wat is the right procedure of automation????

[g_khl]

hi i am a student of electronics third year n i need ur help as i am a begginer in this sector.

there are so many things i want to know

a)how does the automation work really starts-i mean is it that first tags are made then the ladder logic and then the graphics are made and their animation is done??????????????????
if i am wrong then pls correct me and dont get angry on me cos i am trying hard but i still dont know much abt it.

b)i am training in a firm abt automation but i dnt have a desktop there-i have tried my hands at manuals but i think without doing anything practically i cant really get anything,pls tell me some good way of learning and practising.

c)i have done testing of digital n analog o/ps n i/ps till now.i have to make a project in abt 4 months from now pls help me with some choices abt which i can have thorough knowledge as i have only 4 months.



pls kindly help and thanks for showing concern.


thanks

gurleen

 

[jvdcande]

Gurleen,

I sincerely wish you good luck on the thorough knowledge. 4 months is a very limited time to acquire it. This is not to put you aff, but I have been involved fulltime in industrial automation for the last 25 years and every day I become more and more aware that my knowledge is far from thorough. There are a lot more things I don't know than things I do know, even in industrial automation. Why do you think I'm involved in this discussion forum, other than to learn more and more every day?

Automation work really starts out in someones head. He/she thinks up a product to make, beit based on something already existing or not. Once the product is defined, which can take a long time because people have to make prototypes that need to be enhanced over and over again until it reaches it's final agreed state, you can begin thinking about automating the production process. Because people have already made prototype products, it's a good start to see how they did it.

The production process is mostly about handling and assembling parts, so you need to look into it from a mechanical point of view. Once the mechanical part is defined you have to check if it is feasable to make it that way. If not, it's back to the drawing board to overcome the problems that were detected. This stage is repeated untill everything is OK.

(Sometimes it isn't needed to think about the mechanical part of the equation because the machine is already there, but it needs to be automated.)

When the mechanical part is resolved, you can think of automating it. Here you can make the choice: do you control the machine by relays, electronics, microprocessor, PLC, pneumatics or even some other technology? I personally find it very helpfull to draw up a sequential function chart. I didn't write that name with capitals on purpose. If you know how, you can translate a sequential function chart to any above mentioned technology. I have seen magnificent handling machines controlled entirely by pneumatics. And even with my limited knowledge of pneumatics I could understand how it worked, because it was designed using sequential function chart.

Once you arrive at that point you have to check wether the choosen technology can be applied to the mechanical counterpart. If not: back to the drawing board to change one or the other until everything can work together.

Now you can start to fill in the details: what components will you use, how will you connect them together, how will you write the program (if needed).

And don't think you have finished by now! Oh no! Now you have to test the result, make changes for things that don't run as you had imagined, go back to the drawing board (by now this has been replaced by a CAD package on PC), design corrective measures, implement them, test again, and again , and again...

One measure I always take: make sure you know enough people from enough different fields of experience so you can consult other people if you don't get it anymore. Pay them by being around if they are in a similar situation.

It's a hell of a job , but I wouldn't do anything else, even if they would pay me a multitude from what I get now (in fact: don't believe me, it depends very strongly on the amount on the paycheck). It's the best job around: you start with nothing and end up with a working machine. And best of all, you keep on learning day after day without having to pay for it.

What a wondefull world!

 

[g_khl]

thank you so much sir.
i know no knowledge is enough but i am trying to work hard to know abt automation as much as i can and everything u have written has definately helped me.i am trying to learn ladder logic.i have tried converting sequential flow charts to ladder logic programs.

but i am not able to start i mean wat should i try to automate is the question i am not able to answer-i mean since i am still learning ladder logic(actually i have jst started and all i can learn from is net)i want to try something which is easy for the first step thats why i have tried converting sfc.

i hope to get more advise from proffesionals like u in future i will definately stay tuned to this site n group.


thanks for all the help
thankyou
gurleen

 

[Terry Woods]

Gurleen...

It sounds to me like you are putting the cart in front of the horse...

It sounds like you are talking about all of the higher-level stuff without first being well grounded in the basics. At the very least, you can look through the free tutorial at the top of this page. Then consider buying the book. And then... take advantage of the free simulator! (...at the top of the page.)

Your primary goal is to figure out how to automate something... that is, some machine. Machines are automated all the time without any kind of graphics or operator interfaces.

As you indicated, you need to start out small... the simulator will go a long way to help you understand the basics. But, bear in mind, this is a very generic simulator.

So... you might become very proficient on the simulator... but that still doesn't give you the feeling that you are really automating something.

At the very least, you need to get your hands on a small, inexpensive, PLC. Then, you have two options for programming that PLC... using a PC, or using a Hand-Held Programming Device. Either way, you are going to have to spend some money... yours, or someone else's.

Once you have a real PLC, and a means to program it, in hand, you will then need to acquire input and output devices. Initially, just for the sake of learning, Inputs can be satisfied with simple push-buttons and switches. At the very least, Outputs can be satisfied with different colored indicator lights. You might consider adding buzzers, relays,... etc. as time goes on. Again, money.

Up to this point, you are just learning the basics of digital logic.

At some point, you need to think-up a project... nothing big... not expensive... but, something more than just turning on lights and buzzers.

You should be able to get some really small and very inexpensive DC-Motors, and some really inexpensive sensors from a Hobby Shop. Then you have to be a bit "handy" so you can make the system you are trying to automate.

Let your imagination go wild! (However, remember to stay on budget!)

By the way... one of the hardest parts of learning automation is to get your mind around the basic relationships in cause-and-effect.

It's very easy to say... my machine will do this, then this, then that, then something else, and then it is done. The basic questions involve finding answers to questions like... How and why does the process "know" that it is time to do this, or that? How and why does the process "know" that it is time to stop doing this or that?

For some, these can be very hard questions to answer.

Any process is highly dependent on all of the tiny details.

 

[trojan_goat]

Something to Add

Some questions to ask before a System should be automated





Can you remove some danger from a job to increase worker Safety?



Can you make a better product, but still make it at a cost that is competitive?



Can you make the same product, but make it faster and /or less cost?



Sometimes the decisions about how much to automate involve self-preservation. Always leave manual controls; it helps a lot with debugging and set-up.



If the automated system fails, then you can at least limp through on manual.

The manual function also allows for easy troubleshooting and maintenance operation without powering up the entire system.