Superconducting Cyclotron at VECC requires a dedicated cryogenic plant set up for operation of the Superconducting Cyclotron Magnet at a temperature of 4.3 K. The heat load of the liquid Helium Cryostat of superconducting cyclotron system is around 160W @ 4.3K. Also there is a heat load of 70W @ 4.3 K for three cryogenic vacuum panels in the cyclotron beam chamber. Liquid nitrogen flow of 30 litres per hour is also maintained, in the thermal shield of magnet cryostat.

For catering the above cryogenic needs cryogenic plant started in year 2000 with a set up consisting with a helium liquéfier refrigerator of 250W @ 4.5K capacity with its associated sub systems like one helium screw compressor of 50 gm/s flow, one oil removal system of 50 gm/s flow, one 20 m³/hr high pressure recovery compressor, one gas bag of 25 m³ capacity and one helium gas buffer tank of 20 m³ volume and 14 bar maximum working pressure. Subsequently with the progress of cyclotron commissioning and operation job and demand in system reliability some more components were added like one more 50 gm/s helium screw compressor, one more gas bag of 30 m³, two more recovery compressor of 20 m³/hr, two more 60 m³helium storage tanks etc.

One external helium purifier operating at 140 bar and 20 m³/hr flow rate was also added to the system.

Liquid nitrogen set up started with 110 metres of super insulated vacuum jacketed transfer lines, two 2 kilo-litres capacity LN₂ storage tanks and one 12.5 kilo-litres capacity storage tank. Later one more 14.5 kilo-litres capacity tank was added to the system.

Recently one more helium liquefler-refrigerator of 415W @ 4.5K capacity, one helium screw compressor of 50 gm/s flow rate, one oil removal system of 100 gm/s flow capacity and one sub-cooler valve box have been added to the cryogenic set up.

Due to the variation in suppliers and the system introduced the control components and user interface is also different. In this paper the work done to bring everything into a common user interface platform with the help of EPICS is explained.

The first helium liquefier is controlled by means of Eurotherm PC3000 PLC which is handling 130 field I/Os. This liquefier was being controlled by text based user interface. The second liquefier and the sub-cooler valve box are controlled by Siemens S7–300 PLC having 144 field I/Os. The user interface here is having a single station PC based WinCC SCADA supplied by the supplier.

Both liquefiers are designed in such a manner that there is only few initial commands is required to be issued by the operations [1][2]. For example, to operate with the cryostat in cooled down condition the operator gives only the commands of start liquefier and connection to cryostat. Liquid helium level to be maintained in the 1KL Dewar and the maximum allowed turbine speed are set in the set point and can be modified in certain range by the operator. The plant automatically adjusts the refrigeration power according to the demand by varying the turbine speeds to keep a constant level in the Dewar. Once the liquefier is operating in steady state there is no human intervention is required for the operation of the plant.

The impure gas handling, turbine cooling, buffer tanks, liquid nitrogen Dewars and transfer system and turbine water cooling system are controlled by Schneider PL7 PLC which is having 292 field I/Os. [3]. All the systems related to Schneider PL7 PLC are operated based on mode of operation commands from the operator. All sequence, set points and safety features are programmed in the PLC logic. SCADA for this was initially designed using Microsoft Visual Basic software by us.

The external helium purifier is controlled by Schneider Unity Pro PLC having 52 field I/Os. Operators have to select and issue command of the Purification or Regeneration mode and rest sequences are automatic.

Sequential Flow Chart (SFC) language is used for all PLC programming.

Cryogenic plant control can be divided into three layers. The device layer consisting of PLCs and control panels as explained above is the lowest layer. The middle layer is the EPICS input output controller (IOC) layer which works as an interface between the device layer and the user interface layer. The top layer is the user interface layer which is the SCADA consisting of different EPICS user interface applications.

All communication is based on Ethernet based control LAN.

