Bolted connector question -2020

[karthur1]

Trying to learn about the bolted connector.    In my simple assembly, I have a base plate and a bar.... bolted together with one bolt.  I used a bolt connector to fasten the bar to the base plate.  

When I run the analysis, the displacement of the bar (where the force is applied) is large.  So it is like its not constrained properly. Question is, do I have to apply a contact between the two parts or apply a pin constraint in the hole where the bolt connector is applied?

See attached files below.

Thanks,

Kirk

[karthur1]

Trying to learn about the bolted connector.    In my simple assembly, I have a base plate and a bar.... bolted together with one bolt.  I used a bolt connector to fasten the bar to the base plate.  

When I run the analysis, the displacement of the bar (where the force is applied) is large.  So it is like its not constrained properly. Question is, do I have to apply a contact between the two parts or apply a pin constraint in the hole where the bolt connector is applied?

See attached files below.

Thanks,

Kirk

[John_Holtz]

Hi Kirk,

I would help to put constraints on the model so that the 5000 lbf load does not accelerate the assembly to infinity. 😁

And yes, you need to define a contact between the plate and bar. Without the contact, the bolt preload cannot be resisted by anything. That is why you received the warning WARNING R4030: UNABLE TO MAINTAIN MINIMUM PRELOAD IN ONE OR MORE ELEMENTS. (Even if the bolt preload were 0, the results would be wrong without including the contact between the plate and bar.)

______________________________________________________________

If my post answers your question, please click the "Accept as Solution" button. This helps everyone find answers more quickly!

[John_Holtz]

Hi Kirk,

I would help to put constraints on the model so that the 5000 lbf load does not accelerate the assembly to infinity. 😁

And yes, you need to define a contact between the plate and bar. Without the contact, the bolt preload cannot be resisted by anything. That is why you received the warning WARNING R4030: UNABLE TO MAINTAIN MINIMUM PRELOAD IN ONE OR MORE ELEMENTS. (Even if the bolt preload were 0, the results would be wrong without including the contact between the plate and bar.)

______________________________________________________________

If my post answers your question, please click the "Accept as Solution" button. This helps everyone find answers more quickly!

[karthur1]

I am trying to figure this out.  Just to learn more about this, I wanted to create a model and see if I could get it to match my hand calculations.  I have attached my model files below.  The problem is that I cant get the two to agree with each other.  So I made this model (shown below) and this is what the simulation shows for the bolt axial force.

For the model example, my hand calculations show:

Bolt Axial force at 1: 35.5 lbs (vs -1.8)

Bolt axial force at 2: 177 lbs (vs 283)

Bolt Axial force at 3: 354 lbs (vs 522)

My hand calculations are included below as well as the model files. I just dont know what to do to make these match or how to explain the difference that I am seeing.

Many Thanks,

Kirk

[karthur1]

I am trying to figure this out.  Just to learn more about this, I wanted to create a model and see if I could get it to match my hand calculations.  I have attached my model files below.  The problem is that I cant get the two to agree with each other.  So I made this model (shown below) and this is what the simulation shows for the bolt axial force.

For the model example, my hand calculations show:

Bolt Axial force at 1: 35.5 lbs (vs -1.8)

Bolt axial force at 2: 177 lbs (vs 283)

Bolt Axial force at 3: 354 lbs (vs 522)

My hand calculations are included below as well as the model files. I just dont know what to do to make these match or how to explain the difference that I am seeing.

Many Thanks,

Kirk

[John_Holtz]

Hi Kirk,

If I understand your hand calculations, these are the assumptions and results that you are getting:

1. Assume the base plate (the plate with the box beam welded to it) only contacts the wall/foundation at the bottom edge. This enables you to sum the moments about the bottom edge and ignore the contact force.
2. Assume the base plate is perfectly rigid so that the result in each bolt is linear with distance.
3. Result: load in bolt 1 approximately 0. (35 lbs)
4. Result: load in bolt 3 = 2*load in bolt 2. (354 lbs/177 lbs = 2)

And these are the results of the analysis:

1. The base plate contacts the wall/foundation over the lower 16% of the plate, not at just the bottom edge. If doing a hand calculation, the moments about the bottom edge would need to take the contact force into account.
2. The base plate/box beam is not rigid.
3. Result: load in bolt 1 is approximately 0 (-1.8 lbs. Negative is possible because beams take tension AND compression. Real bolts do not transmit compression.)
4. Result: load in bolt 3/load in bolt 2 = 1.84 (= 522 lbs/283 lbs)
5. Result: the contact force over the lower 16% of the base plate requires a larger bolt load to support the combined moment due to the load and moment due to the contact.

