Hi all,

I have a query about the allowable load calculation for a bare pipe when it is in direct contact with support steel.
Usually the conservative method is pipe sizes 10" and less can be directly placed over support steel provided the pipe has no hot/cold insulation. if the pipe is over 10" size, then a wear pad is required below.
But one of our company standard states that it depends on pipe wall thickness and the loads at support point. it has an example of allowable loads for different pipe sizes with different thickness. if the load exceeds, then we need to provide wear pad or shoe.
unfortunately the example covers thickness upto 8mm only.
So, i need to calculate the allowable loads for thickness more than 8mm for different sizes of pipe. if anyone has experience in calculating it, please give me guidance.
I considered a method of limiting hoop stress to 20% of SMYS or longitudinal stress to 50% of SMYS. but it addresses only internal pressure but not the loads due to bending, thermal and occational.
Please refer the attachment and guide me how to calculate this allowable load. thanks in advance.

We use (from Kellogg's):

Allowable span L (feet) =

(2/3)S(d+b)(t^2)
----------------- (divided by...)
(0.644)W(R/t)^0.5

S - Hot Basic Allow. Stress (psi)
d - support beam width, inches
b - bearing width, 1.5% of pipe OD, inches
t - corroded wall thickness, plus any repad, inches
W - unit weight of pipe and contents, lb/ft
R - pipe radius, OD/2, inches

This limits the local stress due to bearing to 67% of S(hot)

Hi, dclarkfive,

I am in a similar situation, having a long large bare pipe on structure.

I have to decied whether I should use pad or not.

Could you let me know more about the Kellog's formula?

Thanks.

Would someone be able to reference the specific kellogg publication for the kellogg formula in Dave's post above please.
Thanks,
Chris

I haven't such reference and it seems that the formula has some troubles to dimensional consistency....

It is likely that the basic model is "the beam on elastic foundation" model, a theory familiar when Kellogg developed such formulas.

As reference please see Peng book, chapter 5.6, where the circumferential membrane stress in pipe is taken as
S=0.643*f*[sqrt(R)]/[t^1.5]. There "f" is the load per unit length of circumference, I guess they considered f as W*L/(d+b), so
S=0.643*[W*L/(d+b)]*[(R/t)^0.5]/t
Limiting stress to 67% of S_hot (a safety coefficient of 1.5), it results something similar to what is posted.

Best regards.

I think a reasonable reference would be Bednar's Pressure vessel design handbook.
In the second edition it is in chapter 7, local stress in shells due to loads on attachments, 7.6 Line Loads.
However Bednar considered that the maximum induced stress under Force on cylindrical shells can be 1.5-2 times larger than the stress computed by the theoretical formula, because of "the smaller rigidity of cylindrical shells in the tangential direction".

Best regards.

I ended up using the L.C. Peng Chapter 6.7.4 Roarks approach. Also the Kellogg Subject 3810 for comparison.

There is a new composite product in Market called "Wear Pad". These are typically epoxyed on the bottom of the Pipe at the support points and provides better support and avoids stress corrosion on the pipe wall.

Dear mariog,

In our project,

40" line is simply resting on the beam with 20 ton vertical load, we provided a Rpad. we have to calculate bearing stresses.

Whether this support is safe with the Rpad or we have to provide shoe type.. on what basis/theory we can say this support is safe or not.

i refered "In the second edition it is in chapter 7, local stress in shells due to loads on attachments, 7.6 Line Loads." but those calculations are for clip supports.. in my case it is simply rested in the beam directly

Dear C. Wilcox.
L.C. Peng Chapter 6.7.4 Roarks approach. in this where i can check the load. because i need type of attachment based on load also.( 20 tonn) what to do.

Dear Shan,

if your problem is solved, PL shared with us also.

Durga,

The reference I made to Bednar 7.6 Line Loads was answering to a request.
For your case, you should decide in your application the interaction pipe-support would be counted as a "line load" type or not, longitudinal or tangential- it is not an exact science there.
See also Roark's 13.7 Pipe on Supports at Intervals.

Best regards.

Thanks Mariog for your Reply,

I refered Roarks 13.7 chapter, but i think in that chapter they discussed about design of saddle.

But in my case

with a particular load, Wear pad is Sufficient or i have to use Saddle.

In this Area any suggestion...

sir, Do u have ROARKS for stress and strain, 5th Edition... i have 7th edition. but i need 5th edition badly.. if u have pl share with me,,

my mail id durga_2045@yahoo.co.in