Weight to pressure calculation 1

[jt2001]

I am struggling to see the source of my error in this simple simulation

I am trying to verify the stresses on a fabrication when a 30 Ton generator is set down on the indicated areas. There are 4 supports assumed to be fully fixed at this stage in the simulation.

My calculations so far:

- For gravitational force of the generator

30000 * 9.8 = 294000N​

- The area of each pad is 0.27125m^2

4 x 0.27125 = 1.085/^2 the total area on which the 30 tonnes is to be distibuted​

- Therefore the total pressure on these pads should be

294000 / 1.058 = 270967.7419Pa​

- My FEA solver allows formula inputs on the face pressure. So to add face pressure proportional to the area of each element face I use the input:

(A/TA)*(294000/TA)​

Where : A = Area of face, TA = Total area of all faces to share the load​

- When I apply this the sum of all pressures match my calculations closely.

But the results of these calculations do not seem to show a realistic output with reaction forces on the restrained nodes extremely low (approx 800N total)

I suspect there is a glaring error here that I am unable to identify.

Kind regards,

JT

 

[TWilkinson]

I would first double-check your applied load in post processing... to verify the total applied force is actually equal to weight. It looks like your formula input is 1/4 (67,742 Pa) what you're attempting to solve for.

Think of the pressure not as the total force over total area, but as each support's force acting on its respective area. Assuming CoG is centered, the force acting on each support is 1/4 the total weight. Therefore, the applied pressure on each support surface is 270,968 Pa.

 

[rb1957]

you say "formula inputs for face pressure", but you're applying the constant pressure you've calc'ed, right ?

your code should provide checks, like load applied, and load reacted (at the constraints).

Nodal reactions may be low, if there are a lot of them.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[IDS]

Therefore the total pressure on these pads should be
294000 / 1.058 = 270967.7419Pa

That is the pressure that should be applied to each face with applied load. If the total area of the faces = 1.085 the total applied force will be correct.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[FEA way]

Can't you create surface partitions in the CAD model so that the pressure is applied as a total value on each pad, without having to account for the number of element faces ? Then you could also try applying this load in other ways such as total force, distributed mass (if supported by the software) and so on.

 

[rb1957]

what FEA code are you using ? It seems we don't follow your calculations ... why calc the force/pressure per element ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[IDS]

what FEA code are you using ? It seems we don't follow your calculations ... why calc the force/pressure per element ?

It looks like Strand7.

I don't think there is any mystery. He has scaled down the pressure on each element in proportion to the area, as if it were a force, and as a result the total applied force is way too small.


Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[jt2001]

Yes IDS although I cannot see why the force is so low. only approx 860N total on the Z axis reactions. Is the method i have used to scale the force incorrect?

Yes rb1957 it is strand 7.

 

[IDS]

jt2001 - You are applying a pressure, not a force. You calculated the applied pressure correctly as Total force/Total area, and that is the pressure you should be applying to each of the loaded faces.


Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[jt2001]

I understand. No adjustment for area is required in a pressure input.

Thankyou. That is the glaring error.

 

[IDS]

jt2001 - as a follow up, I can recommend the Strand7 training courses, if you get the chance.


Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[jt2001]

I will investigate. There is clearly significant capabilities that I am not taking advantage of.

 

[jt2001]

Plausible results.