In response to the question, “I would like to know which chapters would be there for the first midterm?”
My answer is tentative, and there may still be changes. My plan is to have three problems. One problem *must* be solved, but you may choose one from the other two problems to obtain full score.
The mandatory problem is similar to Quiz 5, with emphasis on dynamic response and pole placement. Then you can choose one problem that is block diagram-heavy or the other, which involves examining a given system and finding the constituent equations and transfer functions.
Professor, when is the 1st midterm?
We have not decided yet. I’ll try to consult with the class to see what week we have as little conflict with other courses as possible. To achieve this, we may actually wait until one or two weeks after the official semester midterm (which is probably 2nd or 3rd week of February?)
Is there anyway we can get more worked out examples to help study for the midterm?
Is there anyway we could get more worked out examples to help practice for the midterm?
I will try to include some examples on Tuesday, but there won’t be much time.
I have one suggestion, though. You can access Schaum’s Outlines through our library. (link on my web page). I believe there are problems in there. Look at problems related to steady-state and dynamic response, problems that require you to use Laplace correspondence tables, block diagram problems. This should get you covered.
Why would there be a quiz due the Tuesday after spring break? We literally only have one class period worth of material (this Thursday) and then the quiz? This class is moving way too fast and I’m having a hard time comprehending the material.
We already covered the materials. Check it out, this is a very easy one. It is Routh-Hurwitz, last week’s subject.
And if you have any doubts, please raise them in class. I am always hoping that, if I move too fast, somebody (like you) asks for further clarification. If this does not happen, I have to assume that everybody is with me. So I move on.
We reviewed the purpose of the P controller (the workhorse) and the I controller (handles steady-state error), but could you detail the purpose of the D controller a little more? I know it adds flexibility in adjusting the 2/1/0 order coefficients independently, but what does it do directly to the step response? Does it just serve to dampen overshoot from the I?
In what situation would a PD controller be more appropriate than a PI?
This is a pretty good question. We are not really there yet, although the latest quiz gives you some indication what the D component (or, let’s call it a phase-lead component) does. Most critically, it can add damping. This feature will be at the core of both the levitator and the upright robot projects, because it is *this* controller zero that pulls the closed-loop poles into the left half-plane. We will also look at frequency response (after the midterm), and we will recognize that a phase-lead component in the feedback path suppresses oscillatory tendencies and stabilizes high-gain P-controlled systems (e.g., voltage feedback in power supplies).
As for PI — You *need* the integral component when a zero steady-state tracking error is required. However, you will raise the order of the closed-loop system by one (in most cases). Its purpose is entirely different.
To respond to several questions what chapters are covered. Let me give a few general suggestions.
Your first stop, in my opinion, is to look again at the quizzes, in the order of decreasing importance Q5, Q2, Q3, Q6, and Q4 (two point control will not be part of the exam). If you *understand* why the results are what thy are, and if you understand how you get there, it is already a good start.
Useful preparation would include that you prepare your notes in a fashion that makes them easy for you to access. Annotate your work and put it in a binder for quick reference. Naturally, you also should have a set of Laplace correspondence tables at hand.
As mentioned above, you can access Schaum’s Outlines from the UGA library. I believe there are problems in there. Look at problems related to steady-state and dynamic response, problems that require you to use Laplace correspondence tables, block diagram problems. This should get you covered.
Finally, what chapters? The single most important chapter is Chapter 9 up to p. 133. If you feel unsure about the relationship between Laplace and differential equations, try looking at Chapters 5 and 6. These treat one simple problem in parallel — time domain (Ch. 5) and Laplace domain (Ch. 6). In addition, take a look at Chapter 7 and Chapter 10 (not Jury test, though). These chapters cover most, if not all, of the exam.
Are we to do Problem 1 and then either Problem 2 or 3? Or are we required to do all three problems now?
As a take-home exam, all three, please. Consider that I introduced a 15-point bonus opportunity.
As I am working on the exam, I have noticed that the total points available are 120 rather than the 105 listed on the first page of the exam. How will this translate into scoring and grading?
The total points of the exam is 120. So 30 marks would be bonus?
Let’s see what others think. 30 bonus is probably a lot, so maybe 100 of 120? For a max of 20 bonus?
For part E of problem 1, Do you want a specific number for Kd and Kp, or will a relationship between the two suffice?
You don’t have any values, so you get a relationship between the two. This allows you to eliminate one (see checkpoint)
I don’t know if this is too specific of a question for you to answer without giving too much information away, but for Problem 3 part A is it basically just asking us to find the value of the slope?
Yes, this is absolutely correct. No big deal. Even easier if you took a look at my book 😉
Ok thanks. So we don’t have to get the whole linearized equation? Just the slope
“… as we will examine only small deviations from the operating point.”
On Part F of Problem 1, do we include the impulse response for the entire closed-loop function (including R(s))? Or just the contribution of Fd(s)?
Good question. I think it is fair to assume that the disturbance occurs with the system at equilibrium, which means that you only need to look at the disturbance term.
To answer your question as to why many people did not answer and or finish problem 3 was in part do the fact that the test clearly stated that the problems would be graded out of 90 points. It’s hard for me to speak for others, but I wasn’t exactly looking on the blog much once I received the test. Therefore, I did not see your correction to the point system. Some sort of email or notification would have been a better way to get a correction like that across. Also, part 3 did seem rather difficult. Upon reaching part b, I did not believe my work to be correct, therefore, I did not continue because I knew this would ruin my answers for the remaining parts. Believing that number 3 was purely for bonus (due to the 90 point grading scale) I then chose to quit working, as I could not figure out how to proceed further
Thanks for your post and for your frankness. I appreciate it. I also understand your frustration, and I acknowledge that I made a mistake by re-using the a sum that already contained a 15-point bonus. I tried to make amends by increasing the bonus to 20 while maintaining the score totals.
It occurs to me, though, that you are playing a risky game by relying on bonus points. You did the math: 90 points plus one whole bonus question (how realistic is that?), but you did not take into consideration the stated total of 105 points although you surely must have read it. Once take a closer look at this, with a 15-point bonus and a sum of 105, you’d see that *something* has to be wrong. Either 105 is wrong, or 90 is wrong. The low-risk approach would be to request clarification, as some of you did.
I know this has been discussed a lot, but is there any projection for when we will have our full grades for the midterm? I know a lot of us are waiting to see where that score puts us against the 90% necessary to choose to not include the final exam in our grades. As we get closer to finals time and everything is getting down to the wire, whether or not we need to begin preparing for your final is a big concern based on the subject matter covered in the course. I realize that there is a lot to be graded and that the graders are students too. I’m just looking for some idea of when we might know. Thanks in advance for your response.
I am really sorry, but I have to ask you to direct these questions to the graders.
From an instructor standpoint, I can only suggest that preparing the material at hand is never wrong. A hair-breadth calculation of what you include and what not may go badly wrong (see anonymous post from 3/21 above).
Also, I think that you can at least get an initial estimate based on what is posted in your midterm feedback box, i.e., the partial score that has been graded to-date.
Grading of the midterms seems to be at a standstill not just for myself but for others as well. Do you know approximately when the full grade will be put on elc?
Your email address will not be published. Required fields are marked *