator that can both adjust the short-term transient as well as the steady-state error.

Specifically Not On The Test

• Compensator circuits