# What kind of support should I use at the top of a landing gear for explicit analysis.

Member Posts: 1

Hello Everyone! I am new to the explicit dynamics module. I am doing impact analysis of a main landing gear as it lands at a certain velocity. The top of the landing gear is a beam of sorts which is supposedly attached to an aircraft which is not in the model. I am confused about what kind of support I should use at the top so that it represents its attachment to the aircraft and still have it move downwards to the platform/ground in the analysis. Also I want to apply a weight in the downward direction at the middle but someone told me I am making a novice mistake by applying a force and fixed support at the same beam and it will not contribute to the analysis. A picture of model taken during static structural is and the new model is attached. Can I use these settings for explicit dynamics? And is the force in the middle contributing to the analysis considering there is a fixed support right under it?

I will be very grateful. Thanks in Advance.

• Posts: 11,478Member

You want to simulate this landing gear as it touches down and supports the weight of an aircraft that is not modelled. I see a triad in the top picture, so it looks like Z is up, and X is forward, while Y is lateral.

Let's assume that you only care about the maximum stress during the first touchdown. Aircraft often bounce back up into the air after the first touchdown. You can make a model to show the bounce, but it is more complicated. If you only want to simulate the downward followed by some upward displacement and end the simulation before the craft moves above the initial touchdown height, that lets you build a simpler model.

Let's also assume that the rest of the aircraft provides the stability to prevent the landing gear from rotation about the X, Y and Z axes as well as any lateral motion.

If you were simulating a crash landing where the landing gear is going to be destroyed, then you would want to use Explicit Dynamics.

If you are simulating a normal landing, then all the stresses should remain in the elastic range, then you should use Transient Structural. Let's go with that.

Delete all the loads and supports from the model.

Select the top faces where you had the fixed support, but insert a Remote Displacement Support. Set X and Y components to 0., leave Z Component Free, and set all Rotations to 0. This constraint represents the airframe supporting all motion of the landing gear except for the vertical displacement. We won't simulate the forward motion of the craft in this model. In the Details window, set the Behavior to Rigid. Right click on this Remote Displacement and Promote to Remote Point.

Right click on the Geometry line in the Outline and Insert Point Mass. Set the Scoping Method to Remote Point and choose the Remote Point that was just created. Type in the Mass of the aircraft.

Select the hole where the wheel axle went labeled C above. Insert a Remote Displacement on that face. Set Z = 0. and leave all other DOF Free. Repeat this on the other hole. These two constraints represent contact with the ground.

On the Initial Conditions folder, insert a Velocity. Select the body and type in the downward velocity as a negative number in the Z Component.

Before you finish this Transient Structural model, it's best to know the Natural Frequency of this model. In Workbench, Drag and Drop a Modal analysis onto the Model cell of the Transient Structural. In Mechanical, drag and drop the three Remote Displacements from the Transient Structural to the Modal. Solve the Modal. What is the first natural frequency? Call that value F. Back in the Transient Structural Analysis Settings, set the End Time to 2/F. That means you should see the downward and upward motion of the landing gear in the simulated time.

When you mesh this solid body, you must have at least 2 quadratic elements or 4 linear elements across the thickness of the landing legs. A better model would replace that with a surface model and shell elements.

• Posts: 11,478Member

It should work in Explicit Dynamics also. The mesh is required to be Linear elements for that so you must have at least 4 elements through the thickness. The problem now is that the solution time goes up as the minimum element edge length goes down. Use the Mesh Metric Characteristic Length to find the element with the minimum value and try to improve the mesh to make that value as large as possible. Changing the legs to shell elements instead of solid elements will help.

Since Explicit Dynamics takes longer to solve, you don't need to simulate all the time after the landing gear reaches the maximum displacement downward, which is around the time F/4. So make the end time a bit longer than F/4 but that will be much less than F/2.

• Posts: 11,478Member

There is no fixed support in the Modal analysis. There are only the three Remote Displacements I described above.

F is the frequency in Hz, or cycles/second. A cycle is a sine wave. The period of the sine wave is T = 1/F.

A sine wave goes up then down and crosses zero at T/2 and has a maximum value at T/4.

I made a mistake above, I should have said the time was 1/4F not 4/F. You understand now I think.

• Posts: 5Member

Thank you Sir! I really appreciate all the effort you have put into this answer and it is a life saver. I especially found helpful the way you pointed out for specifying the mass of the aircraft. I have a question if you can answer as well I will be very grateful. If I have to simulate the problem in ANSYS ls dyna or explicit dynamics, do the conditions as specified in your answer change? Can you also give me guidance on that front as well? Again I am very grateful for your help.

• Posts: 5Member

Thank you once again! for your valuable time and advice. I shall try my best.

• Posts: 5Member
edited March 28

Sir! I was just conducting my modal analysis as per your advice. I was confused about where should i place the fixed support in order to get the said frequency? Also by the F/4 and F/2 in the last answer of yours, do you actually mean T/2 and T/4 calculated from the frequency? I am sorry if the question is novice, but please help me.

• Posts: 5Member

Yes Sir! Thank you