Control Engineer Questions

[Kalagaraz 0]

(Go to the * to skip my long story and get straight to the point) Hi, I'm currently starting my 3rd year in a electrical engineering program. Unfortunately I have yet to take any actual engineering classes, as to take them I have to drive 1.5 hours away so I did my first 2 years locally and transferred, I start engineering classes this December. On the bright side I work for an automotive parts manufacturing company and currently work on the assembly line. During a plant meeting the general manager asked me to tell me a little about myself and I mentioned I was going to college for electrical engineering, after that he had the control engineering team come speak to me as they had an internship type program. When the control engineer guys came to talk to me, they said they really needed someone that could come in and just be able to do the job with little training and the position actually involved writing ladder logic for their PLCs. I never heard of those, but I got cocky and told him I would go home and read up on them and would probably be able to come in next day and be able to do it. So all day yesterday I studied up on PLCs and ladder logic, and played around with the software they use there (RSLogix 5000). I think I've got a pretty good grasp on ladder logic, but still a bitty hazy on how the whole PLC system ties in as a whole. Basically I know there are a bunch of inputs and outputs, inputs being sensors and outputs being stuff to be controlled, and I know how to program logic between them using ladder logic. What I don't fully understand are things like PNP & NPN inputs/outputs and more the hardware side of PLCs. Anyways, I came in the next day and told him I think I had a firm grasp on ladder logic and didn't think I would have a problem programming anything with it (I know C++/C#/Java and several other programming languages, so it was originally just a syntax barrier with ladder logic) he said OK and walked off while I finished my days work. At the end of the day, my line leader said to me that he heard that next week I will be moved to the control engineering department. *So that gives me about 5 days to really learn up on control engineering. Which brings me to my overall point, does anyone know any good online sources like ebooks (I'm willing to pay for good E-books) that I can really learn what a control engineer does and how to be one. I'm a fast reader and absorb information quickly (and understand it) so I think (here's my cocky attitude) that with these 5 days I can at least show that I'm knowledgeable enough on the subject that it would seem as if I was in my 3rd year of a traditional EE program. (Again, I'm acctually in my 3rd year, but I've only taken core classes. History, English, Math up to differential equations etc...) Also, reading up on PLCs kind of made me want to wire my house up PLC style and I found some decent expandable PLCs for about $300-$400. The problem is, any time I've googled things like PLC sensors, etc...the sensors are always way expensive. Like $125 for a temperature sensor, I'm wondering if I'm just searching the wrong way or if PLC sensors really are that expensive. Like a simple PLC switch was $10, too expensive to go replacing all the light switches in the house.

[Chris Elston 79]

Start looking in the download section. Dr. Hugh Jack has allowed us to host is ebooks for free is a great start. http://forums.mrplc.com/index.php?app=downloads&showcat=39 Do crazy crap like me with PLCs is another good way. Learn by doing: http://www.plchalloween.com/index.php?ind=gallery&order=2 http://www.youtube.com/user/chakorules#p/u/28/aScNOM0TKgk

[Kalagaraz 0]

Thanks for the links, Yeah I plan on getting a PLC for my house, I'd like to eventually hook up everything through a PLC (all the light switches, all the lights, computers, TVs, Thermostat etc...). That way if I wanted to, I could do stuff like control the AC/Heating from any computer in the house, or any switch in the house depending on how I program it.

[Ken Roach 194]

Professor Jack's book is an outstanding starting place, along with other articles here, on PLCTalk.net, and http://engineeronadisk.com/ From a consumer or IT background the prices of heavy industrial gear are going to cause some sticker shock. When you think about systems that lose thousands of dollars a minute when they break, it starts to add some perspective. That's why you see much simpler stuff in building automation or home automation. While I'm impressed by your brash confidence, don't make the assumption that because industrial control appears simple on the surface that there hasn't been a lot of thought and time put into it. One of the important side effects of an engineering education is an appreciation of the vast scope of what you don't know.

[Kalagaraz 0]

I still plan on completely my electrical engineering degree, I just don't want to walk into the engineering department monday and not know what the hell I'm doing. This is a kind of internship job. The company is paying for my EE degree.

[drforsythe 15]

The main thing to remember is that in this field, there is much to learn. Even seasoned veterans can be stumped by something - not because they are not smart, but because they might not have been exposed to something before. This field provides ample opportunity for those who are willing to learn and aren't afraid to ask questions. This forum is a great place to learn technique, so you have started in the right direction. Don't EVER be afraid to ask questions. I would rather work with someone who is willing to ask questions and learn the right way, than waste my time with someone who thinks he knows it all and ends up wasting time, money or worse. Best of luck in your endeavors.

