how to get cd,cl values of an airfoil correctly?
 

I have worked on these for about two weeks,but i just can't get these values correctly.If i plot these values vs AOA,some particular points(AOA=40-60 or even more)will not go smoothly,and had a rapid change.
I know how a cd,cl graph may look like,but i don't know what the problem is.I have tried changing the viscous models(spalart-allramas,k-epsilon,
k-w and rsm)or decrease the under-relaxation,but it didn't work.Now i will try to get my physical model bigger(inlet and outlet longer),but what i would like to know is,there are a lot of papers that ploted cd,cl values,some of them used fluent,but i don't know how to get a smooth graph.I just get a overall similar graph,but under some AOA,the simulation didn't get a converge solution,even i set iterations to more than 30000.I think many people here knows how to do this simulation,so could anyone give me some advise,
thank you.

my setting:

2-D steady
airfoil=s823
chord length=1m
viscous=k-epsilon
wall function=standard
Re=200000
pressure velocity coupling=SIMPLE
decretization=all second order or second order upwind

Try to use an unsteady solver for big AOA