State of engineering in US

Jiri Toman said:

And may I ask who is "We" ? Do you speak as an educator?
Where are you located? What college? Who are the local employers?

Click to expand...

I am a Professor at Grand Valley State University in Grand Rapids Michigan. The population in the Greater Grand Rapids area is about 1 million. Our local employment base is approximately 20% in manufacturing (one of the highest in the nation). Our local industries are quite diverse including auto parts manufacturers, office furniture, medical, aerospace, etc. You can have a look at the chamber of commerce page for more information.
http://www.grandrapids.org/

Many of our local employers face issues including outsourcing, mandatory cost reduction, increased competition, technology driven innovation, intellectual property driven business models, etc.

Hugh

I am a Professor at Grand Valley State University in Grand Rapids Michigan.

Click to expand...

Jack,
Clearly you are not an impartial third party. You have a vested interest in telling us that all your graduates are immediately employed. I would like to see some supporting statistics rather than just your opinion.
Further more, kids out of school are willing to work for low salaries just to get some experience.

- an analytic approach to problem solving

Click to expand...

If you have not educated your graduates to solve problems then I suggest you give them their money back, being able to solve problems is what engineering is all about, without it you are not an engineer.

- willing to work long/odd schedules

Click to expand...

Of course this is what I am talking about. Pay these kids 40 000 per year and have them work 60 hours per week. It is very common for employers to demand extra hours for no extra pay!

- willing to put a company/team ahead of themselves

Click to expand...

Well my friend you have not worked in a Company lately. It's all about top grading and smart goals. You compete with others for the money allocated to salaries in the budget. All Companies promote team work but yet you are judged on your performance versus performance of others in the department and around the Company.

- good communicators

Click to expand...

This is true, H1B visa engineers are not the best communicators but they are cheap. It takes only two days to fill in the quota of 65000 per year!
So the communication issue is not that important after all.
Besides, many professors at universities don't speak English very well (you are not included), but that doesn't stop schools like Purdue to hire absolutely ridiculous people. One of my friends went through three professors in one semester. The class had to stage a walk-out because the professor was teaching with an interpreter at his side. You are responsible for producing these poor communicators!

In many ways the problem lies with universities that put preparation for research programs ahead of graduating well educated universities.

Click to expand...

I think you meant to say "well educated people". Not trying to nit pick here, I know you what you mean. Thank you for admitting this.

Any article that lumps all engineers into a single reference just doesn't get me excited. I am not simply an engineer, I am an electrical engineer specializing in industrial power distribution systems located in the Midwest.

Please do not try to evaluate my pay scale and employment opportunities the same as you would a process controls engineer in Texas and definitely not the way you would a software engineer designing video game systems in Washington.

Click to expand...

I agree with you that the type of engineering and geographic location makes a difference, but that is all part of the standard deviation. The article refers to engineering and science as a whole, it looks at the bigger picture.

Here is a link to a very cool web site (not my own and I am not affiliated with it)

http://www.automation.com/sitepages/pid2965.php

It gives you results of a salary survey they took, unfortunatelly it is only based on 864 respondents, not enough to be a true representation.

- many engineering programs focus on the theory to the exclusion of the practical. This is an excellent way to prepare students for graduate school, but a terrible way to educate practicing engineers.

Click to expand...

Right on target. This man speaketh the truth. I had to learn how to be an engineer after 4 hard years of a well-known university program. I sincerely wish they had taught me something about what I would later be working on, instead of concentrating on the "pie-in-the-sky" theories and laws.

I think the problem at most engineering schools is that the professors have no acutal engineering experience. Most learned ALL they know in school from other professors. It is much easier to teach Somebody-or-Other's Law, and how to derive all the equations pertaining to that Law, than it is to say, 'Okay, if your boss comes in and says, "We need a conveyor that will carry our product from Point A to Point B. It is your job to make sure it works." Then this is how you would solve that problem.'