The requirement of the SCADA was such that, which can give graphical user interface at a common platform as well as exchange data between systems. Also the user interface computers were required to be put at different locations spread over two buildings. The limitation of number of channels or tags which a common practice with the commercial ones was also not wanted. As the expansion of the system is gradual and flexible the SCADA have to be changed from time to time based on the requirement without any dependency on external agency.

After evaluation of requirements, we selected Experimental Physics Industrial Control System (EPICS) [4] as suitable system which fulfils all our needs. EPICS is a set of Open Source software tools, libraries and applications developed collaboratively and used worldwide to create distributed soft real-time control systems for scientific instruments such as a particle accelerators, telescopes and other large scientific experiments.

We first tested EPICS lOCs (InputOutput Controllers) on simulator software for Modbus TCP protocol for communicating with Schneider and Eurotherm PLC. After testing successfully we implemented this in a linux PC in our network to run the IOC to communicate with Eurotherm and Schneider PLCs. We also used IOC for Siemens PLC with TCP communication. There was no Ethernet port in Eurotherm PLC. So we have put a Modbus RTU to Modbus TCP converter module of Advantech make. Watch dog program is also implemented to generate alarm to operator when there is a communication failure for more than 5 seconds. The control system architecture is shown in Figure 1.

The Graphical User Interface (GUI) is made by using EPICS Motif Editor and Display Manger (MEDM) tool. All the process parameters of all the systems are displayed here for monitoring with a proper navigation buttons. Same GUIs are running in three different PCs so that different systems can be monitored together. Process line colour change occurs according to the process so that the process can be viewed at a glance. Figure 2 shows the GUI of liquid helium system overview.

EPICS Alarm Handler is used to handle the Alarms of the channels observed from all four lOCs to sound an alarm to the operator to take preventative action. For example if the water level in the turbine cooling system tank then the alarm handler sounds an alarm with red coloured indication against the particular alarm. The operator then acknowledges it and takes necessary action to fill up water or change the pump or repair the gland packing leakage etc. Screenshot of alarm handler is shown in Figure 3.

EPICS Channel Archiver is used to archive the values of 376 important parameters from all four lOCs. The archive values are stored to hard disk only when the value changes. Standalone archive server is used to host the data for viewing in EPICS Archive Viewer. EPICS Archive Viewer is available at three computers in the network. Process sequence step numbers are also achieved to diagnose the system behavior for understanding the system dynamics.

Pure gas stock calculation is continuously done by EPICS IOC by monitoring parameters from different systems like pressure and temperature from gas storage tanks, liquid helium level of Dewars, valve box and cryostat etc. The calculated value is the pure gas stock in normal m3 and archived in the Archiver. The trend gives us the idea of helium gas leak rate from the total installation. Screen shot of Archive Viewer showing pure gas stock is shown in Figure 4. The first downward slope in this figure interprets a decrease in pure gas stock which was due to leak due to failure of a pressure gauge in a compressor station. The increase trend towards the end is due to filling of buffer tanks with procured pure gas.

All the applications are set in the startup script so that in case the PC got switched off due to UPS power fail or some other problem all the¹ applications get restarted. PCs BIOS are also set such that PC restarts itself after the power failure.

We had operated the system continuously for 22 months and 13 months without any interruption with the small liquefier. Presently total control system is running undisturbed and without any human intervention, since 19th July 2010 with the new helium liquefier and valve box commissioned. The superconducting magnet of the cyclotron had been cooled down to 4.3K, and filled up with liquid helium and plant is operating in refrigeration mode. The required current is applied in the superconducting magnet and Ne3+ beam tuning in progress. The cryogenic control system is integrated with the main cyclotron control system through EPICS so that cyclotron operator can also monitor cryogenic system status from the central control room.

The author acknowledges sincere thanks to the plant operators and all employees of cryogenic plant at VECC for their continuous feedback and co-operation during the period of control system development, testing and up-gradation. We are also grateful to the entire EPICS community for their ideas, help and support.

REFERENCES