Will a 304 SST Plate Bend Under a 460 Lbs Load at a 21 Degree Angle?

[fishburn7]

So I have a plate connected to a metal wall that is riveted to the wall with 5 rivets. This plate then has a ball joint connected to the one end which has a load of 460 Lbs. at a 21 degree angle. All the drawings have been attached.

I need to prove that the plate will indeed bend under such circumstances and then decide what thickness the plate needs to be to prevent bending.


Thank you!

 

[nvn]

fishburn7: Why would the plate bend about the dotted line? I am currently not following that yet. Could you elaborate? The plate will want to rotate about the centroid of the rivet pattern, thereby loading the rivets in shear. But it currently appears the plate would not bend about the dotted line. The plate will also want to twist at the edge of the underlying wall, due to the ball joint offset. Where is the edge of the underlying wall located? Also, why is there a curved line underneath the 2.5 dimension?

 

[fishburn7]

The curved line in the drawing was a mistake on my part. it means nothing. I do not know why the plate bends at the dotted line, I was given the actual part and that is where it bent. As far as the wall it's connected to it seems that the left and right side of the plate are nearly flush with the wall, whereas the top and bottom are feet from the edge.

 

[nvn]

I was unable to understand your explanation of the underlying wall and underlying edges/supports, so far. Could you provide a sketch? Also, are you sure the dotted line should not be drawn from hole 6 to 4, instead of hole 6 to 1? Can you include a top view of the plate, and show the deflected shape?

If the dotted line should be drawn from hole 6 to 4, and you break rivet 5, then the plate might tend to buckle (bow out) laterally at the dotted line.

 

[fishburn7]

I'm not real sure what you mean by underlying wall but I have attached the partial assembly drawing and the actual plate drawing with a top view. I'm sorry about the upside down top view

 

[nvn]

fishburn7: I think you (or someone) misinterpreted where the dotted line is located. Why are you sure the dotted line is from hole 6 to 1, instead of hole 6 to 4?

 

[fishburn7]

That is where the actual part bent

 

[nvn]

What are you referring to by the word "that" in post 7?

 

[fishburn7]

Sorry, the dotted line is where the part actually bent.

 

[Daever]

oh man, I just finished typing up a big post, but it time out, so I lost it (happens all the time here, I must type slow...)

Anyway, I can't offer any direction on the calculation side of things, because its quite intensive, but it all revolves around buckling loads the part can take, nothing too simple. I failed to see how it buckled where you indicated - I take it the dotted line is where it creased?

I do have some practical suggestions though.

- use much thicker plate, even if its overkill.

- if you want to use the same size plate, put in more fasteners, closer to the lug where the ball joint is attached.

- if it's possible, orientate the plate facing the acting force.

- weld on some stiffening plate co-linear with the force, perpendicular to the plate, kind of like a web plate.

- make sure the surfaces where the plate contacts the wall is clean and paint free.

-put a fillet weld around the plate if the wall is metal

Goodluck :)

 

[nvn]

fishburn7: I am not yet confident that you have not drawn the problem upside down. How do you know, with certainty, that the dotted line is from hole 6 to 1, instead of hole 6 to 4? You did not yet give any compelling explanation or evidence of how you know, with absolute certainty, that the plate is not upside down in your first diagram. Please explain.

 

[fishburn7]

When I say the part bent along the dotted line I mean the dotted line is where the crease of the bend was. If this is not clear it may help to ignore the dotted line all together.

 

[gmax137]

oh man, I just finished typing up a big post, but it time out, so I lost it (happens all the time here, I must type slow...)

Offtopic:
I have the same problem. Sometimes when I know I have been slow in composing the reply, I will highlight my reply, and "ctrl-C" before I click on the 'submit reply' button. That way, If it disappears I can log back into PF and paste the text into a 'new' reply.

I guess the PF system doesn't know you're typing and it just times out the session after some interval.

 

[dave319]

Daever: Can you point me in the general direction for the calculations? I know you say it is intensive but I'm trying to determine bending in a gusset similar to this question posted by Fishburn7.

If anybody can point me in the right direction for this type of "plate bending" I would appreciate it. All I have found so far are calculations for a load "onto" the plate not "along the edge" of it.

Thanks