[Crossbow 295]

My degree is actually in computers. And it's from 1995. Which makes it worth about toilet paper these days, as nobody still uses the computers and languages I was trained on. That being said, I went to work for an automation company in 1998. They couldn't find an EE for their tech support job, so they looked for a computer tech. I had no problem learning ladder logic, and basically sat in the lab and tried to make the PLC do different things. Then I went on-site to some customers, uploaded programs, and reviewed them. Sometimes looking at what someone else wrote (as long as they documented it) can shed light on complex problems. Other times it can make you bash your head against a wall. That being said, by 2000 I was working for my second automation company, and not in tech support. There I did control panel design, tech support, on-site service, machine design, and built and started up complete turnkey projects including PLCs, operator interfaces, and motion. And one note about the wiring your house with a PLC. Plan on being in that house for a long time... It's going to be an uphill battle. I've found most residential electricians have no idea what the PLC is, how it works, etc. And most home inspectors and code enforcement people don't either. If you do your work legally and get permits and such, prepare to spend a bunch of time explaining to them what it is doing. Otherwise they won't approve it. I had to spend 15 minutes explaining to the codes guy what the Siemens Logo was doing to my pool pump and outdoor low voltage lights when he came to inspect my installation. Boy will they be confused next year when I add the automatic level control to the pool. Going to install 2 solenoid valves, one by the hose to add water and one by the pool on the drain. 2 level sensors, and I'm in business. For home automation, something simple and home-user grade like www.x10.com would be good to give you the features you want without the cost of rugged and expensive industrial controls. My grand plan was to wire each outlet in my house 1 side live all linked together for each room and one side a home run to a relay in the basement control panel. Then wire all the light switches 24VDC as inputs to the PLC. This would allow me to connect a touch screen and simply select which outlet the switch turns on, allowing me to move the light switch connection when I rearrange the furniture. It would also allow me to write timers to determine when an outlet would be on. Simple solution, and my house is all wired in conduit (not common) so stringing extra wires won't be hard. Use the right PLC and Ethernet connection from a PC would be ideal too, being able to monitor or change the house remotely. Oh and I plan to add outlets under the eaves for the Christmas lights.

[Steve Bailey 118]

At the risk of hijacking this thread and steering it way off-topic... In addition to the issues with permits and building/wiring inspectors, are you willing to limit the potential buyers of your house to geeks like ourselves? Because a house controlled by a PLC might not be the home of everyone's dreams.

[Kalagaraz 0]

Good point, I did not think of this. I guess I could find some way to have it hardwired as well with a master switch to switch it from PLC control to just hardwired.

[Crossbow 295]

Promote it as a 'smart house' with 'environmentally friendly' automated controls. That will triple the number of people interested....

[Peter Nachtwey 15]

That is why I believe colleges and universities are 'cheating' their students. They offer degrees in tools when they should be offering only degrees in physics and the math required to understand it. The laws of physics won't change and neither will math. Programming languages, PLCs and motion controllers are just tools that will be come obsolete. Knowing control theory and the practical side help with tuning or designing control systems. Machines must be understood before you can control them. I learned how to design with transistors...... almost useless. People use op amps or have special integrated circuits to do the job. Now the biggest hassle when doing digital design is getting the design to work at high frequencies low power and get the final design its CE, UL and CSA rating. If you can program you should be able to learn how to program a PLC with put problems.

[Armadillo852 25]

Where are you located, maybe I'll apply. I love the auto world.

[BobLfoot 302]

Respectiful Counterpoint Peter - If the aim of the school is to make you a productive technician tomorrow who can maintain systems then they need to teach you the current tools and your employer and you need to make sure you stay current and don't become obselete. If the aim of the school is to turn out competent designers and engineers then yes you need all the theory you can grasp your head around and then some. I'll be the first to admit far to many of the maintenance people today are point - click - call tech support - board swapping idiots. My saying has always been "If you don't know how to do it without the computer, don't try and do it with one".

[Crossbow 295]

So few colleges which offer training on industrial controls own anything which could be considered current. I remember one before which had Omron C200H. And I was at a college a year ago that still used PLC-5. Colleges and universities (including technical schools) need to make sure what they are teaching on is still relevant. Why would they consider using 15-20 year old PLCs but use brand new computers? It's siple, they don't understand that PLCs are computers, and need to be upgraded just like computers. Electronics technology changes so fast, and components have a finite life span. Especially when they are tortured by students who wire them wrong, download to ROM every 5 minutes, etc.

