Could an expert please help check whether so many calculations are necessary for designing and verifying the pipeline network?
Thread Content
First, to explain it briefly, it’s a piping network that follows a path similar to the one shown in the diagram. It’s not very large, and its layout is simple as well. The temperature is quite high: the inlet temperature is over 400 degrees, and after being heated by the heater, the operating temperature reaches around 700 degrees, with a pressure of over 2 MPa. The supervisor now requires the calculation of the following: 1) Primary stress, secondary stress, and accidental stress (caused by earthquakes) in the pipeline network, as well as support displacements and forces (calculated using CII); 2) Instability of pipes under compressive loads at both ends; 3) Calculation of forces and deformations at tee joints using finite element software; 4) Determination of the vibration modes of the pipeline network, along with consideration of the rigidity constraints of the supports. Additionally, with regard to external loads on the pipes, friction between the airflow and the pipes must be taken into account, as well as a so-called pressure thrust (I’m not sure what this force is; there are no expansion joints in the pipeline network). Earthquake loads corresponding to a magnitude of 8 must also be considered. Installed indoors; wind loads are not considered. The external loads I applied previously included the frictional force of the supports, as well as the aerodynamic forces resulting from air flow at the elbows (flow rate * velocity), and the thrust generated when the gas was discharged from the pipeline (also flow rate * velocity; however, due to water spraying for cooling, the flow rate doubled, and since the gas was discharged directly into the atmosphere, the velocity increased significantly as well). I have a few questions: 1) Does such a piping network need to consider the possibility of pressure-induced instability? We used Ansys to simulate a pipe segment and found that the critical pressure at which compressive instability occurs is 1,300,000 N. The pipelines in the network are not subjected to such high forces; the value is at least an order of magnitude lower, and there are no pipes in the network that are under pure compressive stress. 2) The right-angled tees used are all standard components, which possess sufficient strength on their own. That type of tee, in which the thin tube is inserted obliquely into the larger tube and welded to it directly, can also be considered for reinforcement under GB150 standards. After calculating the stress values using CII, is it still necessary to use finite elements? 3) Calculating the vibration modes of a pipeline network mainly involves determining its natural frequencies, right? But it should be that a scaffold exists first in order to be considered a mode; how can the stiffness of the scaffold be determined based on the calculated results? Moreover, even if the natural frequency of the piping network is calculated, how can we determine whether it is reasonable? Our airflow conditions are constantly changing, with no fixed excitation frequencies; furthermore, being several hundred meters away from the compressor, it has no impact on this small section of piping network, as there are no other rotating devices or piston-driven mechanisms in between. 4) The frictional force between the airflow and the pipeline is very small (estimated to be around 100 N/m); generally, this force does not need to be taken into account, right? ! To calculate this force, a uniformly distributed load must be applied; in that case, it’s not possible to use the static method to determine the seismic load, right? (In CII, only one set of uniformly distributed loads can be applied, right?) 5) Is it theoretically necessary to add aerodynamic forces and exhaust thrust at the elbow I added earlier? The exhaust thrust is quite high, at several tens of thousands of N, but the force at the elbow is only a few thousand N.
My colleague and I analyzed your problem and were a bit surprised. 1. I’m not sure what you mean by pressure-induced instability of pipes When designing pipes, we pay attention to their stability, but this is achieved by adjusting the pipes’ stability through the use of supports and hangers ； 2. Since the CII calculations have all been successful, there’s no need for finite element analysis anymore; perhaps your supervisors think your workload is too light and want you to take on more work ； 3. GB 50316 provides a formula relating pipe frequency to pipe span; you first determine the range of pipe frequencies, and then use that formula to calculate the maximum span ； 4. The friction between the airflow and the pipeline – isn’t this the frictional loss along the flow path? 5. I don’t understand what you mean. Do you want to install supports at the elbow?
Thank you for your reply. Let me clarify further: the first item refers to beams that are under compression at both ends, and which experience bending instability once the critical pressure is exceeded. What the manager meant was to calculate this value. The fourth item is because the leader believes that the pressure loss force is not taken into account in CII, and it needs to be added. The fifth item is also due to the fact that the model in CII represents a closed pressure pipeline, without taking into account the aerodynamic forces associated with gas flow; therefore, it is necessary to determine this force separately and then add it as an external load.
The purpose of item 5 is to determine more accurately how the scaffold will be stressed under these external loads, so as to establish appropriate requirements for the foundation work
First, do you mean external pressure acting on the pipeline when you talk about compression? If so, then it should be to determine the magnitude of the accidental concentrated load on the pipeline ； Fourthly, the pipeline pressure drop is not calculated using CII; for relatively simple piping systems, it is usually calculated manually, while for more complex systems, software such as Aft Impulse is required ； Fifthly, for the calculation of the exhaust pipe, we generally first calculate the exhaust reaction force and then apply this force to the exhaust outlet before proceeding with the calculation. The effect of fluid flow on the pipeline is not considered.
