I could find some furmula to calculate the shell dimeter based on the number of the tube and pitch distance;but when I want to use NTW baffle I can not put tubes in some part of the tube so I wont have enough area to put all my tubes.
Can any one recomend me an advice or formula in NTW baffle type calculation?
As you may know, there are various methods for designing shell & tube heat exchangers with different complexity and accuracy. I think it is better for you to use a more complex method because of your case. I mean a method that uses baffel parameters (e.g. baffle spacing, baffle type, and baffle cut) for the heat transfer coefficient (and perhaps the pressure drop) of shell-side.
Baffle cut obviously has a significant influence on flow pattern and heat transfer. The effect is complex but we can explain it with some simplifications. Shell-side flow through baffle window is like an opposite flow heat exchanger while the flow across tube bundle is similar to a cross-flow heat exchanger. So from thermal viewpoint, we prefer an opposite flow heat exchanger. Therefore, having tubes in the window increases the effectiveness. Of course, if the baffle window becomes very large, the advantage of opposite flow heat exchanger will be diminished because of a lower Reynolds number in window region. This is explained by a parameter ("correction factor for baffle cut"). This factor (Jc) expresses the ratio of heat transfer coefficients comparing to the value for a completely cross-flow case.
Bell presented a curve for this correction factor. When 40% of tubes are located in window, the value is 1. Higher portion of tubes in the window leads to a smaller Jc (e.g. for 80% of tubes in window Jc is 0.7). On the other hand, lower number of tubes in the window, which is equal to less baffle cut, increases Reynolds number of window region and results in higher Jc. Approximately we have highest correction factor (Jc=1.15) when 10% of tubes are in the window. For your case, when we have no tube in window (NTW), the correction factor is equal to 1. In fact, the heat exchanger is almost a pure cross-flow type.
So I suggest to use for exapmle the Delaware method. If you want to remain your method simple, consider your shell-side as a completely cross-flow heat exchanger. However, there are other correction factors that should be considered such as C.F. for baffle leakage, bundle bypass flow, inlet baffle spacing, etc. As a rule of thumb, all of these correction factors may lead to a factor equals to 0.6 or something close to it. In fact, shell-side heat transfer coefficient is anyway smaller than the value for a ccross-flow heat exchanger.
Thus it is maybe better to use a pure cross-section heat transfer coefficien (across tube bundle) with a constant correction factor (for instance 0.6-0.7). This is not accurate but is simple. From this you can calculate your tube numbers and diameter. By having the tube number, diameter and pitch, you can calculate the tube bundle diameter. Remember that you have no tube in window section but your tubes are nearly filled the shell diameter in horizontal direction. After obtaining the shell diameter, you should recalculate the shell-side heat transfer coefficient and find number of tubes again. Iterate until tube numbers or shell diameter converges.
The design of shell and tube is presented in the book
Ludwig's Applied Process Design for Chemical and Petrochemical Plants
You can find how to calculate shell diameter in NTW baffle shell and tube heat exchanger
The design of shell and tube heat exchanger with NTW baffles was presented in the attached research paper
Effectively designed NTW shell-tube heat exchangers with segmental baffles using flow hydraulic network method.
You can also review its references
@Hamid Abroshan Thanks alot for your complete explanation,your answer could help me well.
@Gamal B. Abdelaziz Thanks for the complete book and useful article that you mentioned; it helped me well.
Atefeh Shiri Thank You
At any time you can tell me about you requirements
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence