Static Analysis of Bolts on Multiple Surfaces

[jsed]

I am trying to analyze if the bolts in my design are sufficient to support the load on a shaft. Unfortunately, the bolts are not all located on one face, and though I know how to analyze each face by itself, but I would like to be able to combine them all into one problem. I have attached a rough sketch of the scenario for your reference, with each of the red dots representing a bolt location. Thanks in advance

 

[AZFIREBALL]

Can you provide us with some idea of the substructure the bolts are going through to hold the beam in place?

 

[jsed]

Can you provide us with some idea of the substructure the bolts are going through to hold the beam in place?

The bolts will be going through a fairly rigid structure of extruded aluminum framing, and the beam is also made of aluminum.

 

[jsed]

Yes, that is almost exactly right, except the bolts are only present on the side nearest us, the bottom of the beam, and the back face of the beam which butts up to the support plate. I tried to put arrows on your drawing to make it more clear

 

[AZFIREBALL]

The bolts will be going through a fairly rigid structure of extruded aluminum framing, and the beam is also made of aluminum.

Like this?

 

[BvU]

I know how to analyze each face by itself, but I would like to be able to combine them all into one problem

That invites the question: what did you do and what came out and why not just add up the three ?

Furthermore I am puzzled why post # 4 can be older than # 5 ?

Are the bolts all the way through -- as drawn in the side view it seems -- ? And they don't interfere with each other ?

 

[AZFIREBALL]

So the top and side bolts do not go through? Is there an angle on the bottom of the beam attached to the plate?

 

[jsed]

That invites the question: what did you do and what came out and why not just add up the three ?

As far as what I am doing to analyze each "face" of bolts, I am basically using the concepts outlined in this web page: http://www.roymech.co.uk/Useful_Tables/Screws/Bolted_Joint.html

That invites the question: what did you do and what came out and why not just add up the three ?

Are the bolts all the way through -- as drawn in the side view it seems -- ? And they don't interfere with each other ?

So the top and side bolts do not go through? Is there an angle on the bottom of the beam attached to the plate?

The bolts do not go through, they just screw into the beam itself, and they do not interfere with each other.

 

[AZFIREBALL]

So the top and side bolts do not go through? Is there an angle on the bottom of the beam attached to the plate

That invites the question: what did you do and what came out and why not just add up the three ?

Furthermore I am puzzled why post # 4 can be older than # 5 ?

Are the bolts all the way through -- as drawn in the side view it seems -- ? And they don't interfere with each other ?

I withdrew my post to fix it...then re-posted.

 

[jsed]

To expand on my thought process: I could use the current analysis methods I am using to analyze each face, but the supports from the other bolts not on the face currently being evaluated would have an affect on the face in question. Originally, I was considering treating the centroid of each face of bolts as an individual support reaction and then splitting up those forces among the bolts on that face, but I am not too confident that will give me an accurate representation of both shear and bending stresses in each bolt.

 

[AZFIREBALL]

Yes, that is almost exactly right, except the bolts are only present on the side nearest us, the bottom of the beam, and the back face of the beam which butts up to the support plate. I tried to put arrows on your drawing to make it more clear

No bolts on the top?

 

[jsed]

No bolts on the top?

Correct, no bolts on top

 

[AZFIREBALL]

Correct, no bolts on top

Is there a reason for this? They would do a better job of resisting the load than bolts on the bottom.

 

[jsed]

Is there a reason for this? They would do a better job of resisting the load than bolts on the bottom.

The beam that we have been discussing is more of a representation of an assembly of a few components. Due to the function of the assembly, I am not able to design it such that the bolts are attached on the top face.

 

[AZFIREBALL]

OK. Now we need to draw force diagrams for each bolt. Number the bolts and show us a diagram of the forces on each bolt as you see it.

 

[jsed]

OK. Now we need to draw force diagrams for each bolt. Number the bolts and show us a diagram of the forces on each bolt as you see it.

This is what I'm thinking. Each bolt provides a force in the same plane as the applied load, and each bolt provides moments that counteract the rotation of the shaft.

 

[AZFIREBALL]

This is what I'm thinking. Each bolt provides a force in the same plane as the applied load, and each bolt provides moments that counteract the rotation of the shaft.

Is the load off center of the beam as you show in the latest sketch?
Now we need to draw in-plane vector diagrams of each bolt's reaction to its load using the group centroids.
How are you handling vertical shear? (Should not be carried by bolts.)

 

[jsed]

Is the load off center of the beam as you show in the latest sketch?
Now we need to draw in-plane vector diagrams of each bolt's reaction to its load using the group centroids.
How are you handling vertical shear? (Should not be carried by bolts.)

