I am simulating shell and tube heat ex-changer i have mass flow inlet and pressure oulet (zero gauge) but my pressure readings are not accurate i have tried velocity inlet but results are same. the boundry conditions i have used are same as previous research papers but somehow by results are off by about 3 Kpa.
Plenty of reasons could be the cause: mass, momentum and energy conservation, fluids' properties, geometry, grid, numerical scheme.
Check the mass flow rate at the inlet and the exit as a starting point and see if the difference is zero.
Hello, are you comparing CFD results with experimental ones? In addition to Ahmad suggestion, I would check how experimental pressure drop was obtained. I mean that if it was calculated there can be many different ways of obtaining friction factor, which will influence the CFD/experimental results. This is what learned from my own experience, I was almost giving it up, but reading some papers I realized that this could be my problem. Hope you get it fixed.
@Ahmad Khalaf Sleiti @Humberto Santos the difference b/w inlet and outlet is zero. convergence is ok. Also to mention heat transfer coefficient and total heat transfer match my hand calculations the only issue is pressure drop seems to be off.
Humberto Santos I am comparing my results with bell Delaware and kern but pressure results are off .... i am rechecking my calculations but so far no improvement.
@Muhammad Saad. In this case, if you have made sure residuals are okay, I would recommend you check out your reference values. These values are important because friction factor is obtained from them. So, right after setting up the boundary conditions, look for the option reference values and take a look at them. Basically, the pressure is obtained from friction factor which in turn is calculated from the wall shear stress, and the wall shear stress is obtained from these reference values. Make sure density and velocity are okay as well as lentgh. Good luck and let us know about your progress. I am facing a lot of problems with my simulations too, some days are really good and others it end up with disappointing results, but we can not give it up, the right answers will come soon. Besides, we are learning a lot from these difficulties.
Humberto Santos
Humberto Santos Indeed we are learning alot. I will check my reference values thanks for your help
In a numerical simulation of steady Navier-Stokes equation, for each finite volume, pressure is a scalar parameter that meets the mass conservation in the continuity equation. Because only pressure gradient directly appears in the momentum equation but no pressure appears in the continuity equation, the pressure gradient can be accurately calculated but the pressure is an integrated quantity based on the reference value.
Sun Kyoung Kim thanks for feeback. what should be reference values. I take density and velocity at inlet. Secondly, length is taken as length of tubes
Muhammad Saad Usually, we set the pressure outlet BC for an internal flow, where you know the pressure. Then, you don't have to worry about the reference pressure. Internally, the algorithm solves a Laplace equation for the pressure. So, you need a Dirichlet condition somewhere in the BC.
The length and the properties would not be a problem. However, it should be understood that force and the velocity at a point are dependent on each other. Set the pressure BC somewhere far from the inlet, i.e., outlet.
Sun Kyoung Kim... I do have applied outlet pressure BC( zero gauge). As mentioned in description.
- Badges
- Science topic
- Similar topics
- Condensed Matter Physics
- Zero