We have better tools, but? 11

[BillBirch]

Recent postings give me a great cause for concern about the lack of knowledge of fundamental principals.

One guy writes that he is designing an ROV and wants buoyancy explained.

Another is designing a pump and wants to know how to calculate the moment of inertia and then discloses that he plans to use three bearings - (indeterminat loading and difficult to align).

Yet another was designing a rotating welding manipulator handling loads weighing tens of tonnes, which was eventually red flagged due to the heat that the thread was generating.

In all cases the posters referred to their solidworks designs. Are we getting blinded by the sophistication of the software and forgetting that garbage in = garbage out applies with software, or indeed any system.

I only know for certain that the last example was from an unqualified, but perhaps over-enthusiastic kid, but I hope that the other two are drafters with a healthy curiosity. If so, I hope that they are given sufficient experienced engineering supervision so that they do not waste too much time, design something impractical, dangerous or all of the above.

On the other hand, if the other posters are qualified, it suggests the following.

- Quality of education. In the examples given, buoyancy is a fundamental of physics that should have been understood in high school. Machine element design has fundamentals regarding constructability that should have been learnt in college.

- Insufficient supervision or inappropriate tasking. Are senior engineers too occupied with management to give the younger guys the necessary mentoring, and related to this, are the young guys being asked to undertake roles above their experience. Similarly, are companies cutting costs by employing graduates only, thus avoiding higher pay rates. This is a common complaint on this website, and although it is heartening to see the younger guys shouldering the challenge, it is also a concern that they may not realise the exposure of their situation and could easily be thrown to the wolves.

- Inadequate recruiting practices. Not everybody was good in all subjects in college and you would expect that career paths would reflect particular strengths. I may be wrong, but I suspect that ptoficiency in CAD seems to be driving a lot of selections. Are the correct skills being overlooked by recruiters who simply see a candidate as a two-for-the-price-of-one find, someone who can draft and do a bit of engineering?
Regards,
Bill

P.S If anybody recognises themselves in the examples, no offence was intended. The purpose of this post was not to ridicule, but to highlight what appears to be unreasonable expectactions on our junior colleagues or associates.

 

[jmw]

I liked your last option for no other reason than that I think this happens all the time; badly drafted recruiting profiles.

Your example is a good one; preference given to candidates proficient in CAD.
Why?
The lack of CAD skills is easiest remedied with suitable training courses.

One company I was at gave preference to recruiting from its competitors. Why? because the recruits were familiar with the product and the market yet these ought to have been the skills the company could most easily impart itself to a new recruit.
What they ended up with were people who were ready to leave their last jobs for some unknown reason and brought the same problems to their new jobs.
If the marine industry was a particular target market then they'd have been better of recruiting someone with solid marine industry skills and knowledge and trained them in the things they didn't know.

Actually, I'm not sure it is a badly drafted recruitment profile I suspect its the total lack of one. They just have a particular vacancy and a standard skills set for the vacancy and recruit the person who most closely matches. It just happens that recruits from competitor have all the right dialogue.

Now I don't suggest this is the problem you are talking about Bill, but it is certainly a problem I've seen too often.

JMW

http://www.ViscoAnalyser.com

 

[zdas04]

I see blind reliance on computer programs as a key piece of what you're talking about. Pipeline models are a good example. The underlying math is a bit complex and the interactions between adjacent pipe sections can be very complex. When I ask client's "how did you pick" a pipe size, route, or compressor, too often I hear "that's what I got from the model". People use pipeline models to extrapolate without ever verifying that the model can adequately represent current conditions (generally called "calibrating" the model). One kid asked me to review a design (at the insistence of his non-engineer boss) and he got really offended when I asked to see his model. Good thing I asked because he had connected the pipe wrong and was going to spend close to $2 million looping a line that didn't have a problem. He was astonished that the model could give him the wrong answer.

When I used to write computer programs (in the 80's) we had a saying "if it's on green-and-white, it's right" meaning that no one questions computer print outs. The attitude was really common then and it seems to be just a common today.

David

 

[Lion06]

I think blindly accepting computer output is a problem, but I don't think it's possible to get away from it (i.e. there will always be some dopes in every profession....... yes, including engineering).
I don't think, however, that it is indicative of the profession as a whole or the quality of education of engineering.

