# Maximum flow with Edmonds–Karp algorithm

I am learning Edmonds–Karp algorithm , I formed following flow network, (capacity is described above arrow, where s is source and t is sink.)

If we first follow path S - A - C - T , we will get max flow equals to 1 as we cannot take path S - B - C - T (residual flow from C -- T became 0). I am also assuming that while doing BFS when we reach to t from c, we are stopping our BFS search and returning path. There is D left in the queue but we will ignore it as we have got one augmenting path. On next iteration we will move to B as S -- A edge residual capacity is 0.

If we first follow path as S - A - D - T and then S - B - C - T , we get max flow as 2.

I think, I am missing some basics, I belive algorithm does not dependend on which path I take first.

I'm sure you have a misunderstanding. Try simulating the Edmonds-Karp algorithm by hand on the first example, then edit the question to show the residual graph you've got after it finishes. I'm sure you'll find there remains an augmenting path from $s$ to $t$ in the residual graph. If you don't draw the residual graph explicitly, you might fail to notice the path -- you can't do this in your head without drawing the graph on paper.