Are you saying that I need to analyse the reactions of the centroid of each bolt group, and then analyze how each of the bolts contribute to that reaction?
As far as shear goes, the bolts will be handling the majority of the shear.

 

[AZFIREBALL]

Are you saying that I need to analyse the reactions of the centroid of each bolt group, and then analyze how each of the bolts contribute to that reaction?
As far as shear goes, the bolts will be handling the majority of the shear.

First question: Yes, that is right. Show directional vectors about the centroid and the resulting load directions carried by each fastener. Non-dimensional (Magnitude of load at each fastener location comes next)
Question 2: Is there an angle clip under the shaft at the attached end where 7 and 8 go through? That could carry the vertical shear load if the shaft sets on top of it.
Is the load off set, or centered on the shaft/beam?

 

[jsed]

First question: Yes, that is right. Show directional vectors about the centroid and the resulting load directions carried by each fastener. Non-dimensional (Magnitude of load at each fastener location comes next)
Question 2: Is there an angle clip under the shaft at the attached end where 7 and 8 go through? That could carry the vertical shear load if the shaft sets on top of it.
Is the load off set, or centered on the shaft/beam?

Thanks so much for your help.
To answer your question, there is an angled clip under the shaft to carry the vertical shear load. And the load is off set as well.

 

[BvU]

The beam that we have been discussing is more of a representation of an assembly of a few components. Due to the function of the assembly, I am not able to design it such that the bolts are attached on the top face.

Changing the rules during the game ? Or just moving the goal posts ? What is it you want from PF ?

 

[AZFIREBALL]

Does this better represent the structure?

 

[BvU]

Nice hand of drawing !
Bottom bolts seem pointless to me

 

[AZFIREBALL]

Thanks.
They are.

 

[AZFIREBALL]

jsed:

Other considerations:
If possible, bolts in tension should have at least two times diameter thread engagement.
If the bolt diameter is 1/2, the thread engagement should be one inch.
Use thick washers under all bolts in tension to reduce ‘pull-through’.
Where applicable, check each bolt connection for bearing failure and edge distance tear-out in its primary and adjoining structures.
Torque all fasteners to spec for type and size.
Consider the use of a head locking device or thread lock material.
Consider increase margin of safety for oscillatory (variable) loading, vibratory environment, aging in harsh environment or dissimilar materials.

 

[Baluncore]

The bolts do not go through, they just screw into the beam itself, and they do not interfere with each other.

Technically then, they are screws not bolts.
A bolt is tightened by turning the nut while the head and shank do not rotate.
A screw is tightened by turning the head, to turn the shank, while the receiving female thread remains in place.

Since screws tend to have a lower clamping force than boltsscrews, with lower friction between the surfaces being held clamped, the analysis comes down to shear of the pins = the shanks that are screwed into the section.

 

[AZFIREBALL]

jsed:

Other considerations:
If possible, bolts in tension should have at least two times diameter thread engagement.
If the bolt diameter is 1/2, the thread engagement should be one inch.
Use thick washers under all bolts in tension to reduce ‘pull-through’.
Where applicable, check each bolt connection for bearing failure and edge distance tear-out in its primary and adjoining structures.
Torque all fasteners to spec for type and size.
Consider the use of a head locking device or thread lock material.
Consider increase margin of safety for oscillatory (variable) loading, vibratory environment, aging in harsh environment or dissimilar materials.

Some additional considerations to add to the list above:
Design the fasteners such that it prevents any threads in bearing.
This is especially important for fasteners resisting a shear load.
Always have the full fastener diameter in the shear plane and the hole size and location as accurate as possible so that each fastener, in a group, takes up the load simultaneously.
Use the root diameter in calculating all tensional stresses.

 

[jsed]

Sorry I have been MIA for a few days. Just wanted to thank everyone for their help with analyzing the stresses, as well as for all the extra information that I will be considering moving forward

 

[JBA]

If possible, the bolt/screw shank should project across the connection interface in order to provide the maximum diameter for shear resistance.

 

[miltos]

It is very difficult to assess the contribution of bolts 1,2. If you want the structure to be rigid, I recommend to ignore bolt 1,2 and just calculate bolts 3...6.

The surface connected by bolts 1,2 develope some friction, which in general is small. In order to calculate that friction, you need to calculate (using FEM probably) the deformation of all compoments of the assembly.

If you want the bolts 1,2 to be stressed by shear load, and have a rigid consolidation you should use pins instead of bolts. But it remains a bit difficult subject to calculate the whole connection.

Best regards