/FAIL/JOHNSON and /FAIL/TAB1 are very similar and work on the same basic idea where a plastic failure strain versus stress triaxility curve has to be defined. In /FAIL/JOHNSON the plastic failure strain versus stress triaxility is created by defining the D1, D2 , D3 parameters which are used in the Johnson-Cook equation.
In /FAIL/TAB1, the user defines a /FUNCT which represents the plastic failure strain versus stress triaxility curves. In /FAIL/TAB1 you can define different /FUNCT which represent different plastic failure strain versus stress triaxility curves at different strain rates. In /FAIL/JOHNSON the strain rate affect is defined via the D4 parameter input.
So you should not need to use both /FAIL/TAB1 and /FAIL/JOHNSON in the same model since they do the same thing. If you have a plastic strain at failure for compresion (trixaility= -1/3) shear ((trixaility= 0) and tension (trixaility= 1/3) then you can do a curve fit to calculate the D1, D2, and D3 paramters. You could also look at /FAIL/BIQUAD which uses two quadratic equations to define the plastic failure strain versus stress triaxility curves and has some built in starting values for some different materials.