You do indeed need to consider sliding forces in the design of the
anchor bolts but you do not need to design for the full magnitude of the
force. Friction at the fixed end will work in your favour.
Under Canadian code procedures we would use the following limit states
Factored resistance >= Factored load
where factored load = frictional force at the sliding end
multiplied by a safety factor (1.5 in Canadian codes)
factored resistance = reliable factored (maximum) shear strength of
the installed anchor bolts plus the factored friction at the fixed
end. (0.85 would be the reliability factor for restraining
friction in Canadian codes).
If no teflon slide plates are used sliding and fixed support frictional
forces would be the same and we would design the factored strength of
the anchor bolts for something like 60% of the friction forces
calculated at one end. For a 500 kip vessel with 0.3 coefficient of
friction this amounts to about 45 kips. Unless you need anchor bolts
this large to resist seismic or other loading you should consider the
use of some type of sliding plates at the sliding end only to reduce the
There are also some other considerations.
1. Large horizontal forces may cause deflection of both piers
greatly reducing the amount of actual sliding that takes place or
even making pier deflection the governing force.
2. Another respondent suggested that your temperature range was
ambient. This might not be true. Propane being pumped into the vessel
may be at a higher than ambient temperature. Propane being withdrawn
from the vessel will cause evaporation inside the vessel which, in
turn, may lower the contents down to the liquefying temperature (about
-35 F. for propane) if large enough quantities are being withdrawn in
a short period of time. This would be ambient for me but probably not
3. Beware of using "building" codes for industrial structures.
Building codes are primarily intended to provide safety for occupants
in structures which are normally occupied, therefore, they do not
always deal adequately with industrial loading situations. In
particular, the practise of using higher allowable stresses or lower
load factors which is commonly applied in "building" codes should be
used with caution with industrial structures. In your case friction,
which is a load case which is thermal in origin, can not exist
independently of the live load (the vessel contents), therefore,
increasing the allowable stresses should not be considered.
Your colleagues may have been lucky to have had no problems in the
past. Although they may not know it they have been relying on the
safety factors inherent in other load cased to protect them against this
load case. They should at least check this load case out before they
decide to ignore it.
As for instantaneous loading, artillery fire provides instantaneous
loading but not negligible results!
H. Daryl Richardson
Padmanabhan Rajendran wrote:
> As far as I am aware of, horizontal vessel, supported
> on two saddles, has slotted holes on one of the
> saddles (sliding end) and standard holes on the other
> saddle (fixed end). Horizontal force of friction due
> to thermal expansion/contraction of the vessel is,
> then, equal to the product of coefficient of friction
> at the sliding interfaces and the vertical reaction at
> the sliding end. Consequently, one of the loading for
> which the anchor bolt on the "fixed" end should be
> designed for is the friction force. Because this
> loading is transient, one may use a higher allowable
> stress or use a reduced load factor. This is what I
> have learned and practiced for about 10 years.
> I have changed jobs and the lead engineers (mechanical
> and civil) in this company tell me that the above
> noted friction force need never be considered in the
> design of anchor bolt. Their reason is that the vessel
> moves on the sliding end (and, probably on the fixed
> end also because the hole diameter on the fixed end is
> larger than bolt diameter)and, any friction force that
> may develop is so quick and instantaneous that the
> bolt will not experience it. The tank in question is a
> propane storage tank with an operating weight of 500
> kips. Based on my experience I figured that the
> friction force will be very large and that in order to
> reduce it, I suggested using Teflon sliding surfaces,
> such as Fluorogold product. The engineers here say
> that Teflon sliding pads do not work well and that the
> pads easily come off the equipment. They also told me
> that it is common to have both saddles with slotted
> holes and that they have never heard of using friction
> reducing sliding surfaces on horizontal vessels! Each
> of them has more than 20 years with reputed and large
> consulting engineering companies, whose design
> standards echo what I have been practising.
> Rick Drake and Christopher, in particular, (because I
> believe that both of you have refinery and pipeline
> experience), am I overly conservative in my design
> approach and is my design criteria in error?
> P. Rajendran
> Do You Yahoo!?
> Get email at your own domain with Yahoo! Mail.