Run Command

Run [/error] [/filename <filename>] [/append <filename>] [/anno]

The Run command does not run a simulation on the currently open schematic but on the specified netlist. Normally a run is initiated using the Simulator > Run menu item. This annotates the schematic then generates the netlist using the Netlist command. Run is then executed specifying the new netlist. The Run command may also be used to run a simulation on a netlist generated by hand or by another schematic editor. Linking Runs The data from multiple runs may be linked together in the same manner as multi-step runs such as Monte Carlo. This makes it possible to develop customised multi-step runs using the script language. Simple multi-step runs may be defined using the simulator's built in features which cover a wide range of applications. The simulator's multi-step features allow the stepping of a single component or a parameter which can define several components. But it doesn't allow, for example, a complete model to be changed, or any kind of topological changes. The script language may be used to control multiple runs of a circuit with no limit as to the changes that may be performed between each run. In such situations it is useful to be able to organise the data in the same way that the native multi-step facilities use. This can be done by linking runs using one of two methods. The tradiontal method compatible with version 8.1 and earlier is to by using the /sweep, /append and /label switches. From 8.2 a newer and simpler method is to use the OpenLinkedRun and CloseLinkedRun commands. By running simulations in this manner, the data generated by the simulator will be organised using multi-division vectors which are similar to 2 dimensional arrays. Care must be taken when making topological changes between runs. Names of nodes that are of interest must always be preserved otherwise the data generated for their voltage may be lost of mixed up with other nodes. Note also that the data for new nodes created since the first run will not be available. The same problems arise for device pin currents. Note that the netlist for a linked run must specify a single analysis only. E.g. a single.TRAN or .AC but not both. Also, do not add .OP lines to the netlist. Linked Run Example using OpenLinkedRun/CloseLinkedRun

** Open liked Run
OpenLinkedRun

Run /label "Run=1" netlist.net... changes to netlist

** second run
Run /label "Run=2" netlist.net... changes to netlist

** third run
Run /label "Run=3" netlist.net... changes to netlist

** fourth and final run
Run /label "Run=4" netlist.net

CloseLinkedRun

In the above, the /label switches are not actually necessary as "Run=n" labels will be generated automatically. Any label, preferably in the form name=value may be used. Linked Run Example using /sweep and /append This is the old method. The newer OpenLinkedRun/CloseLinkedRun is simpler.

** First run
Run /sweep start /label "Run=1" netlist.net
** save group name
Let grp1 = (Groups())[0]... changes to netlist
** second run
Run /sweep continue /label "Run=2" /append {grp1} netlist.net... changes to netlist

** third run
Run /sweep continue /label "Run=3" /append {grp1} netlist.net... changes to netlist
** fourth and final run
Run /sweep finish /label "Run=4" /append {grp1} netlist.net