How about providing your model, an example calculation, a picture, a sketch, really anything to show the issue, your description is quite vague as to what you are exactly modelling, how you are calculating stress to compare, etc, etc.
I don't mean to be rude but do you think you could determine the cause of what you are seeing from the explanation you gave?
Further out in the span is the stress more in line with a hand calc?
I suspect due to the discrete supports and lack of meshing, that load is concentrating or tracking towards these points, so it won't agree with a hand check right where your cursor is located. Further out in the span I'd expect with some fine meshing that it would be closer to your hand check where the stress is more uniformly distributed.
You don't say what hand calculated stress you are expecting to compare to the 65 (MPa presumably) to?
Try dividing it so it's a 1:1 aspect ratio. Keep dividing it until it starts to agree with your hand calculation somewhere out in the span remote from the support. Stop checking it at the support for the reasons I noted in my second reply, it won't be correct.
Your finite element model is quite wrong, so it is only normal for you to not get the expected results.
1st, the aspect ratio of the finite elements should be close to 1:1, as mentioned by the previous poster.
2nd, you should input restraints to all nodes on the boundary condition side of the shells.
I believe you need to study a lot more about finite elements and FEA before you venture yourself in these tools.
