Configuration

IOC Shell

The FastCCD driver is created with the FastCCDConfig command, either from C/C++ or from the EPICS IOC shell.

int FastCCDConfig(char *portName,
                  int maxBuffers, size_t maxMemory,
                  int priority, int stackSize,
                  int packetBuffer, int imageBuffer,
                  const char *baseIP, const char *fabricIP,
                  const char *fabricMAC)

The packetBuffer and imageBuffer parameters specify how many packet frames and images to buffer in the libcin driver. These are directly passed to libcin on initialization.

The baseIP and fabricIP are the ip addresses of the base and fabric interfaces on the camera interface node (CIN). A null “” string sets the default as defined in the libcin driver.

The fabricMAC parameter causes the libcin driver to set the MAC address of the CIN on initialization. This can be used to broadcast the image data by using the FF:FF:FF:FF:FF mac address.

For details on the meaning of the other parameters to this function refer to the detailed documentation on the FastCCDConfig function in the FastCCD.cpp documentation and in the documentation for the constructor for the FastCCD class.

Setting the image size

Due to the multicolum nature of the FastCCD, setting the image size is not trivial. This is due to the nature of the image descrambling and if overscan is used on each supercolumn. For this reason the image size is governed by two paramaters, $(P)$(R)OverscanCols and $(P)$(R)ADSizeY.

The definition used for the image orientation in this driver is that columns are along the X direction and rows are along the Y direction. As the number of columns per supercolumn is fixed in timing to 10 columns, the only free parameter is the number of overscan cols to include and specified by the $(P)$(R)OverscanCols PV. For this reason setting of the $(P)$(R)ADSizeX PV is ignored by the driver. The number of rows defines how many rows per column to process and is set by the $(P)$(R)ADSizeY PV.

The actual image size is therefore given by:

• X Direction (row wise) $(P)$(R)ADSizeY [usually set to 2000 to have some overscan]

• Y Direction (column wise) (10 + $(P)$(R)OverscanCols) * 96 [for no overscan 960]

Auto Configuration of the FastCCD

The FastCCD can be configured by using the built in firmware and settings in the libcin driver. This is controled by the PVs defined in the section FastCCD Auto Setting Commands. The PVs $(P)$(R)TimingName[0..9]_RBV show strings with names defined in the libcin driver corresponding to the modes 1 through 10. To boot into one of these modes, the user should set the PV $(P)$(R)TimingMode to number of the mode required (0 through 9). The PV $(P)$(R)TimingName_RBV will indicate the mode selected.

To boot the FastCCD, the user should write 1 to the PV $(P)$(R)Boot. This will cause the FastCCD to reset the CIN (by power cycling it) and upload the firmware and timing configuration. At this point the user should be able to see the test pattern from the CIN by triggering the detector. The user can then configure the fCRICs by writing 1 to the PV $(P)$(R)SendFCRIC. The bias configuration can be upload by writing 1 to the PV $(P)$(R)SendBias. The latter two steps require the camera head to be powered. This is out of the scope of this driver.

Connection management

As the FastCCD CIN only uses UDP sockets, there is no permanent connection, the driver relies on being able to read such parameters as the serial nnumber of the software to be able to confirm that commuincation with the driver is occuring.