I am so grateful for finding this video, as I've been seeking for awnsers to this specific case and exercise! Thanyou really much for producing such great content and videos!!
Hi if you're still responding to comments on this video, i'm in a situation where i have to tune a PI controller of a third order system. using the dominant poles method, this had first order dominance and so was reduced to only one OL pole in the raw root locus. Question 1: will i need to add another pole to this in the design as it doesn't break away as the desired poles are in the complex plane? Question 2: Is there any way to just tune a PI controller from this or would it have to be PID with the need to place a zero someplace to the left of this to allow it to pass through the desired CL poles. Thanks.
May I know how would you get the angle of the poles if the poles are complex numbers such as -1+1.73i. Is is still the same as imagining a horizontal line on the pole and calculating its angle anticlockwise?
I recommend sketching it on your diagram so that you can see the triangles. But yes you want the angle measured anticlockwise from the positive real axis.
Not sure about your zeta issue. Perhaps it's a brackets issue? Not sure where your other equation comes from - I've not see that before. It's very similar to the one that approximates settling time which is 4/(zeta*wn)...
So how would you find the desired/dominant poles for the PID controller after getting the location of all poles and zeros? I tried to use the angle condition, but I am not getting the correct values.
Thank you very much for sharing! I was wondering if this can be used for third or higher order systems. Also, how would affect having a pre-compensator. Thanks again!!
How would you separate the PI and PD controller processes? Such as when just designing a PI controller with no settling time but given a steady-state error due to a step percentage?
It's the same sort of process - you just only add in the part of the controller than you want though. I don't think have an example of designing just a PI or PD though sorry.
Isn't it required to close the loop prior to the maths? Sometimes it gets really confusing wheter or not the system should be in close or open loop. Very nice example, though.
It depends what exactly you're calculating as to whether you need to use the open or closed loop transfer function. There is also often more than one one to do the calculations meaning that depending on your approach, you may use one or the other.
@@theryderproject5053 What was your reasoning to use an open loop transfer function here? On the your previous playlist you used closed loop transfer function to solve for the variables (zeta and Wn). What was your reasoning for using a closed loop transfer function there? Thank you, I am having trouble understanding the approach.
This is the PD stage where the first zero is placed. In the PI stage which comes later, a pole is placed at 0 (the single s) and an extra zero is placed close to 0 to achieve pole-zero cancellation.
Please, why you put two zeros? How did you know it was going to be two zeros? I've seen methods that use only one zero and one pole. Are there different ways of doing thae same PID design?
If it's only getting one zero and one pole then it isn't a PID controller. There is a recap video in the playlist for the various types of controllers. I recommend looking at that.
Mam for implementation in simulink we need kp,pi and kd.for this we need to calculate these parameters.Can you plz help me, how to calculate the kp,ki and kd value from the derived pid equation.
You'd need to go through a trial and error process (likely using a computer because it would be much quicker) to find that solution. The manual calculations make some simplifying assumptions that impact the result. I think I talk about the specifics of this toward the end of the video.
This video doesn't reflect reality. It is easy to make formulas have a settling time of 0.1 seconds, but the control output is limited so the control output will saturate. The response will not be as desired. In reality the controller and amplifier only have so much power. The zeros are creating the overshoot. The zeros can be eliminated by having the PID gains act on the changes in the actual state only, not the error. Zeros can be placed too.
This half-hour video is a better PID design video than my prof's 10 hours of lectures.
I will go ahead and say it , this has taught me more than 15 hours of binge studying and watching lectures. I love you so much for this! Thank you!
hi, i'm from brazil !!! very clear and ellegant way of describing and solving the problem, very thankful for this video !
You are a savior I spent almost a full day trying to figure this out. Thank you so much !!! This was EXACTLY what I needed haha. You're the best!
I am so grateful for finding this video, as I've been seeking for awnsers to this specific case and exercise! Thanyou really much for producing such great content and videos!!
Thank you very much MAAM for your service, much appreciated. God bless
You make root locus design easy. Great lessons. Thank you very much!
Absolutely perfect pure education.. I cannot thank you enough. God bless you
Hi if you're still responding to comments on this video, i'm in a situation where i have to tune a PI controller of a third order system. using the dominant poles method, this had first order dominance and so was reduced to only one OL pole in the raw root locus. Question 1: will i need to add another pole to this in the design as it doesn't break away as the desired poles are in the complex plane? Question 2: Is there any way to just tune a PI controller from this or would it have to be PID with the need to place a zero someplace to the left of this to allow it to pass through the desired CL poles. Thanks.
You save my day. I nailed a really harsh test thanks to you!!!
May I know how would you get the angle of the poles if the poles are complex numbers such as -1+1.73i. Is is still the same as imagining a horizontal line on the pole and calculating its angle anticlockwise?
I recommend sketching it on your diagram so that you can see the triangles. But yes you want the angle measured anticlockwise from the positive real axis.
Your voice and explanation are perfect! Ty so much
Not sure about your zeta issue. Perhaps it's a brackets issue? Not sure where your other equation comes from - I've not see that before. It's very similar to the one that approximates settling time which is 4/(zeta*wn)...
You are using 0.05 for OS but you should be using 5. The percent overshoot equation assumes units of percent for OS.
Thank you so much for your efforts, I completed successfully my final exam thanks to you
Im a girl and I think your voice is sooooo cute and soothing! And youère a great professor too! From Canada!
more useful than my professor that said teaching is a passion for him
you are a life saverrrrrr, My teacher just says draw by scale and use a ruler to guess haha
I Don't know how many times I have to say thank you 🙏.
