Christopher -
Your comments help me see a path ahead a bit better. Thank you.
The primary stresses are below the limits. It is the secondary stresses that exceed the allowable limits. And I understand that the primary stresses will not have a significant effect on fatigue strength for carbon steel piping experiencing less than 50,000 cycles, and this system has probably seen less than 10 FT cycles.
I have performed site investigations to verify the boundary conditions. I think the restraint modeling is realistic and have looked at a range of friction assumptions.
My observation is that the center of gravity of the reducing station is almost right on thrust line of the anchors to either side of it. So, the station is overly retrained due to the external piping configuration. Additionally, since there are multiple legs and operating scenarios the expansion conflict between hot and cold legs is too great for the compactness of the station. So, the internal piping is overly restrained too.
What I understand you to suggest is that a remaining life can be determined for a low cycling system (under 7,000 cycles) with high overstress, provided I can obtain reliable thermal cycling history data.
The way I would know to do that is to use ASME B&PV Code Section 2 and a pipe stress program to calculate the fatigue stress and the cumulative usage ratios for each node. From this I can determine whether a new system will exceed the allowable fatigue stress or not. I've not done this before, but perhaps through an iterative process determine how many more cycles the system can endure before exceeding the fatigue limit. This would give an estimated remaining life. Is that one "standard technique" that could be used?
However, if modifications are made, then new cyclic stresses will develop with different strains than in the initial installation. I don't see how I can add those fatigue cycles to the first.
(See also my follow-up to Mr. Luf's response on this topic.)
Ken A. Nisly-Nagele, P.E.
Project Engineer, Mechanical
Applied Engineering Services, Inc.
7999 Knue Road
Indianapolis, IN 46250
317-585-8920
-----Original Message-----
From: PipingDesign@yahoogroups.com [mailto:PipingDesign@yahoogroups.com]
On Behalf Of Christopher Wright
Sent: Friday, February 24, 2006 12:13 PM
To: PipingDesign@yahoogroups.com
Subject: Re: [PipingDesign] Modifications to B31.3 Piping w/ Overstress
On Feb 24, 2006, at 10:32 AM, Ken Nisly-Nagele wrote:
> Stress modeling of existing pressure reducing station that is within
> the
> scope of the piping receiving the planned modifications indicates
> expansion
> stresses are 7 times and occasional stresses are 9 times above the
> allowable
> stresses in portions of the station.
If these are primary or secondary stresses, you don't need any further
comment. At the very least you're headed for a significant problem. If
I were to guess, your system is over-restrained or your restraint
assumptions in the FE model aren't realistic. The latter situation
isn't all that uncommon, but don't just assume that's the situation.
> My understand is that just because the piping has held up for 10 years
> does
> not mean the piping is okay,
At best it means that the first failure is 10 years closer to happening
than it was when the system was new. At worst, it means that there have
been failures and either no one remembers or they're not talking. Or
maybe that you're asking the wrong people.
> 1. Concerned about the fatigue stress that is resident in the
existing
> piping, it seems prudent to not simply make restraint or piping
> modifications that reduce the stress, but to demo the piping that is
> shown
> with expansion overstress. Comments?
You should estimate the remaining life and show that additional stress
cycles after the mod is installed will satisfy system life
expectations. You can estimate the remaining life of the system with
more or less standard techniques. Your problem will be getting reliable
estimates of loading cycle from plant people. I've had some unfortunate
experiences with selective memory on the part of plant people or people
who and simply made up things in an attempt to be helpful.
> 2. It also seem prudent to remove the piping that is indicated as
> overstressed by occasional loads too unless it can be established that
> occasional loads have not occurred. Comments?
Again, you should make a good-faith attempt to mitigate the overstress,
and show how it affects system life.
I'm sure there are plant engineering people reading this who now think I'm an impractical techno-snob, who just wants to turn a simple upgrade into a PhD thesis. They wouldn't be the first. But I figure I'm straddling a line between making impossible demands on a client and aggravating an unsafe situation. Either way it's problems for me and the client, and it's very much a judgment call. I tend to put a lot of weight on my impressions of the client before I start using his input to make engineering assessments.
Christopher Wright P.E. |"They couldn't hit an elephant at chrisw@skypoint.com | this distance" (last words of Gen.
.......................................| John Sedgwick, Spotsylvania1864)
Yahoo! Groups Links Received on Mon Feb 27 16:17:00 2006
This archive was generated by hypermail 2.1.8 : Mon Oct 27 2008 - 20:24:10 EDT