The flange components consisting of flange, gaskets, and bolts may have been adequately designed but their performance to specifications will be affected by assembly conditions.
Flange Surface Finish
Flange surface finish is critical to achieve the design-sealing potential of the gasket. Again, gasket-leak tightness is dependent upon its operating gasket stress. Flanges that are warped, pitted, rotated, and have incorrect flange gasket-surface finish will impair the leak tightness of the gasket.
Flanges out of parallelism and flatness should be held within ASME B 16.5 specifications. This will ensure that the uniform bolt loads translate to uniform gasket stress.
The resiliency and compressibility of the gasket are affected by flange surface finish. Recommended flange surface finishes for various gasket types are shown in Table A7.18.
Never reuse a gasket. A gasket’s compressibility and resiliency are severely reduced once it has been used.
Check the gasket for any surface defects along the contact faces that may im- pair sealing.
Keep the gasket on its storage board until immediately prior to assembly.
Do not use any gasket compounds to install the gasket to the flange, as it affects the compressibility, resiliency, and creep behavior of the gasket. Consult the gasket manufacturer when installing large diameter gaskets for a recommendation on how to secure them to the flange during installation.
Bolts and nuts may be reused providing they are in new condition. Ensure bolts and nuts are clean, free of rust, and that the nut runs freely on the bolt threads. Install bolts and nuts well lubricated by using a high quality anti-seize lubricant to the stud threads and the nut face.
Methods of Bolt Tightening
Once the total bolt loads (W ) are calculated for the flanges, specifications, and procedures should be adopted outlining how to achieve the design bolt load.
The total bolt load (W ) for the flange is divided by the number of bolts to determine the individual bolt preload ( Fp).
To achieve improved leak tightness sufficient and uniform gasket stress must be realized in the field. This obviously requires uniform and correct applied bolt load. The higher the requirement to reduce leakage, the more controlled the method bolt tightening.
The common methods of bolt tightening are:
Each method has its own assembly efficiency. Bolt tightening methods and their assembly efficiencies are shown in Table A7.19.
Hammer, Impact Wrenches Method
This method remains the most common form of bolt tightening. The advantages are speed and ease of use. Disadvantages include a lack of preload control and the inability to generate sufficient preload on large bolts.
Torque wrenches are often regarded as a means to improve control over bolt preload in comparison with hammer-tightening methods. However, as indicated in Table A7.19, significant variation in stud-to-stud load control is still evident.
Much attention is given to the level of torque that should be applied to a specific application. However, it is not the torque that is important but the end result of the torque-bolt preload. Control over bolt preload is the factor for ensuring proper gasket-seating stresses are achieved.
Torque is the measure of the torsion required to turn a nut up the inclined plane of a thread. The efficiency of the nut’s turn along the bolt thread to generate preload is dependent upon many factors, including thread pitch, friction between the threads, and friction between the nut face and the flange face.
In general, only about 10 percent of the applied torque goes toward providing bolt preload. The rest is lost in overcoming friction: 50 percent in overcoming the friction between the nut and flange faces, and 40 percent in overcoming friction between the threads of the nut and the bolt.
Another variable to overcome is the elastic behavior of the joint as illustrated in Fig. A7.15.
As the bolts are tightened creating the desired preload, the flange will partially compress. As additional bolts are tightened, the flange joint will compress a little further. The continuous deflection of the flanged joint reduces the stretch (or preload) of previously tightened joints. This phenomenon is referred to as cross talk and is a result of tightening a multistud flange one bolt at a time.
A typical wide variation in bolt and bolt preload is experienced using torquing because of the uncontrolled effects of friction and cross talk, as illustrated in Fig. A7.16, ‘‘Typical Load Scatter of 28 Bolt Heat Exchanger Flange.’’
Torque Calculations. The amount of torque that is required to generate a specific bolt preload is calculated by
Torque Procedure. Torquing should be applied in multipasses following a cross pattern to reduce warping of flange, crushing the gasket, and to minimize cross talk in achieving bolt preload.
1 ¹⁄₃ of final torque (T ). Start at bolt no. 1 and follow cross pattern
2 ²⁄₃ of final torque (T ) following cross pattern
3 At final torque (T ) following cross pattern
4 At final torque (T ), start at highest bolt number and tighten in a counterclockwise sequence
The cross pattern is easily followed once the bolts are numbered in the flange. Randomly select a bolt and designate it as bolt number 1. Proceed in a clockwise motion to the next bolt and add four to the previous bolt number. Moving clockwise, the next bolt number would be 1, 5, 9, and so on.
This system of adding four to the previous bolt number continues until adding four to the previous number exceeds the total number of bolts in a flange. At this point, start again at bolt number 3. Continue in the same clockwise direction, numbering bolts 3, 5, 11, and so on until again this number is larger than the total number of bolts in a flange. At this point the next number is 2; continue as previously described: 2, 6, 10, and so on. The last series of bolt numbers start with bolt number 4 and continue 8, 12, 16, and so on.
A sample 16-bolt flange showing a typical cross pattern is shown in Fig. A7.19.
Many of the variables that reduce the control of bolt preload using the torque process are eliminated using hydraulic tensioning systems.
Hydraulic tensioners are hollow hydraulic compact cylinders that are threaded onto a protruding section of the studbolt generally using a pulling device. A bridge supports the hydraulic head straddling the nut and reacting against the flange while hydraulic pressure is applied to the hydraulic head. Under the applied hydraulic load, the bolt stretches at the same time as it compresses the flange and gasket. Residual bolt load equivalent to the desired preload (Fp) is achieved by manually turning down the nut under the tensioner bridge during the applied hydraulic load. Applied bolt load is directly proportional to the hydraulic pressure and the area of the hydraulic cylinder. There are no frictional losses associated with tensioning,
as compared to torquing. A cross section of a hydraulic stud tensioner is illustrated in Fig. A7.20.
The residual load (preload Fp) = Applied Load — Load Loss Factor.
The load loss factor is dependent upon the stud stress realized, bolt diameter, and effective length of the bolt. For each application its load loss factor can be precisely calculated to determine the necessary applied load to generate the residual preload. Development of thorough procedures is essential to maintain the accuracy
of hydraulic stud-tensioning process.
Cross talk is significantly reduced by utilizing multistud tensioning. Generally 50 percent of the studs in a flange are tensioned simultaneously by using multiple tools interconnected with a high-pressure hose tied into a common pump source. Many flange configurations allow for 100 percent of the studs to be tensioned simultaneously. This completely eliminates cross talk.
Hydraulic tensioning provides the most controlled tightening method for achiev- ing specified bolt preload.
Controlled bolting is the method where the loading-stress of the flange bolts is measured using ultrasonic equipment to ensure that the correctly specified bolt preload is achieved.
The application of torque alone to the flange is not controlled bolting, as there remain many uncertainties about the actual bolt load.
Torquing in combination with ultrasonic measure provides necessary controls to achieve the required bolt preload.
Multistud tensioning following established procedures provides a high degree of control over bolt preload. In critical application, multistud tensioning should also be combined with ultrasonic measurement to verify that all specifications are met. #Little_PEng