I have an airplane and have done a gust response analysis - for a discrete vertical gust. I used the nastran example from chapter 9 section 1. How can I do a side gust analysis - gust 90 degree angle to gravity vector? Can you point me to an example.

Thanks again Sanjay,

My first thought was that I should orient the aero axis from the side ---- then I talked myself out of it – mistakenly. I confused myself into thinking that I had to also re-orient the basic coordinate system. That’s one of the bad things about working by myself.

I have to think through this, but I believe that if I do one run with a vertical gust, and a second run with a side gust, I should be able (by linearity) to construct a load case for a gust at any other angle relative to the flight path. I expect a square edge gust from any direction (of the clock) should work – similarly to the vertical gust case in 9.1. An issue to be determined is whether the gust needs to be reversed as for the vertical case, or whether some other shape is needed.

“Around the clock” gust loads is standard analysis for aircraft certification. I have used those codes, but have not created a load case myself until now. Using Nastran to do all the work is what I was hoping to do.

Thanks, I am going to ponder this a little more.

scott

Sanjay,

I thought about this a little more. The aero entry indicates the flight direction. Both a vertical gust and a lateral gust are orthogonal to the flight direction. Hence, a vertical gust changes alpha, and a lateral gust changes beta.

It appears to me that the Nastran gust capability is applied to vertical gust in example 9.1, and the resulting loads are due to 1g gravity loads plus the increment in the vertical direction due to the change in alpha due to gust.

For a lateral gust, the fight direction is the same as for vertical gust. To compute the result of vertical gust, I need to neglect the 1 g gravity load, and just compute the lateral load due to beta. This affects only aero surfaces that are not horizontal. So I believe my first plan for computing lateral loads in Nastran was on the correct path – the lateral gust along with the flight speed and direction defines the lateral load --- that is side force due to beta. Vertical forces need to be eliminated from the side force calculation, and so this side force must then be applied “by hand” by the user. I think that the user must apply the side gust force in a separate run, and save the time dependent side force with all vertical force removed. This side force is then dependent on time, and must then be applied to the relevant aero suraces as a time dependent forcing problem. The relevant aero surfaces are vertical fins, wing panels with nonzero dihedral, etc.

I do not think that Nastran does the lateral gust in a single step. Nastran does do the vertical gust in a single step.

Does this sound correct?

Thanks,
Scott this is my reply on Saturday.

For both the vertical and the lateral gust, the vehicle is being accelerated by the gust – so the full weight is part of the problem setup. So the gust load I would report to stress engineers would be 1g trim plus the set of around the clock (incremental) gust loads. Various locations of the vehicle will have different critical gust directions for the gust loads.


Sanjay – thanks for responding. The dialog is helping me straighten out my thinking.

scott