1. Lateral displacement is not as simple as "horizontal movement": first clarify the force model
When many people hear "adjusting the metal expansion joint horizontally", their first reaction is: Isn't it just that the pipe is tilted to the side, and the expansion joint is bent with it? Wrong. The lateral displacement is not the translation of the pipe, but the relative displacement at both ends of the expansion joint that is perpendicular to the axis. This action is far more deadly for bellows than axial compression or stretching.
If you think about it, bellows is essentially a thin-walled flexible shell. When it is subjected to axial force, the crest and trough of the wave are uniformly deformed; However, once the lateral displacement occurs, one side of the bellows is stretched and the other side is compressed, and the stress distribution on both sides of the wave crest is completely uneven. Coupled with the circumferential stress generated by internal pressure, the bellows is actually in a multi-axial stress superposition state. This is why the same bellows may allow the axial displacement to be ±50mm, while the lateral displacement is often only given to ±20mm or even less.
So before selecting, first find out the displacement mode of your pipeline system: is it pure lateral, or is it a composite displacement of lateral and axial? If the pipes at both ends expand thermally in the same direction, the middle expansion joint bears pure lateral displacement; If one end is fixed and the other end is free, then the axial vector may also be superimposed. Under the compound displacement condition, don't just take the lateral displacement value to set the sample. That data is usually measured under the condition of no internal pressure, pure transverse and cold state, and it should be discounted in engineering.
2. Compound hinge transverse expansion joint vs. compound straight pipe bypass pressure balance type: what is the difference
There are many products for lateral compensation on the market, and the most common two types are compound hinge transverse expansion joint and compound straight pipe bypass pressure balance expansion joint. The name is long and easy to get dizzy, but the core difference is one sentence: the former relies on hinges to force the bellows to swing only laterally, while the latter relies on bypass pipes and pressure balance structures to eat the internal pressure thrust by itself.
A short pipe in the middle, with a bellows at each end, and connect the two ends with hinges. The role of the hinge is to constrain axial displacement, allowing only angular displacement to occur. As soon as the two bellows swing, the middle pipe section moves laterally. The advantage of this structure is that the internal pressure thrust is borne by the hinge, and the pipe support does not have to bear the blind plate force; The disadvantage is that it cannot absorb axial displacement, and the center of rotation of the hinge must be precisely designed, otherwise the bellows will suffer.
What about the double straight pipe bypass pressure balance expansion joint? It is to add a bypass pipe on the duplex structure, lead the medium to the outside of the bellows, and eliminate the internal pressure thrust by the principle of pressure balance. This structure is a little more "advanced" than the hinged type, because it can not only do lateral displacement, but also absorb part of axial displacement at the same time, and the thrust on the fixed bracket is almost zero. The cost is that the structure is complex, the cost is high, the space occupied is large, and the flow resistance is higher than that of the hinge type.
How to choose the working condition? If you only need to absorb the lateral displacement caused by thermal expansion on the steam pipe, the fixed bracket can withstand moderate thrust, and the double hinge lateral type is enough; However, if the internal pressure of the pipeline is high, the diameter of the pipe is large, or the position of the fixed bracket is limited, the thrust must be removed, and then the double straight pipe bypass pressure balance type must be used. There is another point that is easy to overlook: if the medium is smoke or slurry with particulate matter, dust is easy to accumulate in the bypass pipe, and the hinged type is more worry-free at this time.
3. Guide tube, tie rod nut, hinge structure: which configurations really affect the transverse adjustment amount
The difference between how much two expansion joints with the same nominal lateral displacement can actually be adjusted often lies in the detailed configuration.
Let's start with the deflector. Many design institutes only add guide tubes to axial expansion joints, thinking that the transverse type only has transverse displacement, and the guide tubes are unnecessary-this is a misunderstanding. When transverse displacement, there will be disturbance between the inner wall of the bellows and the medium, and the guide tube can straighten out the flow field and prevent the vibration caused by vortex. Especially when steam or high-speed air flow is passed, there is no transverse expansion joint of the guide tube, and small cracks are easily eroded at the root of the bellows. However, note that the guide tube itself should also swing with the bellows. If the design gap is not enough, the lateral displacement will be limited. Therefore, the lateral expansion joint of the guide tube is added, and the actual allowable lateral displacement may be smaller than the theoretical value. It should be confirmed with the manufacturer when selecting the type.
Look at the tie rod nut again. The hinge and tie rod of the double hinge transverse expansion joint are two sets of systems. The hinge is responsible for constraining axial displacement, and the tie rod nut is used for cold tightening, or pre-biasing. Loosen the tie rod nut during installation to allow the bellows to be in a free state; After the pipeline is in place, adjust the nut position according to the calculated cold tightening amount. The tie rod nut is not casually screwed-too much screwing is equivalent to applying extra preload to the bellows, and less screwing will not achieve the effect of pre-bias installation. When leaving the factory, the length of the tie rod and the position of the nut are calibrated according to the design displacement. It is best to measure it before moving it on the spot.
