Working with InterfaceTry.jl. It was not k, but the adaptive time stepping...

Working with InterfaceTry.jl. It was not k, but the adaptive time stepping giving stability for D>1. For now easy fix.

eff_Drop.jl

@@ -225,7 +225,7 @@ function solve_D(geometry,IC,AG,B,u0E,P,D_i,D_o,Δx,Δt,
         Nb = Int((BP-L)/Δx+3); # Position of left ghost point
 
     elseif AG=="Y"
-        if D_i<1
+        if D_i<1 || B=="E"
             Δxi=sqrt(D_i/2*Δt)
             Δxo=sqrt(D_o/2*Δt)
             Ni=Int(round((BP-L)/Δxi))
@@ -362,6 +362,8 @@ function solve_D(geometry,IC,AG,B,u0E,P,D_i,D_o,Δx,Δt,
 
     t=2*Δt
     i=3
+
+
     #time stepping
     while t<=T
        #b(i) and A(i)
@@ -371,7 +373,6 @@ function solve_D(geometry,IC,AG,B,u0E,P,D_i,D_o,Δx,Δt,
             A=make_A(geometry,B,α_i(D_i,Δt,Δxi), α_i(D_i,Δt,Δxo), 
                 Δxi, Δxo, x, D_i, D_o, length(x), Nb, P, 
                 BE_t(round(Δt*(i-1),digits=length(string(Δt))-2)), k)
-            # giving problems in InterfaceTry
         end