Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: slip joint/friction weld

[Subject Prev][Subject Next][Thread Prev][Thread Next]

I would resolve the dead load into the component vector normal to the friction plane, divide it over the surface area in contact and multiply it by its static friction coefficient (0.74.) to determine the ultimate PSI.  What the safety factor should be is another question…

From there I would determine the stress at the extreme fiber in bending + compression just the same as normal (e.g. F=M/S) – this would need to be less than the allowable determined from the friction coefficient*FS. 


I would imagine one would want to assume a fairly large FS as this calc relies on 100% perfect contact between the faces… but what kind of code reference addresses this I haven’t a clue.


From: Gray Hodge [mailto:ghodge(--nospam--at)]
Sent: Tuesday, February 16, 2010 3:34 PM
To: seaint(--nospam--at)
Subject: slip joint/friction weld


To all,


I have been asked to review a connection which the manufacturer refers to as a friction weld but which looks more like a slip joint to me.  The connection is for two sections of a light pole, and I’m sure you have all seen something just like it.  The pole consists of two sections of tapered pipe so that upper section of pipe slips over the top end of the lower section of pipe, and there are no mechanical connectors at the joint.  The overlap is roughly equivalent to twice the average diameter at the slip connection.  There is no tension at the connection – only compression in the pole and bending due to lateral loads.  The task seemed relatively simple, at first, but then I tried to analyze it.  I can reason out that the vertical load is transferred by the upper pipe section bearing on the lower section and that the moment will be transferred by a resisting couple across the joint.  I suppose that friction resistance will become important if the taper is severe enough, but these pipe sections have a fairly gentle taper.  Is anyone familiar with a methodology for checking a connection of this nature?