[Armadillo852 25]

The only defense I can use with the PLC-5, is that Ford still uses them. Yes I know AB doesn't support them anymore. I just did a project for Ford last year and there were about 20 PLC-5's on a new application. I do not see anything wrong with teaching with the PLC-5's. AS LONG AS they have a ControlLogix or CompactLogix also. Really, is Logix5 really different then Logix500? I can see the need to teach the 2 different software packages. 5/500 and 5000. If we are talking AB, that is. Now as far as just teaching PLC's in general, the PLC-5 is still pretty powerful for anything you are going to do in school. School is to give you the basics, not make you an expert. Just my 2 cents worth.

[Leitmotif 0]

Re your statement I still plan on completely my electrical engineering degree, I just don't want to walk into the engineering department monday and not know what the hell I'm doing. This is a kind of internship job. The company is paying for my EE degree. I think you would be best off walking in Monday tell them you studied your head off and are still quite ignorant. They will appreciate the honesty and truth up front instead of finding the truth in a week or so and they WILL -- THEN the question for you is will you be able to overcome the black mark?? Dan Bentler

[Guest]

my apologies for going further "off the topic" of the thread – but it seems to have gone sort of stale anyway ... in defense of the PLC-5 platform, I'll go along with Armadillo852 ... the PLC-5 may be "old" but at least 50% of my industrial customers are still demanding training for it ... my ControlLogix business has definitely been growing lately – but be-that-as-it-may, there's no way that I'd even consider scrapping my PLC-5 training stations ... steel mills, paper mills, lumber mills, and you name it – there's still a TON of equipment controlled by PLC-5 systems out there – and SOMEBODY has to troubleshoot and maintain all of that equipment ... I tell my potential "walk in" students that if they don't have any specific reason to pick one platform over another – then go with ControlLogix ... BUT ... "do your homework" before deciding on the best fit ... if there's a big juicy plant right down the street that just happens to be chock full of PLC-5 systems, then you need to carefully consider which peg you want to hang your hat on ... you could show up for a job interview and be fully revved up on ControlLogix – but one or two questions about the Block Transfer Reads and Writes involved in the PLC-5's analog signals could quickly weed you out of the running ... the main thing that is often overlooked is that there are TWO major categories to PLC knowledge ... (1) programming skills ... (2) technician skills ... and I'd be willing to bet that for every single "programming" type job available, there are probably 500 (maybe more) "technician" type jobs that need to be done ... anyone who is interested in "PLC training" needs to carefully consider just what their objectives are – and then concentrate on learning MARKETABLE job skills to meet their goals ... one of the major complaints that I hear from new "corporate" customers about tech school and college PLC training programs is that they concentrate almost exclusively on teaching "programming" skills ... I wish I had a nickel for every time I've heard: "My plant's PLC programs are already written. I don't NEED a programmer. I just need someone who can go online with the PLC and figure out why the blasted pump won't run." ... yes, you'd think that anyone who had ANY understanding of how to program a PLC should jolly-well be able to troubleshoot a PLC-controlled system too ... according to my customers, that doesn't usually hold true ... more than 90% of my "corporate" students have already been through "other" PLC courses – and yet they still don't have the skills required to get the "problem-solving" part of the job done ... main point: the PLC-5 platform is indeed "old" – but there are still some customers out there who are specifying that any new systems they purchase must be fitted with these "old" controllers ... here are a few major reasons: (1) the plant's existing spare parts will fit ... (2) the same "old reliable" RSLogix5 software will still work ... (3) the plant's technicians are already familiar with the system ... a case in point from Armadillo852: party on ... Edited 7 Oct 2010 by Ron Beaufort

[Peter Nachtwey 15]

I somewhat agree but you said PLC knowledge, not control engineering knowledge. A real control engineer knows how to do algebra, geometry, calculus and differential equations. A real control engineer knows physics and how it applies to his system. Too many think control engineering is PLC programming but there is so much more.

[paulengr 44]

Peter, please realize that the skill set you are referring to is extremely important when it comes to motion control systems in discrete manufacturing. In the chemical industry for instance although the math is still important, the base knowledge shifts more towards chemistry (including especially surface chemistry as well as bulk chemistry), chemical engineering, and materials science, while physics only insofar as looking at fluid transport, and not so much F=mA. In many discrete manufacturing operations though the motion controls tend to be bang-bang oriented and "real" motion control is less important. I agree on the other hand with Ron (and probably you as well) that when it comes to control engineering, both technicians and engineers tend to do to much of automatically digging straight into the PLC which is least likely to be where the actual problem lies, and not even doing step #1 of investigating the problem by doing a survey of the area and the situation.

