The monitoring of interlock conditions will be performed on an FPGA target to be detemerministic and fast.
One output interlock signal, TTL level, is generated if the valid range of at least one process variable is exceeded. [NEW COMMENT JUN2012 (JPO):] I do not agree. The control system should be more inteligent than this. E.g. if a noise pulse is encountered in the Macro Pulse an interlock should not immediately be genereated, but trigger a "possible trouble condition". Target temperature and avarage beam intensity should be checked - if either is close to their safe limits then an interlock is immediately generated. On the other hand, if they both are well below limits we can afford to defer the interlock to after the next macropuls - if this is ok then it probably was noise. If it also is high then we generate and interlock. There is a lot of "Artificial Intelligence" (AI) which can be built into the FPGA which probably will be able to take care of more than 90% (even 99%) of the interlocks currently (E119 run) needing a manual respons. See separate sub-document for more on TASCA AI
Additional 8 TTL interlock signals are provided for external systems.
The state of the interlock signal depends on the following process variables:
Macro pulse
Macro pulse frequency The frequency must be less than 50Hz.
Macro pulse width The width of the macro pulse must be greater than 0,2ms and greater than width limit configured by user.
Trafo currents of DT2 & DT3
The electrical currents must be derived from 8MHz signal in combination with range signals signals.
The current is monitored for interlock conditions
Fast IL: Current during macro pulse -> Period measurement
Slow IL: Averaged pulse current, average time is one MP(<8ms).