Blast Design in FEA - what acceptance criteria

[EngineerMickeyMouse]

Hi All,

My question is in regards to blast design in FEA, e.g. using ANSYS Explicit Module.
I will create steel structural system consisting of shell/plate elements only, this means any girders will be simulated by few shells.
Loading will be transient with peak value simulating the blast.

Please let me know, against which code and allowable criteria would you check stress results values (for shell/plate elements from FEA) to satisfy any blast code requirements?

 

[JGard1985]

Hi again

Im waiting for my LS-DYNA explict run to followup.

In blast engineering you expect elasto-plastic response, so generate elasto-plastic material curves for steel. Its not about stress so much as it is about for deformation.

Since blast engineering is all about ductility, you want to measure/look at strain and make sure it isn't past the ultimate tensile strain of the material (rupture). In short look at effective plastic strain and compare to maximum allowable plastic strain of material.

Good primer on blast engineering is the US's Unified Facilities Code for blast engineering, has some hand examples and concepts.

Hope this helps
Jeff

Jeff
Pipe Stress Analysis Engineer

http://www.xceed-eng.com

 

[JGard1985]

Also forgot to add

Here is example of one of my past projects. Its a prefabricated blast wall panel and with of effective plastic strain in stiffening members

Finite Element Mesh
Finite Element Mesh in Prefab Blast Panel

Effective Plastic Strain
Effective Plastic Strain in Prefab Blast Panel

Deformation
Prefab Blast Panel Deformation



Jeff
Pipe Stress Analysis Engineer

http://www.xceed-eng.com

 

[IRstuff]

Performance of sandwich composites subjected to sequential impact and air blast loading might be useful

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[EngineerMickeyMouse]

Jeff, once more, thanks for helping me in two parallel threads, which are related to each other! Much appreciated.
Figures you have attached are clearly describing the example.


I have one question, lets please focus at plastic strain plot of yours. Let's assume 1.5e-02 is the max allowable value given in the code. Plot is a snapshot from the 0.075 time range, as such we know that earlier in time some smaller areas of your structure started to be red of strain values 1.5e-02. The question is how do you know how long / to what extent you may allow red areas to extend over your structure? Perhaps you are measuring also some angular deformation to be less than 2 degrees, or more probably you have some additional information from the code what length of the area with red values (1.5e-02) is allowed?

 

[EngineerMickeyMouse]

JGard1985 , I will much appreciate your response on the question above.

 

[JGard1985]

Glad to help.

In the picture you're refering to I clipped the contour plot of effective plastic strain (EPS) at 1.5E-02 (to get a nice looking plot). So basically I set the contours for the fringe plot. One locally stressed element had an EPS of 0.18, so my cropping of the contour limits makes the figure "look pretty" for the customer. It shows highly stressed regions in the framing members. We really weren't concerned with local yielding at corners of windows and doors. Also I don't remember if this was the final design (satisfactory?)

Regarding your question, your analysis needs to be strain dependent. For your analysis, determine maximum allowable strain and corresponding EPS. This is a function of material properties (Percent Elongation)

Then compare your FEA predicted strains and EPS against the maximum allowable strain.

Duration of the ductile response really isn't a factor as long as you have captured the maximum displacement of the structure (look for a rebound).

You can also check rotation but, I prefer to do this by taking some manual measurements and doing a little trigonometry.

Jeff

Jeff
Pipe Stress Analysis Engineer

http://www.xceed-eng.com

 

[EngineerMickeyMouse]

Thanks Jeff!
Am I right to say that your EPS value of 1.5E-02 (as per example) was lower than maximum allowable strain?
I thought EPS value was equal to maximum allowable strain, that is why I have asked what area size of the max utilized strain value is within the limits.

 

[JGard1985]

Yes my upper bound fringe plot threshold 1.5E-02 is lower than maximum allowable strain. Depending upon material your maximum allowable strain could be in range of 0.22 to 0.3. EPS is calculated from FEA and needs to be less thn allowable strain

Can you clarify what you meant by: I have asked what area size of the max utilized strain value is within the limits.

Thanks
Jeff



Jeff
Pipe Stress Analysis Engineer

http://www.xceed-eng.com

 

[EngineerMickeyMouse]

Certainly, the best will be to cite DNV non-linear code as this is not something I have invented, rather that is code clause which is based on experience and laboratory tests:

The strain limit for gross yielding reflects that real structures will include elements of inhomogeneity that
will not be accurately modelled in the analyses. This will mean that the strain measured over a long length
of a real structure will in average not reach the values that can be found in standardized tensile tests.
With gross yielding is meant that plastic deformations with strain above 2% are taking place over a zone
lyz > 20t in the direction of the maximum plastic strain.
The maximum gross yielding strain in any integration point, in any element within the yield zone, should
be limited to the gross yielding critical strain.

 

[JGard1985]

Oh I see where you went, DNVGL-RP-C208, right?

I don't design in accordance with this code (mostly design in accordance with US Army Corps of Engineers facilities design criteria and AISC).

Its pretty common in blast engineering to expect large gross yielding, so I haven't really checked against this requirement in DNVGL-RP-C208. I avoid irregular geometry (bolt holes) in my flexural members that could produce localized yielding or failure. I do check against an upper bound limit state (EPS) and ductility limits prescribed by the respective code.

Jeff
Pipe Stress Analysis Engineer

http://www.xceed-eng.com