Stress Engineer !! 3

[nidhee18]

Dear All,

I have one question related to the term used by many in the Industry and Job Market.

Many times I have seen particularly for Aerospace industries term called " Stress Engineer " How does it different then CAE Analyst?

Thanks
Nidhee

 

[corus]

CAE, I believe, stands for Crap Aeronautical Engineer, Stress Engineer isn't an acronymn and as such is a clearly understood term. That's the difference.

corus

 

[johnhors]

Corus,

Perhaps you could make it more general with "Completely Awful Engineer" !

I do agree with the point you are making, why replace a title of long standing recognition with a cheap tla (three letter acronymn), but then CAE which I believe stands for "Computer Aided Engineering" is probably very apt to describe a so called analyst who cannot sketch a free body diagram and label the applied loads correctly ! Or in simpler terms an "engineer" who is 100% reliant on computers and cannot perform a hand calculation.

 

[rb1957]

if the OP is thinking about replacing "Stress Engineer" with "CAE Analyst", I'd suggest that the idea is almost on the level of HR suggesting a new title for lead and chief ... "coach" ... seriously.

but seriously, "Analyst" is used sometimes to avoid the professional implications of "engineer" ... around here you have to be a licensed engineer to have a job title of "engineer".

 

[nidhee18]

The reason I asked this question is I have seen people from Aerospace industries normally use Stress Engineer but from Automotive Industries professional normally uses CAE Analyst..

 

[corus]

Being serious for a mo, I think that a CAE analyst is what was commonly referred to as a draughtsman but who now uses a computer. These days a TLA is the snappy thing to call them as nobody has the time to say a long word such as draught...
A stress engineer on the other hand will calculate and analyse stresses, sometimes with a computer, sometimes not. I believe it will be a profession that will soon die out though unless someone can quickly come up with a TLA for the job.

corus

 

[johnhors]

Corus,

I suspect it is already too late! Do you know of any genuine stress men under the age of 40 ?

 

[cbrn]

Hi,
Johnhors, with all the due respect, you're being offensive towards MANY "young" professionals who, first under the supervision of "old-school" mentors like the "true" stress engineers you say, and then by themselves, are trying to do the best they can in order to ensure that machines and structures which can VERY easily kill people will never do this in all their potential life.
In the office where I work, a part of a multi-national company, the engineering department is ALL made out of <40-old persons, and the results we achieve are acknowledged not only internally by the "old-and-experienced" gurus, but also externally by the customers.

That said, I agree that removing "engineer" from the significance of the acronym can be misleading, but probably all this is not intentional. A serious tech dept knows what it needs, call it CAE-analyst, CAE-specialist, safety-engineer (yes, I've heard this also!), etc etc etc...

Regards

 

[johnhors]

cbrn

I apologise for any offense. The point I was trying to make is that I do see some very clever highly gifted young engineers. However in general they have too much faith or confidence in what their computer is telling them. As a result designs tend to get very highly tuned or optimised with little or no reserve. I never see a quick ball park hand calculation as a sanity check (I'm as guilty as anyone for being lazy on this as well, though I used to do this I now trust experience to tell me if something ain't right).

One thing that catches just about every young stress analyst out (and quite a few old ones as well!) is when they are asked to calculate the normal modes of vibration using their model which was built and meshed to provide stresses. As very few people work in strict SI units, many of these engineers seem surprised that the natural frequencies are incorrect. A hand calculation using SI units is thus very valuable in this case.

Over reliance on computers has caused some very embarassing and costly problems, eg. the "wobbly" millenium bridge, the current A380 debacle. Then consider that Concorde was designed and analysed with minimal input from very primitive computers as was the Saturn V Apollo rocket.

 

[prost]

It would be impossible to argue that hand calculations and other sanity checks aren't a necessary part of any design process (personally I never trust anything out of a computer without running it through my 'nonsense' filter). However, those aerovehicles that have been mentioned, the Concorde and the Saturn V, had much higher margins of safety than today's designs. It would be impossible to squeeze today's safety factors out of current designs with hand calculations.

 

[crisb]

Just to add another flavour to the discussion; I remember similar comments being made by my elders about the use of slide rules and calculators.
No doubt their forbearers made similar comments about the use of log tables and abacuses.
Actually I find it much easier to make mistakes when I am doing hand calculations than using computers.