Another thing that is often overlooked is the turning clearance of the hinge. There must be a proper gap between the hinge pin shaft and the ear plate. If it is too small, the rotational resistance will be large, and the actual lateral displacement will not be reached at all. Too large will create impact loads again. This thing looks inconspicuous, but it directly determines whether the horizontal adjustment amount can be realized.
4. The three most common pits during installation: how to cooperate with cold tightening, guide bracket and fixed bracket
Two days ago, I met an engineer who was doing power plant maintenance, saying that the transverse expansion joint on one of their DN500 steam pipes leaked in less than one maintenance cycle. When I removed it, I saw that the bellows trough was full of cracks. After asking around, the problem was with the installation.
The first pit is cold tight made into a "dead pull". The purpose of cold tightening is to make the expansion joint bias a pre-displacement in advance in the cold state, so that the displacement can fall in the middle of the bellows stroke in the hot state, so as to avoid unilateral overload. However, in many scenes, the pipe is directly pulled to the installation position with an inverted chain, and then the tie rod nut is hard locked-this is called forced countermeasure, not cold tightness. The real cold tightening must control the displacement according to the design value. Generally, check the direction and size of cold tightening with the manufacturer before installation, instead of "almost the same" on the spot.
The second pit is the improper setting of the guide brackets. Many people think that fixed brackets are installed at both ends of the expansion joint. In fact, the transverse expansion joint has stricter requirements for the guide bracket than the axial type. The guide bracket should ensure that the pipe only produces lateral displacement near the expansion joint, but no torsion-once the pipe rotates, the hinge bears torsional force, and the bellows will twist and twist. The guide bracket is generally arranged at both ends of the expansion joint, and the spacing is calculated according to the diameter and stiffness of the pipe. It is not just welding a shelf.
The third pit is the fixing bracket being "cut corners". Although the internal pressure thrust of the transverse expansion joint of the compound hinge is borne by the hinge, the elastic reaction force caused by the deformation of the bellows, the friction force of the pipeline and the thrust force caused by the flow of the medium still need to be carried by a fixed bracket. In many projects, the fixed bracket is designed according to "only bearing the reaction force of the expansion joint". As a result, the extra load generated by the thermal expansion of the pipeline is piled up during actual operation, and the bracket is deformed, resulting in the deviation of the actual displacement trajectory of the expansion joint. In the force analysis of the fixed bracket, the whole pipe system should be counted, and we can't just look at the single product of the expansion joint.
5. Calculate an account: How to weigh transverse stiffness, fatigue life and selection margin
The selection of lateral adjustment metal expansion joint is essentially a balancing technique.
Low lateral stiffness means that the bellows is "soft", the reaction force is small under the same lateral displacement, the pipe support is lightly stressed, and the fatigue life is longer. However, there are also problems with too low stiffness: first, the bellows has thin walls and decreased pressure resistance; second, it is easy to produce flow-induced vibration in the flow field; third, it deforms at a slight touch during installation, which makes the construction difficult.
Conversely, the stiffness is a little higher, the pressure resistance and stability are better, but the reaction force generated by the lateral displacement increases exponentially-the stiffness does not change linearly, and the lateral stiffness of the thin-walled bellows is about proportional to the cubic of the wall thickness. If you increase the wall thickness from 1mm to 1.5mm in order to withstand pressure, the transverse stiffness may be more than tripled. When the reaction force is large, the fatigue life will drop. Here is a classic empirical formula: the fatigue life of a bellows is inversely proportional to the square of the displacement. As long as the displacement is increased by 40%, the fatigue life is halved.
Then how to keep the margin of selection? It is recommended that the transverse displacement be selected according to 70% of the actual calculated value, that is to say, when the transverse displacement of the pipeline is calculated to be 20mm, the specification of nominal displacement of 30mm should be selected. If the margin is too small, it is easy to exceed the limit if the hot working condition fluctuates slightly; If the margin is too large, not to mention the size and cost of the expansion joint, the transverse stiffness may decrease the stability because the waveform of the bellows becomes larger. In addition, the stress caused by internal pressure should also be counted in the fatigue check-when the pressure is high, the allowable displacement of the bellows will be obviously smaller, which will be missed by many people who choose the type.
If your working condition has high pressure, large temperature fluctuations, or frequent lateral displacement (such as units with many starts and stops), don't get stuck in sample data selection. Ask the manufacturer for a specific fatigue life calculation book, see the allowable stress amplitude of bellows material under cyclic working conditions, and then make a decision. Adjust the metal expansion joint laterally. If you choose the wrong model, at most, the expansion joint will be broken early, but if the selection logic is wrong, the whole pipeline system will be broken.