I am simulating one liquid injecting into another liquid in a tank and trying to determine the transient jet front and shape. I noticed there are different jet length when I use different values of volume fraction in the jet contour plot. Same thing happens while using different jet velocities for determining the jet boundary. This led me to think about the criteria for determining the jet shape in previous jet experiments. How to match the test data with CFD?

Thanks!

Well, how does your test data determine it?

I am using previous test results, in which they used a high-speed camera to visualize the dyed jet with color. Obviously, the color will disappear as the concentration of dye or dyed liquid decreases. In CFD, I modeled the injection as dispersed fluid and process the results using the volume fraction of the injection fluid. Thanks!

Dispersed fluid!? Thats your problem right there.

Use a single phase flow with a non-diffusing scalar to tag the jet flow.

Thanks Joe.

I tried the scalar approach before and the transient jet length I got is much less than experimental data, almost half of the measured jet length. Then I switched to dispersed fluid approach. Do you have any suggestions what parameters in the scalar approach I can adjust to increase the predicted jet length to match the test data.

Thank you so much for help!

How about you provide proper information such as fluid type, Reynolds number, meshes, ... Then you can get meaningfull input.