This morning PC-S changed firmware and DAQ software to that of the RIKEN setup

Firmware version /MIDAS/AIDA/FEE_Riken_Apr14_26.bin

DAQ software has been changed from AidaExecV8 to AidaExecV7

nnaida12 is not playing ball -> ASIC4 has stopped outputting any data and ASICs2+3 are missing multiple channels. Not running hot

Suspicions over the power cable as the source of the problem (previous nnaida12 developed noisy ASICs) and tests found this not to be an issue. Neither are the MACB HDMIs as they were swapped for new ones last night.

Then changed the adapted PCB to see if that was a possible cause. No change -> FEE or mezzanine problem?

With old kaptons attached, peak widths in nnaida11 are ~180-190ch with shaping time 2us. Increasing this to 6us brings FWHM to ~170ch, and increasing further to 8us reduces the FWHM to ~140ch.

FAO PC-S -> After lunch, upon clearing the spectra, nnaida12 stopped responding (terminal screenshot below - attachment 19). Brought system to halt. Power cycling resolved the isssue.

A series of tests over the afternoon found peak widths to vary only by ~10-20% when the second length (detector end) of kapton was removed. Removing the connecter PCB showed no improvement over this.

However, removing the Cu tape put on as shielding, reduces the FWHM by a factor of 3.

see nnaida11_wav_newCable.png (attachment 17) for before waveforms -> the Cu tape is clearly doing something we are unaware of.

Attachment 18 shows improvement in just removing the Cu tape.

Re-adding connector PCB, increased widths by ~5-10% and re-adding second length of kapton increased widths by x3-4.

Second length of kapton was tested on its own and was found to have roughly the same widths as the first length, no issues with this cable -> increase in FWHM just due to increased capacitance (possibly...)

Connected a 45cm kapton to each FEE, 3 with the ground connection broken, and the grounded one connected to nnaida13 (attachment 23).

Varied shaping time and found 0.5us to consistently provide lowest FWHM (summary graphs to follow).

With just one length (+connector PCB) of kapton connected to adapter PCBs, waveforms all look very good, too good to explain the higher than expected FWHM.

Upon connecting 90cm of kaptons, noise creeps in to waveforms. Also observed reoccurring periodic transients in some waveforms (attachment 20).

Rise time with new kaptons connected (45cm) noticeably faster that with the old cables -> new show rise time of ~140ns, old cables has rise time ~3us.

To do tomorrow:

connect detector to FEEs using 4 45cm kaptons, assess affect on FWHM.