The pipeline pressure drop was calculated manually, and our piping network is not very complex either. Well, the force per meter is calculated by multiplying the pressure loss by the pipe area and then dividing by the pipe length; this value is then added as an additional continuous load. The problem is that I think this force is quite small, so it seems unnecessary, right? I need to find ways to come up with reasons to convince the manager right now
I checked 50316, but I didn’t see the relationship between the natural frequency and vibration Which chapter is it roughly in? For earthquakes, should static analysis or dynamic analysis be used? For static analysis, is the acceleration determined by referring to the local basic seismic acceleration as specified in the building seismic code, or is it calculated using formulas according to American standards? If dynamic analysis is used, what input parameters are required? Please let me know if you know, thank you!
Pressure drop multiplied by the pipe area divided by length – what type of force is this value that you calculate? I don’t remember which code links the loads on supports and hangers to internal pressure. GB 50316 indeed does not specify a relationship between frequency and span; I got it wrong – that information is in the book \"Stress Analysis of Pressure Piping\". Natural frequency = SQRT(g/deflection, 0.5)/(2π). According to GB 50316, pipes must not experience resonance; therefore, the span of the supports and hangers only needs to meet the requirements to prevent resonance. I don’t have any requirements regarding seismic loads, so I haven’t carried out any calculations; I can’t help you with that. However, Appendix F of DL/T 5054-2016 recommends the static method, which provides formulas for calculating acceleration.
Thank you for the reply! 1) According to the momentum theorem in fluid mechanics, the force exerted by the side surface of a pipe on the fluid volume equals the difference in impulse at the inlet and outlet of that volume. Considering a pipe as such a volume, the impulse is equal to (PA + Wv). Ignoring changes in flow velocity and gravity, the force acting on the side surface of a pipe with a constant diameter, which is essentially the frictional force between the fluid and the pipe, is equal to (P1 – P2)*A = ΔP*A. Dividing this force by the length of the pipe gives the value of the internal friction per unit length, with units of N/m; this represents a uniformly distributed load. I understand what our LD means: since the CII calculation does not take fluid flow into account, the software likely does not consider this force. Yet this force has a tendency to \"push\" the pipeline away, which definitely results in stress on the supports. But this force is too small; it’s generally not something that needs to be taken into consideration, right? For the same reasons, LD also requires the inclusion of the aerodynamic force generated by the airflow turning at the elbow, as well as the exhaust reaction force at the outlet of the exhaust pipe. 2) I also found the formula for the natural frequency, and I learned that the specification in 50316 stating that the deflection must not exceed 15 mm was set with the consideration that the natural frequency should not be lower than 4 Hz. Conversely, as long as I can calculate the deflection, I can also determine the natural frequency; and if the deflection does not exceed 15 mm, then the natural frequency will be greater than 4 Hz. However, this raises several issues: a) In this case, the deflection can only be calculated separately for each pipe diameter, resulting in different natural frequencies. So what are the results obtained by the dynamic modal calculation module in CII? Are there different frequencies for different pipe diameters, or is there one single frequency for the entire pipeline network (without considering the order of modes)? The software version in our unit is too old; it’s still version D. I can’t perform these calculations, so you might want to try it yourself to see what the result is. b) The natural frequency of the pipeline network is related to the stiffness of the supports. Whether they are fixed supports or sliding supports, in the software they are treated as having a +Y constraint; in other words, their stiffness is considered to be infinite in the software. Yet in reality, supports do have specific stiffness values. So, is it necessary to impose limits on the stiffness of the supports in order to achieve the calculated natural frequency, by determining a minimum stiffness value for them? (This is how we in LD should think about it, but I guess since there are design specifications for brackets, as long as I know the load they need to withstand, then a bracket designed according to those specifications should be pretty accurate; there’s no need to worry too much about the exact stiffness of the bracket.) ）
Let’s return to your initial question: what is required for the verification of pipe supports and hangers? In other words, what aspects are involved in the loads on these supports and hangers? The main ones include: 1. The weight of the pipe components and the insulation structures； 2. Gravity of support and hanger components ； 3. Gravity of the medium transported through pipelines ； 4. Gravity of the medium used for hydrostatic testing or pipeline cleaning ； 5. Forces generated by internal pressure on pipe compensators ； 6. Thermal stress calculation: forces and moments acting on the support and suspension nodes ； 7. Friction force generated on movable support brackets due to pipeline displacement ； 8. Snow load on outdoor pipelines ； 9. Wind load on outdoor pipelines ； 10. During normal operation, the pipeline vibration forces caused by various reasons ； 11. Sudden changes in the momentum of the fluid within the pipeline, such as transient forces caused by steam hammers and water hammers ； 12. Emission reaction force generated by fluid discharge ； 13. Seismic forces generated by pipeline installation in seismic areas, without considering the situation where seismic and wind loads occur simultaneously. So your first calculation regarding the force with which the fluid in the pipe \"pushes away\" the pipe, as well as the aerodynamic forces on the elbows, is meaningless. What needs to be considered is actually item 11 on my list, that is, the impact exerted on the pipeline by the airflow resulting from the instantaneous opening of valves at the start-up of the system; this requires calculation using water hammer simulation software. Regarding the fixing frequency of the pipes, you’re overthinking it; a fixing frequency of more than 2.55 Hz is sufficient for the pipes outside the device. If you want to be extra cautious, aim for a deflection of less than 10 mm.
Okay, thank you. In our systems, the valves are not opened suddenly at all; they are opened gradually. Moreover, in the initial stage, the valves are opened first before air is allowed to enter, so there should be no need to worry about air hammers