THANK YOU 😊
Love from VN
How did you take the desired poles @ -40+- 42 i , that is confusing for me.
Perhaps you can make a similar video for a unit ramp response as well? This was super helpful thank you
Thank you so much for such nice explanation!! Can u pls tell what this method is called? Pole placement method or something else?
Thanks for your videos! Very clear explanation and u answered all of my questions :)
So how would you find the desired/dominant poles for the PID controller after getting the location of all poles and zeros? I tried to use the angle condition, but I am not getting the correct values.
Thank you so much! watched pretty much all your videos and I cannot thank you enough!!! :D
Thank you very much for sharing! I was wondering if this can be used for third or higher order systems. Also, how would affect having a pre-compensator. Thanks again!!
Yes you can apply this method for higher order systems and if you have a pre-compensator.
How would you separate the PI and PD controller processes? Such as when just designing a PI controller with no settling time but given a steady-state error due to a step percentage?
It's the same sort of process - you just only add in the part of the controller than you want though. I don't think have an example of designing just a PI or PD though sorry.
Great effect at 8:28 - 8:34
Isn't it required to close the loop prior to the maths? Sometimes it gets really confusing wheter or not the system should be in close or open loop. Very nice example, though.
It depends what exactly you're calculating as to whether you need to use the open or closed loop transfer function. There is also often more than one one to do the calculations meaning that depending on your approach, you may use one or the other.
@@theryderproject5053 What was your reasoning to use an open loop transfer function here?
On the your previous playlist you used closed loop transfer function to solve for the variables (zeta and Wn). What was your reasoning for using a closed loop transfer function there?
Thank you, I am having trouble understanding the approach.
what if we only have damping ratio? will we use hit and trial method to find Wn or is there any other way?
i just love it. great work.. wanna more videoes.. ♥ from Bangladesh
hi ma'am what happened to the single 's' at the denominator s(s+1)(s+2) @11:07 when solving for the angles? Thanks!!
This is the PD stage where the first zero is placed. In the PI stage which comes later, a pole is placed at 0 (the single s) and an extra zero is placed close to 0 to achieve pole-zero cancellation.
I really loved you. You did save my life
My too ... You did save my life
Incredible explanation, Thank you very much.
Hello, great video. Why didn't you solve another angle for the pole at zero?
Please, why you put two zeros? How did you know it was going to be two zeros? I've seen methods that use only one zero and one pole. Are there different ways of doing thae same PID design?
If it's only getting one zero and one pole then it isn't a PID controller. There is a recap video in the playlist for the various types of controllers. I recommend looking at that.
@@theryderproject5053 thank you very much! I will definitely watch that
do you have a video where you do this example ( or similar) with the characteristic eqn and the K = 1 / GH ???
I can't really remember but there is a playlist that you could look though.
It didn't work for my system. 492.61e6 / (s^2 + 10^3s + 12.195e6)
I wonder why?
Mam for implementation in simulink we need kp,pi and kd.for this we need to calculate these parameters.Can you plz help me, how to calculate the kp,ki and kd value from the derived pid equation.
Amazing video! Thank you so much
is it for open loof system??
not close loof?
This is great! Thank you
Really good explanation! Ty
please make video for complex poles as well
Just wanted to say: Thank you
What does the equation you're writing for zeta say? The one where you include the overshoot.
zeta=sqrt( ln^2(OS/100) / (pi^2+ln^2(OS/100) )
We need more videos like this! Thank you very much ^.^
and the desired pole answer is 40 . . . 2 (hitchhiker's guide to the galaxy) quote
Very well explained.
Thank You
Can you use this for a first order system?
Thank you thank you thank you❤😍❤😍❤❤❤❤❤i needed this so much,u are an angel😙😘
what if ive got 16/S^2 + 0.2S +1 as my plant transfer function? my roots will be -0.1+-0.995i
You should be able to follow through the process shown in the video using this alternative transfer function.
Why there is nothing very interesting information during most replayed part ??
Does anyone know what 'In' for zeta equation is? 1:46
Final tomorrow, you are the greatest
thanks you saved my life
why you put a-42 for theta 1 and the ones before you put 40 -1 and 40 -2?
The poles at -1 and -2 came from the original block diagram to get the system's transfer function.
is there a way to decrease that 17% overshoot to 5%? because that is our requirement.
You'd need to go through a trial and error process (likely using a computer because it would be much quicker) to find that solution. The manual calculations make some simplifying assumptions that impact the result. I think I talk about the specifics of this toward the end of the video.
You two look really good :-)
I am from India, thanks you very much
hello may i ask how to find desire pole for 3rd order system? :(
It's going to be harder to do this manually the higher the order of your system...
@@theryderproject5053 what about finding the desired pole for a first order system?
you just saved my life
damn thats a clean video
Thanks! From 🇸🇦
Thank you so much.
Excellent !!!!
amazing!
incredable
3 hours left for my final I got the basics now or never if I will understand this video I can maybe pass that shit
Amazing! :)
thank you
thanks for posting the video some doubts I have can you share your contact info
thank you mam :*
You’re the real MVP 😤
GJ
I find a gold 🥇
why cant you write clear :) ,cant understand what have you wrote ? when you calculated zeta from max overshoot of 5 %.
it's your problem dude, it was clear af
This video doesn't reflect reality. It is easy to make formulas have a settling time of 0.1 seconds, but the control output is limited so the control output will saturate. The response will not be as desired. In reality the controller and amplifier only have so much power.
The zeros are creating the overshoot. The zeros can be eliminated by having the PID gains act on the changes in the actual state only, not the error. Zeros can be placed too.