Discussion Forum

Hi. I'm trying to learn how to model phase change and I've been following this 1D example about melting ice:
www.comsol.com/showroom/gallery/474/

I have it mostly working, but in my results even half of the ice hasn't melted by 1200s, unlike in the example figures where it seems to be completely melted by 60s. I think I've understood everything about it except for the phase change pulse D. In pages 2-3 it says that the integral of D must equal unity to satisfy a condition.

Currently I have just declared the expression D = diff(H,T), which I understand is the derivative of H with respect to T. H is defined by the Heaviside function H = flc2hs(T - T_fusion, dT). I've tried changing the dT from 0.1 to 2, but it doesn't seem to have an effect.

The D is supposed to be normalized, but if it doesn't come like that automatically, I don't know how to do it. I tried checking the value of the integral through Subdomain Integration, but I don't even know what variable it is integrating it about. Just telling it to integrate at 1200s (the final moment) gives the value 5,25e-5 which doesn't really tell me much, other than not being 1 as the integral over the whole T range should be. Could this be the cause of the slow melting?