Initialise Station Test
Purpose
This test performs a repetitive bring up of the system by repetitively initialising a station of TPMs. This is intended as a soak test to uncover any intermittent issues with station initialisation.
Initialisation executes the following steps:
Establishing a connection to the TPMs over 1GbE
Powering ON the TPM FPGAs if they are OFF
Erasing and programming the TPM FPGA Firmware onto the two FPGAs
Bring-up of the TPM 40GbE links
Synchronisation of station to UTC time using 1 PPS
Configure the destination for LMC data traffic
Configure the destinations for CSP data traffic
Configuring calibration coefficients
Configuring 1 second integration time for integrated LMC data
Initialising tile beamformer and station beamformer
Before finishing the routine then performs the following checks:
Verifies the TPMs in the station are synchronised.
Checks the ARP table on each FPGA is populated. This ensures each TPM has been able to communicate with its neighbour TPM and also the LMC and CSP destination, over the 40GbE links.
Methodology
The station is initialised, including all of the above steps and more.
To verify that the initialistion is sucessful, the data rate of packets being streamed to the CSP destination is checked against expected values.
For each TPM in a station, several status bits are checked within the firmware to ensure the signal chain is online.
Recheck that the data rate of packets streamed to the CSP destination matches the expected values after a set number of seconds.
Check that all FPGAs in a station do not have any discarded, flaged or corrupted packets.
The whole test is repeated a specified number of times.