rubber pad area required to secure horizontal boiler

huangyhg

rubber pad area required to secure horizontal boiler
hello all;
     i have a cylindrical storage tank that rests horizontally on it's side in four cradles along it's length.  i am trying to figure out if the packing material (novo elastomeric bearing pad material) will have a high enough coefficient of friction to keep the cylinder from wanting to slide off the cradle supports in the event of an earthquake.
     i've done some looking around on the net, eng-tips, and reading text books, however the traditional coefficient of friction theory just doesn't seem to fit my situation.
     setting aside the idea that it should just be bolted down (please,there are a number of very good reasons that this cannot be done) and also the ultimate non-frictional solution of providing half cradles at each end, how would you go about designing this?  how do you determine an appropriate amount of bearing material and what coefficient of friction to use?
     i'm leaning towards asking for half cradles at each end to physically restrain the cylinder under uls earthquake loads, however i would be very keen to have someone explain how the force of friction can possibly be independant of area, as well as a good reason to not rely on the coefficient of friction (in case the client balks at the additional "unjustified" cotst).
thanks in advance,
ys
b.eng (carleton)
working in new zealand, thinking of my snow covered home...
earthquake acceleration is both vertical and horizontal.  ties are required to prevent excessive movement.  these ties can take the form of cables that limit the horizontal movement or slots that stop the movement at the limit of the slot.  
thanks for the reply civilperson, but i am quite familiar with seismic design.  i am looking for assistance with the interpretation/application of friction to my design.  can you comment/shed light on my question?
vertical accelerations are virtually never greater than g, unless near-fault (not the case here).  they will not actually cause the tank to lift in this case, but will decrease the vertical load causing the friction.
this is a very large tank.  in my honest opinion, i do not think it's possible for it to slide off the cradles.  that said, i will need to satisfy myself as to the validity of that argument, or provide end cradles.
anyone else care to address my original question?
thanks all,
ys
b.eng (carleton)
working in new zealand, thinking of my snow covered home...
if anyone is curious, i believe that another engineer here in the office and i have solved my problem...
after setting this issue aside for a while and working on other projects, it suddenly occured to me that while the friction formula may be independant of area, the load carried by each cradle is not.  if we had cradles of differing sizes, their stiffness under loading would differ, and thus attrat different amounts of load, which would give rise to different amounts of friction force in the traditional formula.
i would think that with limit state design factors and checking any logical limit states (aka failure mechanisms), such as shear rupture of the bearing pad, slipping off the cradles due to insufficient friction, etc) we should have a workable and reliable solution.
civilperson:  thank you again for your reply.  i forgot to give credit for your raising the important point about the vertical accelerations.  these are particularly key here since they will reduce (and also increase!) the load normal to the pads which give rise to friction.
regards,
ys
b.eng (carleton)
working in new zealand, thinking of my snow covered home...