[BITS N BYTES 66]

And design to specified codes & standards... And thoroughly understand short circuit protection... And know what wire sizes/types to use ............ And know how to efficiently lay out a panel ............. And know how to size components for the specific application............ And follow required safety standards for the application [so you don't get sued]................. And know the how to correctly size enclosure thermal properties so components are within specs.................. And install and startup system with minimal supervision........................ And do this at the lowest cost!!!! .......................... The ANDS go on forever, COSTS are always the first priority!!

[TimWilborne 108]

Jacked http://forums.mrplc.com/index.php?showtopic=20368

[Peter Nachtwey 15]

There are too many questions asked on the forums I visit that deal with the physics of the system or math. Look at the recent thread about Knut's winder on plctalk.net or Erdemsrvi's filter. One person could figure out the relationship between the feed speed and the angle of the winder ONLY after I gave a big clue. Many people are so clueless that they don't even know what questions to ask or what is important to providing an answer. In these cases we must pull teeth to get the information. The hydraulic industry is even worse. There you have relatively uneducated people designing system and they think 'flow makes it go' . The worst happens when a PLC programmer is asked to work on a hydraulic system that wasn't designed to be controllable. The poor PLC programmer busts his butt trying to get the desired results but he doesn't understand that the system will not physically do what is desired. I have seen lots of frustration, time lost, and money wasted. Paul, since we sell motion controllers we end up doing a lot of tech support but mostly application support. If we sell about 500 axes and 99% go in without a hitch then there is still 50 problem axes that have poor designs. That is almost one a week that is totally screwed up. Either it will be costly to do right or expectations must be lowered. We are talking about Control Engineer on this thread but the topic keeps changing. There are many hardware things that can do control besides PLCs. Yes, this is important but it is not Control Engineering except for the design issues.

[Chris Elston 79]

I find the term "Controls Engineer" to be broad. There are some Controls Engineers that know PLCs, and how to integrate with other servo systems, vision etc. Then there are some Controls Engineers that know nothing about Hydraulics as you say, I know because I am one of those people. But I applied "some of the knowledge" of basic servo system tuning with alittle direct coaching from Peter on my first massive hydraulic system. As a Controls Engineer though, I often will look at is it worth my time to create a control system that can control a hydraulic system or servo motor direct from the PLC rack? I often answer that with a NO. That's why companies like Delta Computer Systems, Emerson, Trio Motion, all robot companies exist. They are providing a VALUE to the industry with PRE-ENGINEERED SOLUTIONS. Auto tuning, PIDs, etc. Because of these value services, Controls Engineers would rather purchase these COTS (common off the shelf) components to put together their controls systems. This means they don't have to be a PhD in order to program that crazy hydraulic system. That's my theory anyway. But I do believe they should be able to understand the BASIC CONCEPTS of how it works. Like know you have enough resolution in your feedback device etc. I think and hope that is the point you are trying to drive home Peter. For me, I purchase products that speed up installation, lessens my time I have to write code to a device. I would rather send a "GO" bit to a RMC controller and have the that controller crunch the math, and do the amazing wiz bang things it does then tell me "I AM DONE" then I can get on with the next project. I can clearly justify some of my lack of knowledge, especially the hydraulic parts to make my life easier yet still come out a hero on a project because I was smart enough to know my limits of trying to design a control system for a hydraulic system from scratch or buy an off the shelf product I know will work. And still saved the project a TON of money because I would have struggled with the controls system and spent more time in engineering time, software time and ordering parts than the cost of a PRE-ENGINEERED SOLUTION. On a personal note: I still need to write up that review on the RMC controller as my FIRST EXPERIENCE articles, I promise it will be good. Over time in this industry, I've had to solve some pretty hairy controls problems. I agree that it comes down to the math. I tell my high school robotics kids that I want you to prove that your controls using that gyro and the three drive motors you can control the system via a math equation. Please simulate in MathCAD or Labview for me to see.... Once you have the math down, writing the code is easy, I tell them. Most first time programmers don't get that BASIC CONCEPT either. Do the MATH FIRST..... In PLCs, I don't face math problems everyday, but mostly LOGIC problems. Physic problems inside PLC ladder logic is few and far between. But GOOD machine knowledge is a necessity in programming an automated machine. Edited 19 Oct 2010 by Chris Elston
fix some grammer