[Vetroc_daq_sbs] Status

[Carlos Ayerbe]

Just a few remarks:

- we were aware about the short period pulser issue during the fastbus test where we were trying to achieve the maximum rate the PMTs can support. I could affirm that we don't know that answer but we know that the system (as we have) cannot tolerate that. As Evan mentioned, it could be a GTP problem. 

- are you sure HIL is 0.8V? I thought it was 1.8V last time, but you were the one who set it. I think 0.8 is kinda low... but I never played with the LED.

----------------------------

Do you want to introduce modifications in the minimal decoder for the ADC? I have a working decoder (attached) for such ADC. Well... it worked 3 years ago and Alex told me that it is good.

For the rest... totally agree. Thank you!

-Carlos

----- Original Message -----
From: "Evan McClellan" <randallm at jlab.org>
To: "Vetroc_daq_sbs" <vetroc_daq_sbs at jlab.org>
Sent: Friday, October 7, 2016 3:48:43 PM
Subject: [Vetroc_daq_sbs] Status

Everything is back to normal and working properly. Here are a few discoveries:

1 - I am an idiot. My "minimal" decoder had been decoding the same .dat file, regardless of the run number command line argument. This is now fixed. This was the cause of my confusion about the reference channel timing not changing when cable lengths were changed.

2 - THE DISPLAY ON THE PULSER IS UNTRUSTWORTHY. When changing the "high level" and "low level" of the channel A output, we noticed that the numbers on the display were often different that the buttons we were pressing. Checking the output on the scope, we found that the actual output matches the buttons we pressed, NOT the displayed number.
(e.g. try entering 0V for the low level. The display will show 2.0V, but the actual low level of the output is 0.0V)

3 - Setting the pulser period too short will cause the system to not see any triggers. I suspect that this is due to the GTP using 32ns wide bins internally. However, we did not fully investigate this issue. Setting the period to 250 us works fine.

After noticing #2 and #3 above, we tried to find settings on the pulser which produced the best-looking PMT pulses. We arrive at LOL = 0.0V, HIL = 0.8V, WID = 9ns.

-------------

Next steps:
- Modify event-display to read class-based event format (Evan, Scott)
- Connect some signals to ADC, try taking some runs, check raw data, attempt to time-in analogue signals
- modify minimal decoder to handle ADC data
- test out "time-over-threshold" gain-matching idea
- Think about most useful way to store raw data (hit time & length rather than rising edge and falling edge times?, etc.)

Cheers,
Evan


R. Evan McClellan, PhD
Hall A Postdoctoral Fellow
Jefferson Lab
_______________________________________________
Vetroc_daq_sbs mailing list
Vetroc_daq_sbs at jlab.org
https://mailman.jlab.org/mailman/listinfo/vetroc_daq_sbs
-------------- next part --------------
A non-text attachment was scrubbed...
Name: decodeEel.C
Type: text/x-c++src
Size: 3415 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/vetroc_daq_sbs/attachments/20161007/09db2f39/attachment.bin>