I have read and applied the recommendations in this arcticle before my first post.
Still it doesnt work.
Hmm, well I thought that would have been the problem. The fact that you were able to get the cluster configured in the first place tells me that the communication over the heartbeat is working. The cluster configuration uses unicast traffic. Usually when both firewalls are active, it means that the heartbeat is not getting passed to the other firewall (heartbeat uses multicast).
I think that tcpdumps on both firewall's heartbeats would make sense. Also it might make sense to engage support at this point.