Hi all,
does anyone know what the "bearing stress" is? ASME31.3, paragraf 302.3.1 b) limits it (1.6 times value from Table A-1 or A-2) but does not define it.
A typical example is the "point" load from the support when checking large diametar water piping requirements for bearing pads. Local stresses are shear, bending and compression. Is the bearing stress combination of these?

Hi Ranka,

In the cited Code paragraph, there is mention of several "pure" loadings that are really very rarely encountered in the real world. These are offered for "completeness". The bearing stress is a pure compressive stress. There are also allowable stress magnitudes provided for pure tension (e.g. the stress in hanger rods under ideal conditions) and pure shear (be aware of this limit if you have a very heavy valve near a support). These are not intended to be combined stresses.

The Codes address thermal (displacement) stress ranges with equations for calculating combined stresses using Tresca failure theory. In the High Pressure section of B31.3, Von Mises failure theory is used in the equation for required wall thickness.

Of course, the above is just my opinion and it is not necessarily the opinion of ASME or any Code Committee.

Best regards, John.

[ November 15, 2001: Message edited by: John Breen ]

Thanks John for your reply
In the mentioned, high pressure section of ASME B31.3, the compression and bearing are both mentioned between other "pure" stresses. Then the bearing stress must be something else.
I am looking at this to evaluate local stresses at support "point" for 36" and 44" (wall 9.5mm thick) cooling water lines and to size the bearing pads if they are necessary. I am using basic stress formulas S=F/A for compression stress and comparing it with ASME B31.3 allowable. Is this too simplified? What about local bending and shear? If I calculate combined stress, which allowable must I compare it with?
If anyone knows a manual method to resolve this (except for FEA) please response.
Thanks

Hi Ranka,

I think that your instincts are correct. There will be compression, local membrane bending and shear all acting in the pipe wall at the point of contact with the support. Of course when we use beam theory (as does CAESAR 2) we calculate beam bending stresses over the gross section of the pipe. When you are using large diameter pipe (especially with heavy contents) this may not tell the whole story. You will have to look at these local (to the pipe wall) stresses as a combined stress using the most appropriate theory of failure (may I suggest Von Mises). Just remember, when you go away from B31 beam theory methods, it is no longer valid to use the B31 allowable stresses - you will want to look at the ASME Section VIII, Division 2 methodology and associated allowable stresses.

To properly analyze the local stresses in the unstiffened pipe wall you will need to take your CAESAR 2 calculated loads and use them in a finite element analysis (PIPE/FE, e.g.). The pipe will want to ovalize at the point of contact and a "doubler plate" might help the situation at the local contact point - but to truly understand the stress pattern around the ovalized section, you will need FEA.

You might look at how the folks that design penstocks support these big water pipe loads. They like to use ring stiffeners and support the pipe on columns that straddle the pipe. The ring stiffeners keep the pipe (somewhat) "round" at the point of support and the rings make convenient attachment points for hanger and supports (columns). If you can get a copy of AWWA Manual M11, you will see some manual methods for designing water pipe supports. There are also some manuals issued by the AISI that detail steel water pipe support schemes. I think the ring stiffener design methods for horizontal vessels are similar and these are described in the classic pressure equipment design book by Brownell and Young. Using these methods will preclude the need for FEA but they will provide a more conservative design margin ($$$).

You will probably end up having to design some steel base plates if you use short columns (for carrying the load to the ground) and you will find that you must also consider compression and bending on the plates (they want to "curl" under load - you will have to know how big the foundations are under the base plates). I think we all use the AISC Steel Design Manual method for designing base plates and anchor bolts. If you use this method, also use the allowable stresses given in the manual.

Good luck with your project.

Regards, John.

We look at the local stress at the supports for every pipeline we design. This allows us to use maximum spans. The pipeline are to B31.1 but we calculate the local stresses using BS5500 appendix G and threat the pipe line a long vessel.