Hi Tivadar,
Let me quickly answer these:
• Q: no. and type of links to be readout
- 2 VTRX links, 2 LpGBT, 4 H2GCROCs in total. Basically, what you were testing in December with Shihai times 2. This would cover the latest prototype
• Q: how do you plan to analyze the data
- Best would be already O2, this is why we are trying to understand where is the design currently. Even if it is in raw data format on the FLP, that would be good enough for us to analyze it already. Some of the O2 will require also some development from our part, that’s one of the reason why are are pushing for it.
• Q. in particular, what do you need in terms of slow control, fast control, + trigger operation
- As far as I understood, most of the slow control requires python scripting. Shihai and Ian Pascal would be great help there, as they are experts in python. Shihai already developed a slow control GUI for the KCU fw, but this could be adopted also in the CRU with the underlying communication changes. For trigger, I think we would like to receive some external trigger for the CRU and then send it through the CRU towards the LpGBT-H2GCROC. In principle it is also covering the fast commands. The H2GCROC requires the fast command to start to operate, then we could test in the lab internal injection (which is two fast commands for pre-pulse and then trigger readout). If we would see some internal injection pulses, then we are convinced we could progress towards testbeam.
• Q: In general we would like to give as much of the final service from the central team(s) as possible, instead of supporting / working with your own developments. I guess you would also be happy if we provide the tools, and you do not have to do a lot of developments on your own. There are several functions already in place, we are stuck with orthers, and still a lot of testing and verification is to be done until the full stack (CRU WF + low level SW + higher level services) can be released. However if the CRU and FLP teams know your short term requirements, they can be more focused on providing you with the required functionality by the summer.
- Our main motivation to ask you also to know where is the current progress from the CRU and DAQ side. Hardware-wise, we are ready, you saw it in December, we could have two of those setups connecting to one CRU. This would cover the whole FoCal-H and we could compare it with the KCU readout and reconstruct the performance of the prototype. At this stage, it is all through the VLDB+ board, so we can access the slow control through the raspberryPi also, but as far as I understood it was working through the CRU already in December. Also, the synchronization is ready in the CRU after the December test, so it looks like to me it was almost ready.
- My idea was to send again Shihai and add Ian Pascal to the team to help to make an internal injection readout. Technically, it would require sending 1 prepulse fast command, then wait about 16 clock cycles, then send 10xtrigger fast command now to readout the full waveform (this will be one trigger later in the real running). There is also a CRU hw and FLP sitting with Tommaso at CERN, it might be good to make that also operational, so then we could make some readout tests in CERN. This is all basically preparation for the Aug testbeam.
Best regards,
NN
