I'm trying to extract LTP and LTD graphs via Sdevice after doing a SPROCESS design for gate-all-around, but I'm struggling with an error that says Newton didn't converge, trying again with smaller timestep.

Some of my solve code is

NewCurrentFile="tmp"

Quasistationary(

InitialStep= 0.0010 Increment= 2 Decrement = 2

MinStep= 1e-5 MaxStep= 1

Goal { Name="drain" Voltage= 0 }

){ Coupled {poisson electron hole eQuantumPotential Contact } }

Transient (

InitialTime=1 FinalTime=2 InitialStep=0.010 MaxStep=0.05 MinStep = 1e-5

Goal{Name="drain" Voltage=3}

)

{ Coupled { poisson electron hole eQuantumPotential Contact }}

Quasistationary(

DoZero

InitialStep= 0.0010 Increment= 2 Decrement = 2

MinStep= 1e-5 MaxStep= 1

Goal { Name="gate" Voltage= 0 }

){ Coupled { poisson electron hole eQuantumPotential Contact } }

NewCurrentFile="V-0"

Transient (

InitialTime=1 FinalTime=2 InitialStep=0.010 MaxStep=0.05 MinStep = 1e-5

Goal{Name="gate" Voltage=-4}

)

{ Coupled { poisson electron hole eQuantumPotential Contact }

Plot(FilePrefix="n@node@_VG_0" Time=(range = (0 1)) )

}

How can i solve this problem??