PID Trainer, issue on setup

[MaxK]

GIT, a few questions:
1. Why this thread?
2. 2 questions.. determine the order yourself
- at what price* would you rate the services rendered to you in this thread IRL?
- on what basis do you consider yourself entitled to ignore / sabotage the advice offered to you?

* I do not ask for money (read below)

You are an ordinary, I believe, quite adequate, and not an ignorant person.
Therefore, with the adjustment of this circuit, you make all the classic mistakes of an ordinary person. (You posted a video of a rolling ball - I guess the consumer likes it, but it's pointless; you managed to remove the only data that wasn't useless from the graph - CV)

Given
-240 posts,
-specialist support
result = 0

Question:
Who are you going to sell this PID trainer to?

 

[geniusintraining]

1. Why this thread?

As stated several times I thought it was quite obvious, I have tuned several PID's in the past and built one device (PID Trainer) but I dont know as much as you guys about the PID setup... SO I started this thread in hopes to get some help (and I have) I sell tools, I do not teach, I only need to make sure my tools work as intended (NOTHING MORE)

The point and only point of this thread is/was to make sure the trainer can be tuned and I am selling a product that works

you managed to remove the only data that wasn't useless from the graph - CV)

what the hell can you tell from the CV going from 0-100 every .03 sec?

Question:
Who are you going to sell this PID trainer to?

People that want to get more comfortable with PLC's and learn how they work

Max you dont have to read this thread, I appreciate you and anyone else taking time out of their lives to help or try and help me but this site is voluntary and anyone can or can't help if they wish... trust me I wish Peter or Dr could wave their damn magic wand and get this damn thing working 16 pages ago and most people know I would be glad to hire and pay someone else to do it and yes I have hired and paid several people on this site that have helped me in the past

 

[Peter Nachtwey]

As stated several times I thought it was quite obvious, I have tuned several PID's in the past and built one device (PID Trainer) but I dont know as much as you guys about the PID setup... SO I started this thread in hopes to get some help (and I have) I sell tools, I do not teach, I only need to make sure my tools work as intended (NOTHING MORE)

drbitboy correctly stated that you don't tune PIDs, you tune systems. Just because you tune a PID doesn't mean you know how to tune all systems. I have shown in above links how the symbolic formulas for the controller gains are derived. Each different type of system has a different set of formulas for each of the gains so knowing one PID does not fit all.

The point and only point of this thread is/was to make sure the trainer can be tuned and I am selling a product that works

That was your original intention but you didn't heed my post #8. The rest is just stubbornness. I can admire that to a point but this project has gone too far.

People that want to get more comfortable with PLC's and learn how they work

This project will not do that. PLCs are not the right tool for motion control. I haven't seen any ball and beam videos on the internet that use a PLC. They all seem to be using Arduinos.

trust me I wish Peter or Dr could wave their damn magic wand and get this damn thing working 16 pages ago and most people know I would be glad to hire and pay someone else to do it and yes I have hired and paid several people on this site that have helped me in the past

We could but there is something about your system that makes it difficult to control and drbitboy and I are not sure what it is. A proper system identification needs to be done to get a true model.


However, it makes no difference if you get your ball and beam balancing project to work. It won't work for your stated purpose. It is too difficult for beginners. I warned you about that in #8 and NO ONE LISTENED.



At least you didn't break Paula's "learning by doing thread" record.


One more thing. Way back in 2005 I modeled Ron Beaufort's hotrod temperature control systems. I didn't is on a ML1500 and SLC5/05. I wrote the simulator in ladder. The PID still needed to be tuned but since I wrote the simulator I need all the system parameters so it could be tuned perfectly using my formulas. No one seemed to care. It sat in the MISC are of this forum's library for years. All that is needed is a SLC5/05 or ML1500 but the code could probably be migrated to a newer ML easily enough.

 

[MaxK]

People that want to get more comfortable with PLC's and learn how they work

trust me I wish Peter or Dr could wave their damn magic wand and get this damn thing working 16 pages ago and most people know

Is my English that bad, or do you think it's so unnecessary to delve into what is written?

