ble leakage flow. However, once the compressor is up to speed the faces separate
leaving a very small gap. The size of the gap is very critical as it controls the
leakage.
The main problem associated with designing a dry gas seal is calculating the
forces on the seal faces during operation. The forces have to be known so that the
seal faces can be designed to open at the correct operation conditions and the clear-
ance between the faces maintained at the required value. Under normal operating
conditions there is a force balance between the separating forces and the closing
forces. Friction within the O-ring type seal, which is incorporated into dry gas seal
design, plays an important role in the balance of forces, especially during the start
up when the gap between faces is being created. It is essential that the separating
forces can overcome not only the closing forces but also the frictional force in the O-ring seal located behind the static face. Until this force is overcome the seal faces
will not come apart.
The present paper presents results of experimental studies [5] where the primary
objective was to measure the frictional force in an O-ring seal at the commencement
of linear motion as a function of pressure using a configuration close to that of a dry
gas seal. There have been a number of studies, mainly of theoretical nature, on the
O-ring type seal. Green [6] used the finite element method to analyse the behaviour
of an elastomeric O-ring seal in compression. The state of deformation of an
unpressurised elastomeric O-ring seal inserted into a rectangular groove was stud-
ied by Dragoni [7,8]. Stress fields in a compressed unconstrained elastomeric O-
ring seal were investigated by George [9]. He compared his computer model predic-
tions with experimental results. All these studies have concentrated on the static
performance of an O-ring seal. Narumiga [10] studied an elliptically deformed seal-
ing ring for a shaft seal. Although he considered dynamic conditions, the emphasis
was on the state of deformation within the O-ring seal.
It can be safely concluded that the public domain information about the level
of friction in an O-ring seal at the moment when the shaft commences its linear
motion is rather scarce. As the object of our study was a real, full-scale O-ring seal,
interpretation of the results in terms of precedents, such as dry friction of rubber or
the lubrication of elastomers, is practically impossible and will therefore not be
attempted in this paper.
2. Experimental apparatus and procedures
2.1. TEST APPARATUS
In order to test the frictional performance of the O-ring type seal under condi-
tions similar to those encountered during operation of the dry gas seal, a suitable
apparatus was required. The main conditions to be simulated during testing are as
follows:
(i) unlubricated contact between seal and contacting material,
(ii) variable pressure environment with maximum gauge pressure of 10 MPa,
(iii) minimal leakage through the O-ring seal,
(iv) linear motion of the shaft, O形密封件摩擦性能英文文献和翻译(2):http://www.youerw.com/fanyi/lunwen_1699.html