Mechanical Engineering

• entry
1
7
• views
503

# Strength calculation

1,547 views

hello All,

i am working as a design engineer and i am working on sheet metal structures. i am looking at the calculations to decide on the thickness of sheet metal to carry a particular load.

i calculated the sectional modulus of the design based on the cross sectional area of sheet metal and it is well within the sectional modulus of material. I would like to know werher the approach is right or are there other calculations which i should do before deciding the thickness?.

request to suggest on the above

please find the image below for reference

Hello! I would make a FEA analysis on the solidworks simulation platform... If you have a simple situation where the load is precisely known, your method can achieve good results, however, do not forget that some parts needs to be calculated for support a certain level of stress and some parts needs to present small deformation between certain limits that are essentialy dependent from design considerations.

halo
would like to speak about the field of
sheet metal forming
And questions in interview at a factory works in the field of sheet metal

thanks

16 hours ago, Haytham Fathi said:

halo
would like to speak about the field of
sheet metal forming
And questions in interview at a factory works in the field of sheet metal

thanks

On 3/3/2017 at 7:35 PM, Guilherme Lima said:

Hello! I would make a FEA analysis on the solidworks simulation platform... If you have a simple situation where the load is precisely known, your method can achieve good results, however, do not forget that some parts needs to be calculated for support a certain level of stress and some parts needs to present small deformation between certain limits that are essentialy dependent from design considerations.

`

On 3/3/2017 at 7:35 PM, Guilherme Lima said:

Hello! I would make a FEA analysis on the solidworks simulation platform... If you have a simple situation where the load is precisely known, your method can achieve good results, however, do not forget that some parts needs to be calculated for support a certain level of stress and some parts needs to present small deformation between certain limits that are essentialy dependent from design considerations.

we know the load values and apart from FEA i would like to know the basic calculation to evaluate the design.

@ Guilherme Lima : we know the load value and before taking to FEA i would like to know the basic calculation to evaluate the design. something like initial validation of design and i was asked about it in one of the interview

I would like to dispel the idea that FEA is required for every design. This is simply foolishness!! The USA has sent men to the moon using equipment largely designed without FEA (yes, I know about NASTRAN, but it was not used for everything by any means!).

There is no reason at all why this simple plate cannot be adequately analyzed using traditional mechanics of materials methods. We have well established methods in use for over 100 years, and they cost a whole lot less than FEA in both money, modeling time, and computer resources.

One observation I would make about the problem at hand is this: This is a plate bending problem, not a beam bending problem. Plate bending (in one axis) can be handled much like beam bending with some simple modifications. Look in an advanced mechanics of materials text (such as Timoshenko, Budynas, Seeley & Smith) under wide beams for more details. FEA is simply not called for here, and to jump to that simply shows incompetence.

DrD

Dr D. I used Seely and Smith 2nd Ed for an advanced strength of material class as and undergrad.  I need to review chapter 8 on plates to see how the results would vary from beam analysis. I usually refer to Roark's handbook for plates.

Back to your point on FEA. Every class, seminar, or article presented by someone not selling a particular brand of FEA, will tell you to verify the FEA results with hand calculations. If you think you can skip understanding the basics you will get into trouble.

On a particular online FEA course the instructor provides solved problems for free hoping to get you interested in purchasing the course. I recall a simple bracket example that from his experience you better know the approved hand calculation method. If you create an FEA model it would not put you in a good light with the Sr. engineers who did most work by hand before FEA was so easy to access.

On one project my hand calculations said a part would be over stressed. The FEA said it was OK and by a wide margin. If this happens to you, you need to be concerned.

The FEA was run for me by a friend who did not claim to do stress analysis or use the FEA that was embedded with his CAD package. The FEA set up was menu driven. I had my friend read to me over the phone each option as he stepped through the set up. I found where he went wrong and once corrected his results also showed an over stress. If you don't understand the basics you will likely not know how to set up an FEA model or have a gut feel for the answer.

In another situation we were working with an FEA package with very limited capability. The engineer was trying to model something similar to a 4 leg chair. The model's limitations would had only allowed him to fix all 4 legs to the ground. This would not represent a chair leg.

In another situation I found a different engineer inputting not only the applied loads but the reactions forces at the locations he should of been entering constraint conditions. He calculated the reactions by hand, not understanding how the FEA program works.

JAG, thanks for the useful real-life commentary. FEA has been grossly oversold as the way, the only way, to solve problems. Actually, in many cases, it simply creates more confusion.

On another forum recently I saw a question from a young engineer who had done a rotordynamic analysis using FEA. He got back the mode shape, in a normalized form. Not understanding modal analysis and the meaning of NORMALIZED, he thought he had gigantic deflections, 0.6 m or so, when that was not what the result was telling him at all. He just did not know what he was doing, did not know how to interpret the results the FEA gave him.

One of the biggest problems that I have seen over and over with FEA is the matter of boundary conditions. They are usually an essential element of the FEA solution, and yet it is often less than obvious just what they should be. Too much constraint and the structure becomes too stiff; too little constraint and it is unrealistically flexible. FEA is hardly the cure-all that it is thought to be.

DrD

×   Pasted as rich text.   Paste as plain text instead

Only 75 emoji are allowed.

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×

×
×
• Create New...