Hello all,

I have a query regarding ductile iron push on joints like the one shown in the following link:

http://www.pscipco.com/tyton_joint.html

Now I know there are lots of discussions regarding slip joints, thrust forces etc but I am still having trouble specifying this joint even after reading them all. They have certainly helped me though.

The model I have is that of buried pipe, with thrust blocks at each bend, with settlements of 25mm except in one place where there is no settlement, so the pipe is angled down in that section. The fluid temperature is ambient so there should be no expansion effects.

I have run the buried pipe modeller and have put displacements (dy=-25 and 0 for all the rest) at each bend, however I have not modelled the push on joints, and I get failures in the sustained case, and very high moments at the thrust block, which I believe are due to the pipe not being able to bend at each joint in the model.

So here are my questions:
1. Does the slip joint as detailed on page 5-21 in the Caesar II App Guide allow rotational movement at each joint as well as translational? I'm assuming that it does as no rotational restraints have been imposed. If so, is this the appropriate model for a push on joint and how would you normally go about estimating/sourcing stiffness and joint friction data?
2. Most vendor data I have seen for this joint specify the maximum deflection allowed at each joint (refer link above), is there some way of telling Caesar that the node for each joint can rotate up to that maximum angle, rather than having to specify stiffness. Or does the deflection angle depend on the joint frictin thrust (which is not specified in vendor data).

Anyway, I would really appreciate your help, I haven't modelled buried pipe before, and am still on my training wheels at work, so please bear that in mind when you're cringing.

Thanks.

1a) Yes the joint on page 5-21 does allow rotation.

1b) The best source of stiffness and friction data for this push joint is the Vendor. Ask them how this joint should be modeled in a flexibility analysis.

2) The only way I can think of to limit the rotation is to put a rotational restraint at these locations with a gap (in degrees) of the appropriate size. Once the rotation reaches the specified amount, things should lock up. (Test this on a small model first, and satisfy your self that this technique behaves as you want it to.)

Thanks Richard, after some thought, I think I need to specify ecah joint as a combination of the two.

I need to specify the expansion joint to model the movement due to the thrust force impact at each joint, and then I need to restrain the rotational movement to ensure each joint does not rotate (due to thrust forces and elevation differences between anchors) more than the vendor's specification.

In that case, does that mean that at each push on joint I need to apply a force due to thrust?

If I can just jump in here...

I think of it this way: if you get inside the pipe and look upstream, the surfaces normal to your line of sight are the surfaces that are axially loaded by pressure - it could be an elbow, the annulus of a reducer or bellows convolution, a blind flange or a closed valve face. These points are where the thrust load will act to produce structural response. Do the same downstream.

If you apply at all push-on joints, most of these will just put the pipe into compression (with no resulting "structural-load- causing" response) and be equal along the way. Only the outer loads will "load" the system to cause strutural response.

The added annular cross section of the bell and wall thickness of the spigot would be thrust surfaces if you want to go to that detail. But...

Thanks for your help guys.

I have emailed the company supplying the pipe and I don't think they will be able to provide me with any friction data on their joints.

Based on page 2 of the document linked below I have assumed the joint is frictionless.

DIPRA Thrust Restraint document



So I have modelled it like the slip joint on pg 5-21 of the applications guide, however I have not modelled an axial stiffness/friction.

Anyway, my supervisor is back in a few days so I will run it past him.

While I think it is possible some sizes/fits of push-on joints for ductile iron pipes might contain some degree of “stiffness” (as a result of tight joint fits and/or internal restraint features etc.) particularly when substantially deflected, for modeling purposes it may be generally safest to model such joints as in effect structural “hinges” allowing substantially free rotation. In ductile iron pressure pipelines, it should be noted there will also be a lateral thrust caused by internal pressure actions at each such deflected joint/hinge location that must be counteracted in some fashion so as not to become over-deflected or separated (see e.g. the advisories requiring backfill of pipes intended to be buried in the previously mentioned DIPRA thrust restraint literature and also AWWA C600, and also requirements for external stabilizing means such as strap anchorage or bracketed roller anchorage near any such joints etc. in exposed, aboveground pipe-on-supports applications e.g. see “General Note” 5 at the site http://www.acipco.com/adip/specials/span.cfm). In most normal pressure applications with quite small joint deflections, the amount of external/lateral joint stabilizing thrust resistance required is not large compared to other thrust foci calculations such as larger angle bend fittings etc. The amount of this lateral thrust is also calculable by the conventional bend thrust equation, Thrust = 2PAsin (theta/2), where P is the pressure, A is the effective cross-sectional thrust area of the pipe (based on the outside barrel diameter for ductile iron pipe joints), and theta is the amount of joint deflection (in effect the angle formed by the intersection of the two central axes of the pipes adjacent the deflected joint) in comparable units (see in the DIPRA thrust restraint literature previously mentioned).