It is worth repeating.

- many engineering programs focus on the theory to the exclusion of the practical. This is an excellent way to prepare students for graduate school, but a terrible way to educate practicing engineers.

Click to expand...

For the most part I agree with what is being said.


But as a professional engineer, I would disagree that someone with years of experience (no matter how good they may be) should not have an engineering title (or if they do, then it should be as an Engineering Technician).

But there is also a difference in engineer levels. Unfortunately, the HR and Financial people do not rate Engineering Techs as having Engineering abilities, therefore creating an incorrect sense of class.

There should be a sense of identification that one has credentials of higher education - even though the higher education is not practical skills but theoretical skills.

Let me additionally say that I personally have huge respect for the crafts and those that have risen through the ranks. There is a lot of knowledge in people eventhough they may not have a higher education. But like it was stated earlier, these are the people with the practical skills that are not taught at the university level.

This will probably stir the pot, but that is my 2 cents.

Hugh Jack said:

In summary there is a shortage of engineers with the following attributes, and when you expand the search pool to cover the globe you are more likely to find the right people (or at least better fits).
- an analytic approach to problem solving
- good decision makers
- willing to adapt to a new corporate culture
- a willingness to relocate and travel
- willing to work long/odd schedules
- willing to put a company/team ahead of themselves
- self starters and learners
- good communicators
- etc....

Click to expand...

Yes, and if you can do all that then why work for someone else that can't do all those things? The engineers that I know that have done very well own their own business and are usually system integrators. There is no sense in working long hours on a salary. It is bad for the health and some of the time working in the extra hours should be put into keeping up to date.

A system integrator has to be good and self reliant. When you are 'the man' there is no one to ask when you come across an engineering problem. You have to know how to solve problems yourself.

Add to the list that engineers should have the ability to do research and learn on their own.

Hugh, I know that professors are under pressure to publish something but I find that much of what the professors write it just the same old stuff that is repackaged in a slightly different way. My area of interest is control theory because I am into motion control. When I read about Kalman filters I often find the books disagree or are flat out wrong. I find that I have to read texts from 3 or more people to get a consensus and then work through the problems to verify what has been said. If the books had examples that proved the formulas actually worked it would be better. I have control books that are very good as long as one stays in the s domain. I don't think the author every really had to implement a design and get it the z domain or a difference equation so the control algorithm can be implemented in C. I have control books where the goal is to achieve some sort of control that no customer would accept. In other words the example is bogus. I have control books that neglect to say how to implement anything other than a regulator. They don't say how to implement a integrator, how to compensate for a zero. Why one should use difference equations or state space. They don't say what an observer really does. The books don't explain why the material is important. They don't say where all the nice number in the examples that come from. If the natural frequency is 1 and the damping factor is 0.7 ..... How does one know how to get the natural frequency and damping factor in the first place? I think this should be taught first because this is the first problem an engineer has in the real world. I have yet to see a machine with a transfer function stamped on it and yet the text books let the student think this is a given. My list goes on and on. BTW, I really use that stuff and more.

The poor engineering students pay a lot of money these days. I don't think they get their money's worth when they are taught the same old stuff the same old way. The engineering students are the customers. They aren't treated that way. More like cattle. On the other hand. When I was a clueless student I when through the motions of what a student was supposed to do but I had no idea what was really required. The result is clueless engineers that think the machines have transfer functions stamped on them and every control system is just a regulator.

I am sorry Jiri, I should have recognized your original article as Flamebait. I did a survey of all of the (ABET accredited) manufacturing engineering programs in the country a couple of years ago. The had similar hiring patterns, with the typical starting salary being $55K. The new graduates earning $40K are typically coming from technology degree programs (the subject of a sister study).

Jiri Toman said:

Jack,
Clearly you are not an impartial third party. You have a vested interest in telling us that all your graduates are immediately employed. I would like to see some supporting statistics rather than just your opinion.
Further more, kids out of school are willing to work for low salaries just to get some experience.


