Hello everybody,
I am designing a counter flow heat exchanger embedded in a phase change material. I am using corrugated pipes instead of plain ones with fins attached throughout the length of the heat exchanger. It is quite a detailed geometry. I was wondering, normally how do CFD experts model large heat exchangers? Since the whole exchanger would require at least 100-120 million cells to incorporate all the details (about 6*6*6 feet). I cannot use a plane of symmetry since there isn't one, and I am interested in the transient thermal performance. The two options I had in mind are as follows, but im not sure if they would work:
1. Scaling down the model by e.g. a factor of 10 (but this would result in the same element numbers of the mesh just with smaller sizes). Dont know if it is effective ?
2. Doing only a portion of the exchanger (but no sure which portion to do, since every length in the exchanger would have a different temperature and heat profile). Then somehow finding the thermal parameters per unit length and extrapolating ?
Do I have any other reasonable scientific option ?
Looking forward to your advice, based on your experience for such problems.
Thanks Abdur
Maybe you can start by defining the objective of your study. Based on that information you can motivate the level of detail required.
In any case a 'bottom-up' approach is preferable, where you start simple and incrementally add complexity until the desired level of accuracy is reached.
Hello Horimek Abderrahmane
Thank you very much for your valuable reply. It was really helpful.
To me your third paragraph seems the most relevant. I am not sure how to do this in CFD, particularly using Star-CCM. Do you know of the method to input data from a file into starccm ? Is it alsp possible to iteratively do this, without manual having to give the input file upon completion of every geometry length ?
Thank you
Hello Abdur Rehman Mazhar
As an engineer you should use Reynolds analogy for scaling down
Regards.
Alaa
- Badges
- Science topic