Fatigue simulation by applying 4 separate loads consecutively

[sriveraariza]

I hope all of you be doing well. I got the next situation, so if anyone can help me that would be great.

We are wanting to run a Fatigue simulation with Nastran Module however our requirement is to apply 4 separate loads consecutively rather than concurrently per the picture attached?

We need to simulate Hf1 to the maximum load then release followed by Vf2 to the maximum load then release followed Hf3 to the maximum load then release followed by Vf4 to the maximum load then release. This is a single cycle. We need to replicate this cycle 2,000,000 to simulate the physical test. (The constrain are the 4 bolts).

This is the link of the component. https://a360.co/2Z75Gk3

I really appreciate your help, thanks so much.

Sergio.

[g.ceruti]

Hi @sriveraariza, I am not 100% sure,  but if Nastran uses the Miner rule to evaluate fatigue life, it is a sort of a linear superimposition of separete loads effects. 

So it doesn't matter the loads application sequence, ( first horizontal or first vertical load, neither first max or first min value).

I think you have to apply 2 subacases:

• Subcase1 for the horizontal load, with a load cycle ranging from minHf to max Hf (or viceversa)
• Subcase2 for the vertical load, with a load cycle ranging from minVf to maxVf (or viceversa)

Check the results, because Nastran2019 analyses only one subcase at time,even if your analysis has more subcases (I don't remeber if the first or the last one).  I don't know if the newer nastran versions have solved this issue.

[John_Holtz]

I would like to elaborate on @g.ceruti 's answer.

• Version 2019 does not handle multiple loads correctly. See the article "Multi-axial fatigue analysis with multiple loads not displaying multiple results in Nastran". (The issue with multiple loads was corrected in 2020.0.)
• The results of multiple loads are combined using Miner's rule (or the rainfall method) using linear superposition. The results can be different in some models based on the "order" of the loads since the calculation depends on the magnitude difference from one "peak" to the next "valley" in the plot of the stress history. (I did not study your load history to know whether the order makes a difference in your model.)
• Multiple loads (in a version that works) are not entered in different subcases. All of the loads are entered in one subcase. The load history curve for each load gives how the load changes with time. For example, the tables for the different loads would look like the following:

Time	Load Multiplier
	Load 1	Load 2	Load 3	Load 4
0	0	0	0	0
0.5	1	0	0	0
1.0	0	1	0	0
1.5	0	0	1	0
2.0	0	0	0	1
2.5	0	0	0	0

[sriveraariza]

Thanks you  @g.ceruti . I will keep that in mind, and I will try to do it 1 subcase for the horizontal force & 1 subcase for the vertical one.

Thanks, I really appreciate you opinion.

Sergio.

[sriveraariza]

Thank you John Holtz. 

I really like that idea to use 1 Subcase and using the 4 loads at once. However, I am not sure if my version of Nastran 2022 can do that. Let me show you attached an image I took from my Nastran Version, so you can see the "Table History Table Data", but I could not see the option "Time vs Load Multiplier". Would you mind showing me how do it with my Nastran 2022, please?.

Thank you so much, I really appreciate all your time and effort to help me.

Sergio.

[John_Holtz]

You found it. (Instead of "load multiplier" as I wrote, the actual text is "load scale factor". )