If you have not educated your graduates to solve problems then I suggest you give them their money back, being able to solve problems is what engineering is all about, without it you are not an engineer.


Of course this is what I am talking about. Pay these kids 40 000 per year and have them work 60 hours per week. It is very common for employers to demand extra hours for no extra pay!


Well my friend you have not worked in a Company lately. It's all about top grading and smart goals. You compete with others for the money allocated to salaries in the budget. All Companies promote team work but yet you are judged on your performance versus performance of others in the department and around the Company.

This is true, H1B visa engineers are not the best communicators but they are cheap. It takes only two days to fill in the quota of 65000 per year!
So the communication issue is not that important after all.
Besides, many professors at universities don't speak English very well (you are not included), but that doesn't stop schools like Purdue to hire absolutely ridiculous people. One of my friends went through three professors in one semester. The class had to stage a walk-out because the professor was teaching with an interpreter at his side. You are responsible for producing these poor communicators!


I think you meant to say "well educated people". Not trying to nit pick here, I know you what you mean. Thank you for admitting this.




I agree with you that the type of engineering and geographic location makes a difference, but that is all part of the standard deviation. The article refers to engineering and science as a whole, it looks at the bigger picture.

Here is a link to a very cool web site (not my own and I am not affiliated with it)

http://www.automation.com/sitepages/pid2965.php

It gives you results of a salary survey they took, unfortunatelly it is only based on 864 respondents, not enough to be a true representation.

Click to expand...

Nice horror stories guys. It makes me glad I can into this field with a BS in Geology and a life long passion of IS/IT. I ended up working IT after college, and after 2 years, the automation engineer quit, and I inhereted the job. 11 years, and 90 or so projects later, I am happy to call myself an "Engineer". I dont think the fact that I dont have a engineering degree makes me any less of an engineer.

matt

Originally posted by Jiri Toman:

If you have not educated your graduates to solve problems then I suggest you give them their money back, being able to solve problems is what engineering is all about, without it you are not an engineer.

Click to expand...

This is certainly not universally true. Individuals are graduated with a skill set to allow them to solve problems. Being the best math (or physics or ...) student in my class doesn't directly help me solve a problem. This is because the problems we encounter on a daily basis are not framed in equation form. They are a piece of physical apparatus or process that requires the problem to be framed in an alternate symbol set (read 'modelled') before it can be solved. Typically universities only touch on a specific field very briefly because they cannot touch on them all whith anywhere near thoroughly enough. So you get your skill set (tools, as Peter calls them) and then it is up you you to determine where to use them. I think this is the separation between good and bad referred to previously.

I really hate to get this argument going again, but I don't think it is so much that there are not any engineering jobs out there. I think it is more that companies are filling those jobs with non-engineers who have expertise in a very specific field or function.

Also, EVERYONE, even the Sloane Institute, has an agendy. To believe this is not true is rather naive.

Keith

WOW, Finaly the truth. I am a current Purdue Student studying Electrical and Computer Engineering. What we (students) have to put up with is a joke. All I want is a good education. I work hard, go to class ask questions and even turn in my work on time and completed. I have had professors refuse to answer my questions when I have made several attempts at solving the problem on my own, Professors that can not speak English, Professors that don't show up for class and fellow students that do not care about anything let alone getting a good education.

All that matters where I go to school is how much the university can charge in tution and the many extra expences such as Technology fees, Parking fees, Student service fees, and of corse Lab fees.

The sad part is that questining the management at the university does no good. The school pushes students through the classes -many unable to do basic math. Heck there was an instructor teaching a "statics" class who could not do trig. go figure.

I am a problem solver by nature and you are right engineering is problem solving profession. I see a problem as an oportunity to excel not a pain in the @##.

Both the Universities and the Students must make some changes if the engineering community is going to have any type of future.

