I am making a comparison of anchor loads on a pipe rack in seismic ( load acting axially to the pipe case U i.e pure Seismic ) based on hand calculation and CAESAR output. I am new to pipe stress.

My methodology is :

Total wt.times base shear coefficient - Total Friction load = Total Anchor load and load per Anchor= Total load /nO. OF ANCHORS.

If my hand calculation method is correct, the Anchor load should be equal to zero if the base shear value ( which i have taken as 0.3g) is equal to friction coefficient ( taken as 0.3).

I have used Friction coefficient at all restraints ,all og which are vertical restraints but the CAESAR II calculated values and Hand calculation results do not match.

Kindly correct me where I am wrong.

Baburao,

Please draw the Free Body Diagram and you will know why the results are different.I will give u a hint:In Case U ( pure OCC), What are the loads that are acting on your system?

I have some questions too.

What code are you using that allows you to use friction to dampen seismic loads?

How are you able to predict that your supports will be seeing 100% of the dead weight load during a seismic event?

Most building codes PROHIBIT the use of dubious frictional forces to act as restraints during seismic events.

If I was reviewing work of that nature I would toss it out and have it redone sans friction.

Finally don't even get me started on what the value for MU should be. If you use common sense (as well as science) you soon realize that Mu may be one value at T=0 seconds and another value some time later after the object begins to slide.

I fully agree with John that friction is anybodies guess and that calcs should be done frictionless. However, my personal point of view is that friction cannot be ignored just because it cannot be predicted with any degree of certainty. IMHO calcs should also be done with at least some high end friction values to give a range of results. Each calc might bear little resemblence to reality but together hopefully should bracket the realistic picture.

I agree with Andrew... friction should only be considered as a hinderance never a help. From the standpoint of stresses due to the occasional load set I would anticipate that no friction would result in the highest displacements/stresses.

However to determine if my assumption is correct, build the model with mu, then set the mu multiplier to zero in the load case editor and set up summations for both mu and no mu, design for whichever represents the worst case.