I am attempting to use the standard "Create Polymer Network" protocol supplied under the Materials Studio Collection and am running into a fatal error almost every time I attempt to make a network. I am using the monomer TGGDM and the linker 4,4' DDS to create a polymer network with loadings of 40 and 20 (as a start). In the protocol, after the initial molecules are set up as input, the data flow moves to the Create Amorphous Cell subprotocol. After the cell is completed, an Equilibration protocol is initialed that runs NVT MD. This usually leads to an error

Dynamics calculation - Deviation in total energy (1.33641e+006 kcal/mol) between successive steps was greater than the specified maximum (50000 kcal/mol).

I have tried reducing the time step as is suggested, but it almost inevitably leads to the same error being generated. Essentially, the Amorphous Cell Builder is constructing an amorphous cell in a very high energy state and when the NVT dynamics are run, the energy difference between the initial and subsequent step is large throwing the error. Since the ultimate cause is the Amorphous Cell module constructing a "poor" cell, I would like to branch upon the above error back to the Amorphous Cell module and have it try building another cell to replace the first "failed" cell. An overview of the protocol is reproduced in the first figure. The equilibration subprotocol is where the error is thrown. The second figure shows the details of the Equilibration subprotocol. The initial protocol runs NVT and then NPT dynamics. The error output the subprotocol containing the MD runs is directed to a Throw Error component which stops the protocol and leaves the above error. If the large energy jump occurs, I would like to branch back to the Build Amorphous Cell protocol and generate another configuration. I have simply connected the error output of the NVT dynamics protocol to the input of the Create Amorphous Cell. I realize this is probably incorrect and am curious what the correct way to do this would be. It would also be useful to let the user know that the protocol has looped back to build another cell and at the same time allow the user to specify a maximum number of attempts. Thanks for some help on this issue.