Leakage
From Joints Containing Gaskets
Introduction
Incidents of fluid leaking from a joint containing a gasket is one which many Engineers have encountered. The inclusion of a gasket in a joint to prevent leakage is only effective as long as there is sufficient clamp force generated by the bolts to allow effective sealing. For joints whose design are not covered by standards, establishing the number and distribution of bolts which must be used to ensure effective sealing can be problematical. The complexities of gaskets are such that empirical methods have been developed to approximate their performance and operating characteristics. These methods involve the use of empirically derived factors to approximate the clamp force required to allow the gasket to seal effectively.
There are two practical factors which are commonly used in joint design involving gaskets. The m or maintenance factor is used to establish the clamp force required to ensure an effective gasket seal when the joint is subjected to internal pressure. The y or yield factor is used to determine the clamp force required to be applied to a gasket to ensure that it seats properly to provide a seal. Further explanation, of how these factors are used to establish the bolt clamp force needed for the gasket to seal effectively, is presented in the boxed caption.
It must be pointed out that besides calculation, a large degree of experience is required to ensure a leak free joint. Gasket manufacturers sometimes vary the m and y factors to suit the medium which the gasket is sealing. This is based upon the experience that gaseous media is generally more difficult to seal than fluids. Increasing the factors in this way allows the approach to be valid where overwise it would fail to yield sound results. Consultation with gasket manufacturers is to be encouraged at the joint design stage to ensure that account is taken of all relevant factors.
Another common problem is inadequate bolt spacing. In many joints the bolt spacing is dictated by the gasket pressure midway between bolts. If insufficient pressure is applied to the gasket in such regions, leakage can result. Research into this subject indicates that clamping pressure, in joints without gaskets, quickly decays away from the bolt. In such joints zero pressure occurs between 1.8 and 3.6 times the bolt radius, the actual value being dependent upon the joint details. With a gasket the pressure decays more slowly, however, to overcome this problem many design codes dictate a maximum bolt spacing (which varies with bolt size).
Local crushing of the gasket can occur if the clamp force generated by the bolt is excessive for a particular gasket material. Special pressure sensitive film (such as Fuji film) can be used, once the joint is designed, to determine what the local pressures are within a joint. All gaskets have a crush strength which, if exceeded,will result in excessive creep leading to leakage. The crush strength of gaskets can be obtained from the gasket manufacturers. For example, for premium grade compressed asbestos fibre jointing, the maximum permissible surface stress typically varies from 60 MPa, for a gasket thickness of 3mm, to 190 MPa for a thickness of 0.5mm. The gasket area locally around the bolts can be increased when the crush strength is exceeded. If the gasket has insufficient crush strength for the application, reselection of the gasket or redesign of the joint is necessary.
Ideally all the bolts of the joint should be tightened simultaneously, especially when gaskets are being used. If this is not achievable, a tightening sequence should be specified. A poorly specified tightening sequence will result in uneven load distribution in the joint increasing the chances of failure occurring. (Without using numerical techniques such as the finite element method and making assumptions about the unevenness of the joint faces, it is not possible to quantify the magnitude of this irregular load distribution.) Sound tightening sequences, based upon experience, have been established. If the bolts are in a circular pattern, a criscross tightening sequence is usually specified. For noncircular bolt patterns, a spiral pattern starting at the middle has been found to be effective.
Creep or relaxation of the gasket
material can be a practical problem. This is a particular
consideration in joints subjected to temperatures greater
than 100 °C. The amount of relaxation which can occur
with gaskets is usually far greater than that which occurs
in joints which do not contain such compliant materials. Because
of the magnitude of bolt clamp force loss which can occur
with gaskets, frequently a retightening schedule is specified
24 hours or more after initial tightening. The magnitude of
the clamp force reduction due to gasket creep can be of such
magnitude that is not feasible to design the bolts for this
loss. To overcome this problem, retightening of the bolts
can be specified after a period of time following initial
tightening, frequently 24 hours. Such a schedule may involve
retightening the fasteners on a regular basis to overcome
the problems of relaxation.
The most prevalent controlled method of tightening bolted joints containing gaskets is by tightening so that a specified torque is achieved. This method is generally known as torque control. Without experimentally obtained torque values or analytical tools, specifying the correct torque which should be used in an installation, can be problematical. Leaks from many joints are directly attributable to a poor torque specification. A deficient tightening torque leads directly to an inadequate clamp force. This may well be insufficient to achieve gasket sealing.
By a wider awareness of the potential problems involved when designing joints containing gaskets, and by an appreciation by Engineers of the importance of bolt clamp force, will gasket leakage problems be prevented. In order to assist the Engineer in the specification of the correct tightening torque, Bolt Science has developed a computer program which can account for the relevant factors involved. Using the assistance of online help screens and an inbuilt database, the Engineer can determine the tightening torque and the resulting clamp force for both metric and imperial threaded fasteners. The effect of changes in the fastener design and the prevalent frictional conditions can quickly be established.
The Gasket Factor m
If the joint containing a gasket is to seal an internal
pressure then there must always be a sufficient clamp force
applied to the gasket, to ensure that a leak free joint is
achieved. The m (maintenance or multiple) factor, is the factor
that provides the additional clamp force capability in the
joint's fasteners to maintain sealing pressure on a gasket
after an internal pressure is applied to the joint. It is
a multiple of the internal pressure; the ratio of gasket contact
pressure to contained pressure. Values of the m factor for
various gasket materials can be obtained from pressure vessel
and other similar national and international standards. Values
for the m factor can vary from 0.5 for rubber to 3.5 for asbestos
(values vary with thickness and composition). To provide an
estimate of the minimum bolt clamping force required for joints
subjected to an internal pressure, the following formula may
be used with caution:
Fk = (Fh + P x m x A) / n
where:
Fk = Minimum required clamp force from each bolt in the joint
Fh = Hydrostatic end force acting on the joint
P = Internal pressure acting on the joint
m = Gasket factor
A = Total area of gasket based upon using an effective gasket width (this value is covered in standards such as BS 5500)
n = Number of bolts in the joint
The formula assumes that the same size and grade of bolt is used in the joint and that the distribution of bolts in the joint is sound. To establish what bolt pattern does give a sound joint is largely down to experience.
The Gasket Factor y
Before a leak free joint can be obtained, it is necessary to seat the gasket properly by applying a minimum initial pressure (under atmospheric temperature conditions without the presence of internal pressure). This design seating stress has been given the term y (yield) factor, and it is the stress required to deform the gasket into the irregularities of the joint surface. It is governed by the compressibility of the gasket material. Values of the y factor for various gasket materials are given in the BS 5500 standard. To determine the minimum clamp force to meet this requirement, the following formula can be used:
Fk = A x y/n
where y is the gasket factor and other terms are as given before.
Conclusions
The minimum clamp force required from the bolts, to ensure that the gasket seals effectively, is the maximum of the two values determined using these two factors. Implicit in this method is that the bolt spacing and the rigidity of the joint flanges are such, that problems will not occur because of deficiencies in these areas. This is largely down to experience for those joints which are not covered by existing design codes.
Applying state of the art analytical analysis to prevent gasket leakage
can be complicated. To assist the Engineer in overcoming the problems associated with the use of threaded fasteners and bolted joints, Bolt Science has developed a number of computer programs. These programs are designed to be easy to use so that an engineer without detailed knowledge in this field can solve problems related to this subject.