 

[zdas04]

I don't think that blind faith in some "authority" to whom you can pass the responsibility is in any way new. What concerns me is that the computer programs have gotten so slick and so capable that people don't feel that they need to personally verify that the program does the arithmetic properly any more. When I started, input to a program was on punch cards and output was a stack of paper with graphics that were a small step up from cave drawings. Good engineers scrupulously questioned every result because the interfaces were so crude.

Today even very experienced engineers are lulled into a sense of security by the pretty interfaces. I was recently guilty of this myself. I got a new (to me) pipeline program and was just too busy to evaluate it properly until after the 90 day trial period expired (the software had been on the market for 30 years in both mainframe and PC versions, the one I bought was something like version 14 so all the bugs must have long since been exterminated, right?). When I did get around to trying to use it I found that it had three show-stopper busts in the arithmetic that made it totally inappropriate for the task I was trying to accomplish. Luckily I'm enough of a curmudgen to check for reasonableness in the answers. When I brought the problems up to the programmers (they call themselves "software engineers"), one of them said that he knew about the problems and that I could work around them by "tweaking ..." (some seemingly unrelated parameter). If you don't know to ask you'd never find the quirks.

Blind faith in some "competent authority", be they a grey haired engineer, a text book, or a computer program is truly a bad way to do engineering, and always has been.

David

 

[MikeHalloran]

The experienced engineers can't possibly oversee the results produced by the enfants driving the zoomy CAD seats.

It's not because they can't, or don't want to, or are too busy with other stuff.

It's because they're not there.

More than one outfit paid for the expensive software by discharging its most expensive people. (e.g., me)

After all, if the software has "Engineer" in its name, who needs actual engineers?





Mike Halloran
Pembroke Pines, FL, USA

 

[SomptingGuy]

zdas04:

Point taken.

I work on the solvers behind the pretty interfaces that our customers (you, etc) demand. The GUIs have to be so simple that idiots can (and do) use them. What should we do? Text input files or GUIs?

- Steve

 

[BillBirch]

Mike,
You've probably noticed it worse with software that does not produce a result per se, but "tests" a configuration, such as pipe stress analysis programs. I don't know how many arguments I have had with project managers who maintain that "the budget allow allows one run per line", which assumes very good, experienced and intuitive piping desighers.
As a client rep, I once agonised over a young stress guy trying to reconfigure a relief valve inlet line from a vessel that kept "failing" due to a relatively stiff outlet line. After four or five runs he proudly showed me his solution - another three or four metres on the inlet line to incorporate an expansion loop! I told him to increase the vessel nozzle by one size and increase the flexibility downstream where pressure drop is not critical. It worked.
The stress guy was unaware that he also had pressure drop (or other) considerations when altering pipe routing.

 

[jmw]

Zdas04 has put a finger on the flaw in all this... some of these programs are written by software "engineers" not engineers.

So one assumes they had competent engineers looking over their shoulders all the time?
Yeah, right.

JMW

http://www.ViscoAnalyser.com

 

[MadMango]

I think it is a combination of inadequate tasking and recruitment policies, but it could also be a difference in generational culture. I know plenty of 20-somethings that will not ask those they work with for assistance, for fear of being labeled as a failure, or incompetent. For them it is better to ask an anonymous audience for help and look good. It could also be the curriculum being taught. I am sure if you compare today's curriculum with that of 10-20yrs ago, you will see a lot of subjects combined with others. When that happens, things are glazed over and not taught with any depth.

I have seen my fair share of "the computer said it was all right," and I am guilty of it myself. I've learned from my mistakes and experiences, hopefully so will those that are doing it today. As for a lack of mentors (e.g. MikeHalloran), I don't know what much can be done about that, except for fighting for those individuals to remain. It is always a shame to see anyone being forced to leave simply to get the ledgers back into the black.

"Art without engineering is dreaming; Engineering without art is calculating."