 

[rb1957]

crisb,
i'd spin that into "it's easier to find mistakes in hand calcs"

i've always likened FE to a loaded gun, but perhaps a loaded syringe is a better analogy; only this syringe has a multi-dose system ... you can select medicine (and with the right knowledge, you can select the right medicine) or you can select poison (particularly a poison that feels like the right medicine).

that being said, I don't think we could work today without FE, (there is a baby in the bath water?) we just have to be very careful with it.

 

[cbrn]

Hi,
Johnhors, I understand your point and I appreciate your fair-play.
Foundamentally, I agree with most of what has been said.
But I dissociate from the concept of the engineers being too much confident towards "FEA". A too much confident engineer will be so with every calculation instrument he might use. So I really do believe FEA is NOT the problem.
Rb1957 & al., wouldn't it be like comparing nuclear electric power plants with atomic bombs, if you see what I mean (in Italy so much demagogy has already been spent on this question that any Italian reader will immediately notice my polemics...)? Or also, does not a surgeon manipulate very dangerous instruments? But he saves lives, doesn't he?
If I was to calculate with an acceptable precision (nowadays we can't afford "wasting" material) an entire hydraulic turbine BY HAND, not only would it take me three years or so, but most probably I'd have to have the same calculations of mine checked by three or four other persons in order to be sure there aren't mistakes.
Anyway, I'm a bit diverging from the original discussion. Of course an engineer must know when he needs FEA to accomplish his task, and when he'd better use analyticals etc. Stated that anyway ALL THESE are simplified representations of the reality, NOT the reality itself. I personally prefer finding out analytical laws rather than employing FE for everything, but that's only my opinion. And I'm writing here only because I've learnt A LOT from other people and also because I made my "useful mistakes" as you could call them... A design error can be very embarrassing, of course, but it surely tells a lesson that it is hard to forget afterwards...

Best regards to all

 

[Matlsguy]

I would like to include that to be an FEA analyst you do not need a 4-year degree. You could train someone in CAE with about two years of schooling. Of course, that person would not understand the machinery of FEA or know how to perform the closed-form solutions of simpler models. Mechanical engineers that I know don't do closed form solutions prior to FEA work. This is where we are heading folks. Its cheaper just to hire a CAE person with a two year degree than a mechanical engineer with a 4-year degree- this especially true for big companies like P&W or GE. A mechanical engineer only needs to oversee the work.

I had a physics professor who did FEA EM field modeling once tell me that to be successful as an FEA person you need to know how to write your own codes or modify existing codes. I believe thats true.

Matlsguy

 

[GregLocock]

What you would have then is a mesher.

Cheers

Greg Locock

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

 

[cbrn]

... And probably even not, due to the fact that a good model needs an adequate mesh! So how would a CAE person be useful without knowing anything of the theory underlying? Hope the philosophy of using "blind executors" won't gain further credits, or two scenarios will open:
- either companies will use two persons for a single task
- either they will use the blind executor instead of the engineer (cheaper... :-( ).
This, however, is valid in any field, not only FEA and not only engineering.

Regards

 

[GregLocock]

We have mesh builders, who generate reasonable meshes of all parts from the solid models. These are then turned over to experienced CAE/stress engineers to build into models, refine and debug the mesh, and analyse.



Cheers

Greg Locock

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

 

[rtmpxr]

It's nice to see that engineering community retains it's sense of humour Now, seriously, I see a CAE engineer like that:
-design the product in Soliworks or Unigraphics (for example)
-import the model and do an analysis with Ansys or Abaqus (for example)
-conduct an experiment to verify his assumptions and validate the model (where is needed)
-submit the model to technical engineers to create the technical drawings
-do a "six signma analysis" based on new assumptions
-receive aknowlegments (a rare case)

 

[Qshake]

While this may be a bit off topic, I will tell you my secret.

For whatever reason, some years ago the company placed me in charge of the structural department. And when I find an engineer, young or old, who can't draw a FBD or at the very least predict the results of his computer analysis by way of magnitude and direction, I usher them to the Marketing Department.

If they find their way back, I may have them try one more time; those that don't make it back aren't missed and I hire a real engineer with the prerequisites I like.

Regards,
Qshake

Eng-Tips Forums:Real Solutions for Real Problems Really Quick.

 

[rtmpxr]

"If you can predict the values then you are aware of the phenomenom" said by JJ Thompson (not sure, maybe my memory let me down ) Important is the concept.