Hi Everybody,
I'm a beginner in CAESAR II,a self teaching one,but I do my best to understand the input,the restraints especialy.It is any manual,except the Application Guide,that contains examples with different type of supports and their modeling?
In regard of the restraints maybe you can help me to clarify the followings:
-to input an "U"-bolt,guide type with 2mm gap lateral and on top of pipe(with Z the axis of pipe)it is accurate to be modeled like this:
Restraint Type:X Gap:2mm
Type:-Y Gap:2mm
-to input an "U"-bolt,fix type,it is correct to be modeled like rigid anchor:
Restraint Type:ANC
Thank you for the reply.

Not published by COADE.

Almost, but "no". You would need this:
Restraint Type:X Gap:2mm this takes care of the lateral direction
Restraint Type: -Y Gap:2mm this takes care of the gap when moving up, but the node is free to move down - therefore you also need
Restraint Type: +Y this prevents the node from moving down

"No", that would fix all six degrees of freedom. Assuming the pipe in the "Z" direction, you would want two restraints:

Restraint Type: "X"
Restraint Type: "Y"

This would allow rotation about the three axis as well as longitudinal movement in "Z".

.........and as Rich alludes, a "U" bolt will NOT restrain any moments.

You may be interested in looking at this discussion:

http://www.coade.com/cgi-local/ultimatebb.cgi?ubb=get_topic;f=1;t=000925#000000

Regards, John.

John
U-Bolt will not restraint any moments ?
Im not sure, you sure ?
A simple example :
A Pipe 1000 mm length outside diameter 30 mm
Node 10 to 20 DX =1000mm
At location X = 500 mm (Node 15) is a U-Bolt welded at one plate (structur).
And now give at node 10 a Z-Force.
Is in this example no restraint moment Y at node 15 ?
That to mean this pipe rotate without reaction free ? Never John.

Hello Mr. Ohliger,

I hope I have the example correct.

Yes upon considering the example, I believe there are some scenarios where the "U" bolt, acting in concert with a structure that is in contact with the pipe for some significant horizontal distance would provide rotational restraint.

I believe that in most cases a single "U" bolt can only act on one plane - the plane on which the "U" bolt lays. I believe that when a single "U" bolt is used at a location it may provide only restraint against the three translational movements (and the restraint against the axial movement of the pipe depends upon friction and should not be depended upon to restrain very well).

In the example, if the "U" bolt is oriented on the Z-Y plane it will provide Y translational restraint (support) and Z translational restraint. It will provide some X restraint but only if the "U" bolt is installed very tightly and even then the restraint would be unreliable. If the "U" bolt is holding the pipe firmly against a structure (for example a vertical wide flange column, say a WF14) that is in contact with (along) the pipe for some significant horizontal distance, then the "U" bolt and the face of the column (that is in contact with the pipe) may act together to provide a moment restraint about the Y axis (but NOT about the Z axis).

If I saw a horizontal NPS 4 pipe supported by a horizontally orientated L3x3x3/8 angle (the angle orientated perpendicularly to the centerline of the pipe) with a "U" bolt going over the pipe and holding the pipe and the angle together (the "U" bolt being on the Z-Y plane), I would assume NO significant moment restaint would be provided by this combination of angle and "U" bolt. If however, I see the same scenario but with a horizontal WF8x31 beam supporting the pipe instead of the angle and with two "U" bolts (one bolted through the upper flange of the WF on each side of the web) then I would assume that the combination of the WF with two "U" bolts would provide rotational and translational restraint along and about the axes that were perpendicular to each other and perpendicular to the centerline of the pipe. I would assume NO translational restraint along the centerline of the pipe and I would assume no rotational restraint about the centerline of the pipe (I do not trust friction).

I suppose what I am saying is that there are many scenarios that could involve one or more "U" bolt(s) and each one should be judged individually.

Thank you for the "reality check". I admit I have negative feelings about "U" bolts.

Regards, John.

"The proof of the pudding is in the eating."

Most users want to build a model that will perform proper under any circumstance. Build it right and you don't have to "worry" about it; just go right to the output. This thread is a good illustration of the gnashing of teeth associated with this "right model" approach.

I believe, and promote, a simpler approach but it takes discipline. I think it best to take a good, quick shot at the model and allow the output to confirm or reject the input. Here, if I assume there is an axial restraint through the U-bolt (the moment restraint is a greater reach but follows the same tack), I would add the axial restraint and run the job. If the results (for certain load cases) show large axial loads, I would assume the model inadequate (for those load cases). If the load is low, no big deal.
The opposite is true. If I ignore the axial restraint in my model but I see large deflection (or large rotation for that moment restraint), then I might want to adjust the model.

In most cases, then, the anxiety over the "proper" model is greatly reduced and you can spend your time worrying about the really hard stuff.

The down side of this approach is the requirement that each of these simplifications must be evaluated before other results are examined. Many do not have this discipline.

Good recipies do not necessarily make good puddings. It's the results that count. Focus on the results.

Proposal for U-Bolt-type supports - here for horizontal pipe axis in x-direction

1. Guide
From 10 to 20, DX=1000mm, support at 20.
a) Prefered variant: Y, Mu=0.3; Z (or Guide), Mu=0.3; without any gap,
b) Alternative: +Y, Mu=0.3; -Y, Mu=0.3, gap=1mm; Z (or Guide), Mu=0.3, gap=1mm
Usualy I prefer variant a) because of normal precision of piping installation work on site and because of simplifying of numerical solution.

2. Fixed U-Bolt
From 10 to 20, DX=1000, support at 20.
a) Simple variant: X, Y, Z (or Y, Guide, Lim); without any gap, without any friction
b) Alternative I: Y, Mu=0.3; Z (or Guide), Mu=0.3; X2, K2=1, K1="blank", Fy=resistance force against pipe movement due to friction between pipe-beam and pipe-U-bolt from additional normal force of fixed U-bolt installation
c) Alternative II: Y; Z (or Guide); X2, K2=1, K1="blank", Fy=resistance force
If the simple variant model is not suitable I prefer the 2nd alternative variant.