I am a non-traditional student, I am 48 years young, begining my second career. I was a Co-Op at the company that I know have been hired at. I am very fortunate in landing my current position. I am learning how to be an engineer at my job NOT IN SCHOOL.

Peter Nachtwey said:

..
The poor engineering students pay a lot of money these days. I don't think they get their money's worth when they are taught the same old stuff the same old way. The engineering students are the customers. They aren't treated that way. More like cattle. On the other hand. When I was a clueless student I when through the motions of what a student was supposed to do but I had no idea what was really required. The result is clueless engineers that think the machines have transfer functions stamped on them and every control system is just a regulator.

Click to expand...

Wow, I agree with everything you said there. Actually, the professors know this as well but it's very hard to change the way things is done for so many years. During the 100 year ChemE get together at U of Washington, this very topic was discussed by the ChemE chair of UW and MIT but I'm not sure if anything came out of it. Oh, I think I did a tranfer function question in the PE exam but my memeory is fuzzzzy now.

As for the original topic which basically boils down to pay and respect. it's always been about supply and demand and the tendency for engineers not protecting their "turf" like other professions. I clearly sense some bitterness in the OP. To me, there's always a matter of choice, i bet less than 30% of my graduating class is still doing "engineering". An engineering education prepares one for many careers.

As for the manufacturing decline in the US, it's true if you only look at employment numbers in the sector. If that's the measure, than China is on a deeper decline than the US. I say blame it on the automation engineer.

Looking backing 20 years, entire industry which didn't exist now are the norm (Google, Amazon, etc). This trend will continue which means more and more service jobs. No sense crying about it.

harryting said:

You are competing with all the engineers in China, India, Eastern Europe anyway.
Speaking of computer, there's no shortage either but they just don't want pay for talent. Microsoft is bring a tons of Indian and Chinese programmers to Redmond.

Click to expand...

I dont think you are competing with Indian or Chinese engineers. Actually you are offloading the dull, boring and monotonous jobs to Indian and chineese engineers.
Also those Redmond Indian engineers just CTRL C, CTRL V the codes of which you(Americans) have become tired off.They are not writing the kernel of Vista.

There real dearth of good engineers, who actually knows the systems inside out.In my 5 years of industrial experience I have so far changed 3 companies but still I have not come across the person who knows the nitty gritty details of the system (Any). And thats the reason India dont have any world famous engineering company,and you guys have thousands.................

As I sat here debating with myself whether or not I should reply, I felt I would also add my feelings into this topic (Ron would be proud, even though he would argue with me).

I have seen Engineers come and go in my company that couldn't engineer their way out of a wet paper bag without the help of the Senior Maintenance Mechanics.!!!!

Click to expand...

Only 1 engineering bash this time, I would have expected more.

The problem is that getting an engineering degree is just the start and not the end of a process. Engineering is like everything else. Use it or lose it.

Click to expand...

Well said. I am amazed at how many people think that attending a university magically gives you all the knowledge. That certificate does not hold all the answers.

There is a real shortage of good engineers, good technicians and good electricians.
Regards Alan

Click to expand...

This I would have to agree with. The company I work for has been looking for an engineer for more than 1 year. We've had all kinds of people apply, many with only a couple of years of experience who want in the $60k to $70K salary. Compared to the techs, this is not unreasonable, but in our world, the majority of industries think of an engineer as "non-value added". As for technicians, we spend a lot of time training them. We are currently using ones from a local tech school trying to get some good ones. Sad thing is these 22 year old guys will be making $60K per year after 2 years. Maybe the 4 year degree wasn't worth it.

Oakley said:

But as a professional engineer, I would disagree that someone with years of experience (no matter how good they may be) should not have an engineering title

Click to expand...

AND

Matthias von Zorn said:

I ended up working IT after college, and after 2 years, the automation engineer quit, and I inhereted the job. 11 years, and 90 or so projects later, I am happy to call myself an "Engineer". I dont think the fact that I dont have a engineering degree makes me any less of an engineer.

matt

Click to expand...