The point I am trying to convey is:

You HAVE the support of Peter and the Dr. You have NOT configured the system
Whoever buy DID-trainer will NOT have the support of Peter and the Dr. What will the simulator teach the buyer? helplessness?


Anyway. The conversation is pointless. Waiting for you to do the correct "time scaling" according to #238 by drbitboy. Then he publishes "what the hell" curves and PIDE settings.

 

[PLC]

GIT

You should box it up and send your demo to Peter or the Dr.
It’s obvious they have nothing else to do.

Over 6000 posts in just a few years. Wow

 

[MaxK]

GIT

You should box it up and send your demo to Peter or the Dr.
It’s obvious they have nothing else to do.

Over 6000 posts in just a few years. Wow

Do you know how to set up the system?

It’s obvious they have nothing else to do.

But really, apparently, we have nothing to do .... they all work fine without us ...


To avoid being misunderstood again:
Give an example of one single phrase, which will be able to explain how the PID controller is tuned.

We learn from you with great desire and zeal

 

[drbitboy]

Over 6000 posts in just a few years. Wow

Yeah, but near-nil SNR . Consider also my post:application ratio.

The best thing about the internet is that it lowers the cost to publish.

The worst thing about the internet is that it lowers the cost to publish.

 

[drbitboy]

Give an example of one single phrase, which will be able to explain how the PID controller is tuned.

@MaxK: to whom is that request directed?

We cast our bread upon the waters ...

 

[Peter Nachtwey]

I explained how the ball and beam PID should be tuned. I derived the symbolic equations for a critically damped response. The formulas are correct. You just need to supply the correct values for each of the variables. That is what we don't know because a system identification was NEVER done.

 

[drbitboy]

I went back over the thread, seeing who might have mentioned the time issue that has been the problem all along.

@Genius.
I remember long time ago ( I am retired now) when I was working with RS500 that the PID instruction has a timer or something similar, have you experimented trying to change that value. ?
I also remind of something called STI that I used to use when I needed a faster response, I am not sure if that is suitable in this case.

I found this. posted within 2h of the OP. It does not address the actual issue directly, but it is close enough you would think someone would have asked the question "How are you scheduling evaluations of the PID instruction?"

d'Oh!

"... misunderstanding all [we] see ..."

 

[MaxK]

@MaxK: to whom is that request directed?

We cast our bread upon the waters ...

This is absolutely obvious: The question was asked by @PLC , so I am waiting for an answer from him

 

[padees]

GIT

You should box it up and send your demo to Peter or the Dr.
It’s obvious they have nothing else to do.

Over 6000 posts in just a few years. Wow

I agree, but GIT should send it to both. Bit is theoretical and never has worked in a real world application. He only has a ML1100 with free software to play with. When he posts of a real world process factory he engineered, built and programmed by himself...


Peter is retired and think he still has the itch. Most people I know who retire loose the itch in a few years. Much better things to do with your time.

 

[drbitboy]

Bit is theoretical and never has worked in a real world application.

https://www.gocomics.com/peanuts/1955/04/21

 

[MaxK]

I agree, but GIT should send it to both. Bit is theoretical and never has worked in a real world application. He only has a ML1100 with free software to play with. When he posts of a real world process factory he engineered, built and programmed by himself...

I'm guessing it won't take the Dr half an hour to get the system stable.
So what about you?

Like the previous speaker, I invite you to explain the tuning of the PID controller in one sentence.
Or talk about modeling, math, and tuning PID controllers, something we don't know. With your rich practice (according to your words) it will not be difficult.

We cast our bread upon the waters ...

According to the latest posts
We throwing diamonds in the cesspool

 

[thehoneybadger]

i don't want to be that guy. But check out this video.

https://www.youtube.com/watch?v=K-F_T59ZDPw

I know it is not a PLC, but it is amazing none the less.

It does introduce a novel idea that may be useful to GIT. It uses a touchscreen to track ball position.

May be worth considering...