Have you read faq731-376 to make the be

 

[desertfox]

Hi BillBirch

I agree that to many people now solely rely on the computor for answers without any rough hand calculations to check or estimate that the answer from the computor is correct.
But the more worrying thing from my experience with young graduates is that they have no feel for what they are trying to analyse and even worse some of them don't even understand the very principles that they have been taught in their degree ie:- laws of motion, static equilibrum, basic maths ie one couldn't work out the length of an arc traced out by a lever within a mechanism its unbelievable!
( in fact I think some of the graduates got their degree free with 3 tops off Kelloggs Cornflakes packets!)
The other trend I see is that a lot of them can model very skillfully in 3D but struggle to understand correct dimensioning of components from the point of view of fit and function and mention geometrical tolerancing and they think it is a foregin language. Sadly what now happens at a lot of companies is they employ graduates to do the 3D modelling and retain retired engineers to mark up and correct the 2D drawing so that it can at least be manufactured.
The problem stems from the system getting rid of apprenticeships as I believe that the students don't get sufficient training in draughting, tolerancing etc and then they are thrown into the workplace straight from university and a lot of companies don't have graduate training programmes to help they cover the area's they are weak on.
That said when I see graduates that can't do simple maths that one should have learnt at school maybe the education system as a problem too.

 

[zdas04]

StompingGuy,
The only pipeline program I ever use for design has a DOS file interface and I use KEDIT to build and change files. The output looks like a 133 character lineprinter output. It works very well and is the reason I won't be going to Vista. I've verified the math on dozens of systems before using a model to recommend a system change. I don't see anything wrong with not spending money on the pretty, idiot resistant interfaces.

But I'm a dinasaur that can't be long for this world.

David

 

[SomptingGuy]

"Stomping"? One day I'll change my handle to something people can get right ;-)

- Steve

 

[zdas04]

Sorry about that, I've been seeing your handle for yeara and saw a word that i thought I knew. It has all the letters of "stomping". What is a "Sompting"?

I'll not make that mistake again.

David

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The main difference beteen humans and apes is that we have cooler tools

 

[SomptingGuy]

David,

Sompting is a village on the South coast of England, between Worthing and Shoreham-By-Sea. It's where I live. Simple really.

I have a good idea for a prank to play involving our village sign though...

- Steve

 

[ivymike]

so Sompting is your Stomping Grounds?

 

[TheBlacksmith]

It's not cut and dried. Where I work is certainly guilty of blindly looking at resumes and hiring quantity while letting quality go, but then a summer intern caught a mistake by an experienced engineer before it was implemented - maybe the exception that proves the rule.

 

[ivymike]

I suppose I have to jump to the defense of the younger engineers. It seems that if you want to find anyone who can double-check a calculation, you have to find a fresh-from-school kid or summer intern, because most of the so-called experienced engineers are only experienced in eating dinner with suppliers and releasing parts. They prefer to rely on the opinion of "experts" rather than understand the math themselves... the experts are the few in the org who used and built on their education instead of letting it rot, plus those highly-polished supplier experts and consultants who have an answer for everything. Even the simulation and analysis guys seem to rely on procedures more than "does this answer smell right."

Experienced engineers seem to be more likely to believe in alternative medicine, gas-saver gizmos, and get-rich-quick schemes, while a first-year graduate or a business major working in marketing would say "bull*" before the peddler got thru the first sentence. The more experienced the engineer, the more confidently he'll argue with you that his entirely unfounded beliefs are right. Heck, a little reading around here will turn up plenty of examples.

 

[graybeach]

I have a shareware dos cogo program that I swear by. You have to draw your own layout and lable the points by hand. Then you have to type up the commands one by one in a batch file. You know what's good about it - one person can prepare the input and someone else can easily check it. What a concept! How can you really check geometry that was generated in cad? You really can't properly, and there is no factor of safety on geometry.
zdas04-Is it not possible to run dos programs on vista?

 

[TheTick]

It's getting harder for engineers to stay in engineering. Seems like we all get pushed into project management (usually with "remote resources"). I've gone to contract work so that I can keep myself in design in spite of my degree.

SolidWorks/COSMOS is probably one of the biggest offenders in the "Anyone can do this" camp.
<

http://esoxrepublic.com/blog/2008/06/13/learning-cosmos/>