Considering the simplification of the hand calculations, I would say the simulation results match pretty well.

	Hand Calculation	Simulation
Bolt 1 load	Approx. 0	Approx. 0
Bolt 3 load/bolt 2 load	= 2	= 1.84
Bolt 3 load	354 lbs	522 lbs (47% higher)

The hand calculations become more complex once you add a preload. This is why you do a simulation instead of depending solely on hand calculations! (And doing hand calculations as a sanity check that the analysis is correct.)

Good work!

[John_Holtz]

Hi Kirk,

If I understand your hand calculations, these are the assumptions and results that you are getting:

1. Assume the base plate (the plate with the box beam welded to it) only contacts the wall/foundation at the bottom edge. This enables you to sum the moments about the bottom edge and ignore the contact force.
2. Assume the base plate is perfectly rigid so that the result in each bolt is linear with distance.
3. Result: load in bolt 1 approximately 0. (35 lbs)
4. Result: load in bolt 3 = 2*load in bolt 2. (354 lbs/177 lbs = 2)

And these are the results of the analysis:

1. The base plate contacts the wall/foundation over the lower 16% of the plate, not at just the bottom edge. If doing a hand calculation, the moments about the bottom edge would need to take the contact force into account.
2. The base plate/box beam is not rigid.
3. Result: load in bolt 1 is approximately 0 (-1.8 lbs. Negative is possible because beams take tension AND compression. Real bolts do not transmit compression.)
4. Result: load in bolt 3/load in bolt 2 = 1.84 (= 522 lbs/283 lbs)
5. Result: the contact force over the lower 16% of the base plate requires a larger bolt load to support the combined moment due to the load and moment due to the contact.

Considering the simplification of the hand calculations, I would say the simulation results match pretty well.

	Hand Calculation	Simulation
Bolt 1 load	Approx. 0	Approx. 0
Bolt 3 load/bolt 2 load	= 2	= 1.84
Bolt 3 load	354 lbs	522 lbs (47% higher)

The hand calculations become more complex once you add a preload. This is why you do a simulation instead of depending solely on hand calculations! (And doing hand calculations as a sanity check that the analysis is correct.)

Good work!

[delaroca]

@Anonymous 

I don't get why you considered F3=10*F1 and F2=5*F1.

Is this a practical formula for cases like this? From what I see, this is a statically indeterminate problem.

[delaroca]

@Anonymous 

I don't get why you considered F3=10*F1 and F2=5*F1.

Is this a practical formula for cases like this? From what I see, this is a statically indeterminate problem.

[karthur1]

That relationship is from the assumption that the load on each bolt will be proportional to the distance from the pivot point.  The relationship can be calculated by solving for similar triangles.  Not indeterminate. 3 unknowns... 3 equations and solve.

I actually made an error that would change the values for the bolt load slightly.

F3=11F1 

F2=6F1.

[karthur1]

That relationship is from the assumption that the load on each bolt will be proportional to the distance from the pivot point.  The relationship can be calculated by solving for similar triangles.  Not indeterminate. 3 unknowns... 3 equations and solve.

I actually made an error that would change the values for the bolt load slightly.

F3=11F1 

F2=6F1.

[karthur1]

John,

Thanks for reviewing that. I just wanted to make sure I had things setup correctly in the simulation.

I know it would be simple enough to just calculate the bolt stress from the axial load, since I input the diameter.... but, I found this in the Plot settings. It appears that this would not only give me the axial stress, but the maximum stress as well at the connector.  Am I correct on that?

[karthur1]

John,

Thanks for reviewing that. I just wanted to make sure I had things setup correctly in the simulation.

I know it would be simple enough to just calculate the bolt stress from the axial load, since I input the diameter.... but, I found this in the Plot settings. It appears that this would not only give me the axial stress, but the maximum stress as well at the connector.  Am I correct on that?

[John_Holtz]

That is correct about the beam stress.

[John_Holtz]

That is correct about the beam stress.