Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Calculating concentration boundary layer thickness

Please login with a confirmed email address before reporting spam

Hello,

I am running a simulation where I am blowing air over a water surface. My goal is to characterize the boundary layer thickness as a function of position. I was wondering how I can use COMSOL to do this? Here is a photo of the concentration profile obtained from the simulation. I also uploaded my model.

As you can see, the boundary layer is the teal color, and the blue is the bulk concentration. Also notice that the teal layer is getting thicker as we move across the block. So I want to know how I can find out the thickness of the teal line as a function of distance.

Thanks!


3 Replies Last Post 17 nov. 2015, 15:16 UTC−5

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 27 oct. 2015, 23:35 UTC−4
I also wanted to mention that I am getting a warning

Accurate boundary flux is not available while solving. Using domain flux instead.

and I am wondering how I might be able to fix this.

Thanks
I also wanted to mention that I am getting a warning Accurate boundary flux is not available while solving. Using domain flux instead. and I am wondering how I might be able to fix this. Thanks

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 17 nov. 2015, 07:54 UTC−5
I'd also like to know how to do this. Too much of the time I am using a slow, sledgehammer approach to do things, thinking there must be an elegant built-in function to do it.

From a search it seems there was a feature inn earlier versions of COMSOL called postinterp() which looks like it might have something to do with it, but this does not return any hits when searching the documentation or help. Hopefully that means it has been incorporated into some other feature (and it is just a matter of finding what)

I looked at your model. The mesh is quite coarse in the boundary region, and there were only two boundary layer elements. You could specify your own mesh (seems like you have just used a Coarse, Physics defined mesh). Adding more boundary layer mesh elements might be a better way to get finer resolution in the region of interest (boundary layer) without making the whole domain finely meshed (and hence slow to solve).

For a sledgehammer approach I try defining in DataSets a 2D Cut Line, vertical at a series of x positions. Then made a 1D plot group with a 1D line graph that graphed the concentration on each x-value for the Cut Line. All these curves come on the same plot, which you then can export the data from (helps to select "sort" under Advanced). I imported the data into excel and used the Forecast linear interpolation to determine the y-coordinate height at which the concentration was 0,8. Looked pretty linear in the region of c=0,8, but otherwise it is just to plot and fit a trendline and interpolate with the equation. Attached the 2D cut line plot (c vs y for different x) and the resulting BL growth plot (of y for c=0,8 vs x) from Excel. A real pain to have to do this manually, so hope someone comes with the elegant solution.
Post back if you find how to!
cheers
Ross
I'd also like to know how to do this. Too much of the time I am using a slow, sledgehammer approach to do things, thinking there must be an elegant built-in function to do it. From a search it seems there was a feature inn earlier versions of COMSOL called postinterp() which looks like it might have something to do with it, but this does not return any hits when searching the documentation or help. Hopefully that means it has been incorporated into some other feature (and it is just a matter of finding what) I looked at your model. The mesh is quite coarse in the boundary region, and there were only two boundary layer elements. You could specify your own mesh (seems like you have just used a Coarse, Physics defined mesh). Adding more boundary layer mesh elements might be a better way to get finer resolution in the region of interest (boundary layer) without making the whole domain finely meshed (and hence slow to solve). For a sledgehammer approach I try defining in DataSets a 2D Cut Line, vertical at a series of x positions. Then made a 1D plot group with a 1D line graph that graphed the concentration on each x-value for the Cut Line. All these curves come on the same plot, which you then can export the data from (helps to select "sort" under Advanced). I imported the data into excel and used the Forecast linear interpolation to determine the y-coordinate height at which the concentration was 0,8. Looked pretty linear in the region of c=0,8, but otherwise it is just to plot and fit a trendline and interpolate with the equation. Attached the 2D cut line plot (c vs y for different x) and the resulting BL growth plot (of y for c=0,8 vs x) from Excel. A real pain to have to do this manually, so hope someone comes with the elegant solution. Post back if you find how to! cheers Ross


Please login with a confirmed email address before reporting spam

Posted: 9 years ago 17 nov. 2015, 15:16 UTC−5
Hello Ross,

Yes, that assignment was due awhile ago, so it is sort of a thing in the past. I used a vertical cut line and then plotted concentration gradient vs. vertical position, and the location where the gradient is zero is the thickness of the boundary layer. I never found an elegant solution, just this brute force method.
Hello Ross, Yes, that assignment was due awhile ago, so it is sort of a thing in the past. I used a vertical cut line and then plotted concentration gradient vs. vertical position, and the location where the gradient is zero is the thickness of the boundary layer. I never found an elegant solution, just this brute force method.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.