Sources/sinks

April 11, 2000, 07:16

I am currently doing a final year project at university in CFD. I need to plot the closed body shape caused by sources/sinks in a uniform 2d flow. I would like to know how the stream function equation ( = Uy - mtan-1 (2ay)/(x2 + y2 - a2)) can be adapted for multiple sources and sinks as I have successfully done one source and one sink !! Thanks

April 17, 2000, 09:10

I am also doing a similar project and have not got a fucking clue. Perhaps we should meet up so time and talk it over.

May 14, 2000, 23:15

Do you just need to (A) plot the shape or (B) derive an analytical expression for it? (B) does not belong in this forum, I would think. (A) can be easily solved in the spirit of CFD: set up a grid covering the region of interest, sum up the potential caused by each point singularity (source/sink) at each grid point. Once you have a discretized representation of the potential function, you can differentiate it numerically on this grid to get the velocities, and integrate to get the stream function, and then plot the contours to find closed surfaces. This is sort of like a crude panel method ..

April 11, 2000, 07:16	Sources/sinks	#1
john Swallow Guest Posts: n/a	I am currently doing a final year project at university in CFD. I need to plot the closed body shape caused by sources/sinks in a uniform 2d flow. I would like to know how the stream function equation ( = Uy - mtan-1 (2ay)/(x2 + y2 - a2)) can be adapted for multiple sources and sinks as I have successfully done one source and one sink !! Thanks

April 17, 2000, 09:10	Re: Sources/sinks	#2
Paul [think as pig shit] Bartlett Guest Posts: n/a	I am also doing a similar project and have not got a fucking clue. Perhaps we should meet up so time and talk it over.

May 14, 2000, 23:15	Re: Sources/sinks	#3
Cfd Guest Posts: n/a	Do you just need to (A) plot the shape or (B) derive an analytical expression for it? (B) does not belong in this forum, I would think. (A) can be easily solved in the spirit of CFD: set up a grid covering the region of interest, sum up the potential caused by each point singularity (source/sink) at each grid point. Once you have a discretized representation of the potential function, you can differentiate it numerically on this grid to get the velocities, and integrate to get the stream function, and then plot the contours to find closed surfaces. This is sort of like a crude panel method ..