Steve and Kaustubh,
Maybe we were explaining different things at beginning and communicating in different ways. I overlooked heat transfer effects that can cause temperature differences scattered around (and moving about) in the material. I guess an FEA analysis would consider thermal transients, thicknesses, thinning of film layer at inside surface of outer arc, flow induced temperature differences, and some other stuff I know little about. My guess is that analysis would reveal the operating temperatures are normally greater on the outside bend arc than on the inside one. "Imagining FEA saves the cost of running it." I have never done FEA and only had a short course in heat transfer.
In stress analysis, most of us consider the entire elbow at same temperature.
Best wishes,
Ben
-----Original Message-----
From: Steve McKenzie [mailto:Mechproj@xtra.co.nz]
Sent: Saturday, April 20, 2002 5:12 AM
To: <a href="/group/PipingDesign/post?postID=hj0EX6eNhNalOrNDLsaSfxzSZ8ClUweESSnBDIFXplOlC0gEwdPWZpeEGx11I5uM803uAUBGSAWkzohabxbwShbFF-QC">PipingDesign@yahoogroups.com</a>
Subject: Re: [PipingDesign] Another Trick
Dear Kaustubh
you have fallen into exactly the same hole that I did when I started this,
and I am satisfied that for thin wall pipes I was wrong and Ben is right.
Explanation? Here goes:
Consider your 0.005mm/mm or 0.5% thermal strain. All dimensions will
increase by 0.5%, including bend radii, and pipe diameter, which I
overlooked. The bend centreline radius will increase as well. Suppose, for
ease of visualisation the strain was increased to 5% or even 50%. The bend
will simply be a larger version of the pre-strained version, but
geometrically similar. If you are still unsure, draw it out on paper; it
only takes a minute.
However, if the effect of significant wall thickness is added then I am
unsure, and would think some secondary stress could possibly be generated.
However I have difficulty in visualising the effect. A finite element study
would put my mind at rest.
Sorry about the Red Herring. I'll start taking the pills again.
Cheers
Steve McK
> Dear Ben,
>
> When the temperature of the outer surface is more than the inner one, then
the thermal bowing comes into picture.
>
> I try to explain what I think, I may also be wrong:
> Take a 12"NB, 1.5D 90 Deg. bend. The outer radius is 619mm, while inner
radius is 295mm. The outer surface developed length is 973mm, while the
inner surface developed length is 464mm. Now let us take a thermal strain of
0.005mm/mm.
> The expansion of the outter surface is 4.86mm, whereas that of the inner
surface is 2.32mm.
> What happens when the bend is subjected with this temperature? Since the
outer surface wants to expand more than that of the inner surface, and there
is a connecting piping at start and end of bend, then bend bends inward.
There is always a end rotation associated with start & end of the bend. At
the same time there is ovalization of the bend (flatter from top & bottom
surfaces).
> And this happens at all the temperatures increase. Am I sounding logical?
>
> Regards,
> Kaustubh
>
>
>
> _____________________________________________________
> Unsubscribe: <a href="/group/PipingDesign/post?postID=H3YwV5v4EuRp-TYwkaLYppS-zU1lWBOpL_VOvxdWNk6UWvYPaOzdULgRxl-8IJ2CuepKQdXiEUoJj1LAgq1pmJ0kET33TLytSbG3MEAy7SI">PipingDesign-unsubscribe@yahoogroups.com</a>
>
> Your use of Yahoo! Groups is subject to <a href="http://docs.yahoo.com/info/terms/">http://docs.yahoo.com/info/terms/</a>
>
>
>
Your use of Yahoo! Groups is subject to <a href="http://docs.yahoo.com/info/terms/">http://docs.yahoo.com/info/terms/</a> Received on Sat Apr 20 11:42:00 2002
This archive was generated by hypermail 2.1.8 : Tue Mar 04 2008 - 11:40:22 EST