I agree with Torben's assessment of this endgame.  Ra5 is a very passive move, and it doesn't achieve any other goal than protecting the a-pawn, although you still will be winning.  By playing Rc3, Black forces the pawn exchange and gets his rook positioned behind the enemy pawn after Rc3-h3-a3.

   Thanks Torben, this is very similar to what my assessment was (yet you described it more precisely), the reason I asked was that when I was at that point in the game I had to leave and I thought, ok so my plan is to play Rc3 as my next move tomorrow, let's just check what Fritz thinks about it, and I was kinda confused to see that Fritz gave Ra5 a higher eval...these numbers don't matter so much at this point of the game I guess, but still needed to understand why that would be...anyway, I played Rc3 today and it didn't take long till Fritz resigned.

Simon

Ra5 looks awfully passive. In rook endgames, activity is usually more important than material, so I would look at other continuations first.

The idea of 1...Rc3 looks much better, when the sequence 2.Rxa4 Rxh3 is forced. After that, I would expect White to launch his a-pawn to create counterchances by 3.Ra8 and a3-a4-a5-a6-a7. Obviously, once the pawn takes off, you will have to put your rook on a3, and when it has reached the 7th rank, you won't be free to move your king up the board on an open file, because then a rook check followed by a8Q is due. So in case Black follows the a-pawn-plan, I suggest advancing the kingside pawns while keeping the rook on a3. In this process the f-pawn will be used to protect the king from lateral checks as well as from checks from the back-rank: the latter will be especially important once White's a-pawn has reached the 7th rank!

Without checking every detail it seems to me that you should be quite easily wining in that case.