Sorry to stir the pot but I fully disagree. Just because you think you do the same thing as what an engineer does - that does not make you an engineer or give you the ability to call yourself an engineer. If I pull a splinter out of my kid's finger, I don't call myself a doctor. An engineer is someone who completed an engineering ciriculum at an accredited college. Life experience does not count as credit. SORRY!!!


There is a shortage of experienced qualified engineers willing to work for what is currently being paid. I don't care what a magazine article says, I've seen the hundreds of resumes out there and talked to almost that many.

I also fully believe that there needs to be better education in college (with any practical teachings being extremely helpful) and a willingness among newly graduate engineers to spend time "in the trenches" learning from the guys who already know how to do it. When this partnership comes to fruition, then you will have a successful team. Both will benefit and learn.

Peter Nachtwey said:

Hugh, I know that professors are under pressure to publish something but I find that much of what the professors write it just the same old stuff that is repackaged in a slightly different way. My area of interest is control theory because I am into motion control. When I read about Kalman filters I often find the books disagree or are flat out wrong. I find that I have to read texts from 3 or more people to get a consensus and then work through the problems to verify what has been said. If the books had examples that proved the formulas actually worked it would be better. I have control books that are very good as long as one stays in the s domain. I don't think the author every really had to implement a design and get it the z domain or a difference equation so the control algorithm can be implemented in C. I have control books where the goal is to achieve some sort of control that no customer would accept. In other words the example is bogus. I have control books that neglect to say how to implement anything other than a regulator. They don't say how to implement a integrator, how to compensate for a zero. Why one should use difference equations or state space. They don't say what an observer really does. The books don't explain why the material is important. They don't say where all the nice number in the examples that come from. If the natural frequency is 1 and the damping factor is 0.7 ..... How does one know how to get the natural frequency and damping factor in the first place? I think this should be taught first because this is the first problem an engineer has in the real world. I have yet to see a machine with a transfer function stamped on it and yet the text books let the student think this is a given. My list goes on and on. BTW, I really use that stuff and more.

Click to expand...

I agree with this. When I started teaching 'manufacturing controls' (over a decade ago) I started with traditional books and theory. Obviously this did not go down well with our students who had as much as a year of industry experience because of their co-op employment. The classical control theory I was teaching was good to prepare them for following courses, but was virtually non-existant on the factory floor where most of them would probably be working. As you indicate, linear (and variations) are good ways to get grant funding. I decided to shift the focus of the controls curriculum to discrete control systems (PLCs). At that time I also looked for books to support an engineering level course, but only found one that was close by Filer and Leinonen. (It actually used state based design methods.) Over time I have developed my own materials to support a structured design approach that seems to do the job for preparing my students (http://claymore.engineer.gvsu.edu/~jackh/books/plcs/). This course culminates with projects (e.g., http://claymore.engineer.gvsu.edu/~koopsl/Sparky2.html) most of these are done for local companies (e.g., http://claymore.engineer.gvsu.edu/~rockerk/).

In an earlier modeling course we still cover the basic first/second order system models including basic linear control system design and numerical simulation (http://claymore.engineer.gvsu.edu/~jackh/books/model/). However, this material is heavily reinforced with weekly labs (http://claymore.engineer.gvsu.edu/~jackh/eod/egr345.html). The students use microcontrollers that they buy and keep ($35) to control the systems. The end result is that when they are done they are able to design controllers for consumer products.

BTW, this is for our mechanical and manufacturing students.

Peter Nachtwey said:

The poor engineering students pay a lot of money these days. I don't think they get their money's worth when they are taught the same old stuff the same old way. The engineering students are the customers. They aren't treated that way. More like cattle. On the other hand. When I was a clueless student I when through the motions of what a student was supposed to do but I had no idea what was really required. The result is clueless engineers that think the machines have transfer functions stamped on them and every control system is just a regulator.

Click to expand...

This is why it is crucial for people outside the universities and colleges to get involved with the process. Find schools that do what you like and work with them - drop off used equipment, mention them to local suppliers for donations, hire their graduates, talk to the faculty. If there are schools that are doing a poor job then complain. If you are an Alumnus your voice carries much greater weight. If you are a student and you have an instructor who can not communicate complain bitterly to the chair, the dean, the provost, the president, and anybody else who will listen. You will be heard.

As a faculty member I visit these forums because I am very interested in supporting industry, and keeping track of issues. There are many more out there who are also interested in working for progress. I hope you will avoid getting sidetracked by some of the anecdotes about terrible events.

Hugh

A little off topic but.

Hugh, it has been a while since I visited your site. I went to this link.
http://claymore.engineer.gvsu.edu/~jackh/books/model/
There is a lot of good stuff there but while I was looking through the the .pdf files I saw a example that I can relate to. I noticed an example of a mass connected to another mass by a spring connected to the or wall or actuator by another spring. The example shows how to model the system but not the control.

The question I would have is 'Why Bother'.
1. You can learn how to model two second order under damped systems.

But there is so much more.

2. You can learn how difficult it is to control the system. A simple PID will not do if the damping factor is low and a simple PID cannot place all the poles.
3. The mechanical engineer should know that designing such a system will make life difficult for the control engineer. The control engineer will likely feed the mechanical engineer to the machinery if the mechanical or hydraulic designer is around.
4. The mechanical engineer should weigh the cost of being fed to the machinery against the cost of a stiffer design and one with fewer poles. Systems with fewer poles are easier to control.
5 The mechanical engineer may be able to keep from being fed to the machinery if a transfer function is stamped on the machinery and the control engineer knows up front this project will be difficult.
5. The control engineer should be able to figure out by himself that that this system will be difficult to control and warn the mechanical designer and management that there will be additional costs involved. Having to find an new mechanical designer for starters.
6. The control engineer should know that a simple PID will not work and some specialized controller or perhaps a small embedded controller can be used to do the control. This will require much study to figure out how to move a fourth order system quickly and accurately. The documentation and future support will be an added cost.

There is more to this problem than just coming up with numbers.
The numbers mean something. Those that have a gut feel for how difficult controlling this system is will be miles ahead of all of those that can do the math and yet not understand the significance of their calculations. This one problem could take a whole class period discussing the all the options and costs.

Note, last spring I was tuning a system like that. There were two hydraulic cylinders. One mounted on the other. The bottom chipper was mounted to the base of the machine and the top chip was mounted to the bottom chipper. ( Saw mill ). A chipper is basically a mass mounted between two springs, the oil on either side of the piston. There was a note by the hydraulic designer in the spread sheet with the specifications that said "This axis will be impossible to tune". Fortunately for the hydraulic designer he was thousands of miles away.

Peter Nachtwey said:

Hugh, it has been a while since I visited your site. I went to this link.
http://claymore.engineer.gvsu.edu/~jackh/books/model/
There is a lot of good stuff there but while I was looking through the the .pdf files I saw a example that I can relate to. I noticed an example of a mass connected to another mass by a spring connected to the or wall or actuator by another spring. The example shows how to model the system but not the control.

The question I would have is 'Why Bother'.
1. You can learn how to model two second order under damped systems.

Click to expand...

That is an excellent question. On a practical level, there are a number of local companies that deal with passive systems, such as engine mounts that use that theory directly. And, it supports the students that go on to vibrations.

From an academic perspective I start with modelling first to show the students how to develop the equations. This is also a chance for me to go back and work on their math skills (from basic algebra up to diff. eqn.) You will also notice that I show them numerical methods so that they can predict system performance quickly. I put off LaPlace transforms to get to controls sooner.

Peter Nachtwey said:

But there is so much more.

2. You can learn how difficult it is to control the system. A simple PID will not do if the damping factor is low and a simple PID cannot place all the poles.

Click to expand...

While they are studying the material in class I have them doing a separate set of labs that work ahead of the lectures in building control systems. So when we get to feedback control they have already experienced at least one system that is potentially unstable.

Because this is an introductory course (normally it would only be system modeling) I routinely skip Root-Locus methods, but I do show them how to develop and implement controllers, as an example have a look at some of their programs (http://claymore.engineer.gvsu.edu/~volkersc/index.html).

Peter Nachtwey said:

3. The mechanical engineer should know that designing such a system will make life difficult for the control engineer. The control engineer will likely feed the mechanical engineer to the machinery if the mechanical or hydraulic designer is around.

Click to expand...

I have been trying to make the mechanical and manufacturing engineers able to do the controls work. After all, US electrical engineering programs focus on electronics, so the students are not prepared to work with electrical systems. Ironically my mechanical and manufacturing students help the electrical students when they do anything 120-660Vac.

Peter Nachtwey said:

4. The mechanical engineer should weigh the cost of being fed to the machinery against the cost of a stiffer design and one with fewer poles. Systems with fewer poles are easier to control.

Click to expand...

This is a tough one to answer - it is outside my area of influence. If I were to do a redesign of a mechanical program I would have more machine/frame design in integrated courses. Presently the courses are taught as separate topics and never really integrated.

Peter Nachtwey said:

5 The mechanical engineer may be able to keep from being fed to the machinery if a transfer function is stamped on the machinery and the control engineer knows up front this project will be difficult.

Click to expand...

There are a couple of issues here....
- There are no techniques (that I know of) for mapping a transfer function back to a mechanical system. (Although I recall a couple for electrical systems.) Moreover, a real system is fraught with non-linearities that make nice-clean techniques unlikely to arise. As a result the design proces becomes trial and error until there is a best fit. Or, linearization. I hate to say it but many undergraduates are not mature enough to make those choices - but they can be prepared.

- It is also a major chore to get the students to look at a system and identify the main components. They often get mired in the complexiy and are unable to negate effects and develop clean models. Trying to go backwards is another step up.

- Some undergraduates have a natural knack for joining theory and practice, but they are the exception, not the norm.

Peter Nachtwey said:

5. The control engineer should be able to figure out by himself that that this system will be difficult to control and warn the mechanical designer and management that there will be additional costs involved. Having to find an new mechanical designer for starters.
6. The control engineer should know that a simple PID will not work and some specialized controller or perhaps a small embedded controller can be used to do the control. This will require much study to figure out how to move a fourth order system quickly and accurately. The documentation and future support will be an added cost.

There is more to this problem than just coming up with numbers.
The numbers mean something. Those that have a gut feel for how difficult controlling this system is will be miles ahead of all of those that can do the math and yet not understand the significance of their calculations. This one problem could take a whole class period discussing the all the options and costs.

Click to expand...

Anecdote - I recently had a student who was out on a co-op assignment decide to use the material learned in the first modeling course. The controller was for a sheet steel roll feeder. They wanted to control the sag to flatten the sheet. They did use a PID (or similar simple controller) and even had to deal with things like current in-rush during startup.

But, this is only one coure in system modeling and controls. Eventually we will add a follow-on course that includes things like filtering, adaptive control, etc.

Peter Nachtwey said:

Note, last spring I was tuning a system like that. There were two hydraulic cylinders. One mounted on the other. The bottom chipper was mounted to the base of the machine and the top chip was mounted to the bottom chipper. ( Saw mill ). A chipper is basically a mass mounted between two springs, the oil on either side of the piston. There was a note by the hydraulic designer in the spread sheet with the specifications that said "This axis will be impossible to tune". Fortunately for the hydraulic designer he was thousands of miles away.

Click to expand...

Hey, at least the documentation was complete