|
Re: list of variables going into the offline conditions database [message #3887 is a reply to message #3870] |
Fri, 15 December 2006 10:12 |
Ken Oyama
Messages: 7 Registered: November 2006 Location: Heidelberg
|
occasional visitor |
From: *physi.uni-heidelberg.de
|
|
Hello Dariusz,
If you want to reduce the number of items:
1. Clock frequency is fixed to be 10.02 MHz and it will never change .
2. status byte of pad is not necessary to be such big. I think pad status will not have more than 10 states. 4 bits (16 states) are fine. I would put the pad status information into MCM status by extending number of MCM status bits to be 64 bits for example. It is better also because "pad status" normally related to ADC status. MCM has 21 ADC channels but if you say "pad", it can contain only 18 channels. How about to use 21 * 4 bits = 48 bits outof 64 bits of MCM status bit?
3. FEE gain also required to have ADC granurality, means 21 channels per MCM. Therefore it can also be included into MCM status by extending MCM status more. It can not be chamber-by-chamber because we will probably apply lower gain for higher noise pads.
4. What is ADC thresholds? Perhaps it should be more precisely defined?
best regards, Ken
Ken
|
|
|
updated list of variables going into the offline condition database [message #3955 is a reply to message #3870] |
Wed, 24 January 2007 10:33 |
Dariusz Miskowiec
Messages: 70 Registered: August 2004
|
continuous participant |
From: lxial30.gsi.de
|
|
This is the updated list of variables which go into the offline condition database, resulting from yesterdays discussion between Raphaelle, Tom, Christoph, Christian, Anton, and Dariusz. The total parameter volume for one run is about 10 MB.
We decided not to switch from pads (18 per MCM) to ADC channels (21 per MCM) (sorry Ken). This would require too many changes in the software. The additional ADC channels, not entering the digits, can be inspected by analyzing raw data. Pad status can be at least dead, noisy, bridged to the left, bridged to the right.
We removed the supermodule and MCM status bytes. Dead MCM's will be marked by setting the dead bit of all the involved pads.
Config summary file will be generated while configuring the electronics, and put on the file exchange server FXE. The shuttle should get it from there and translate to the individual variables needed for the offline CDB. For the unlike case that more information about the configuration is needed the variable "config id" is a name of the complete configuration file which will be stored in a subversion repository. Tom will define the format of the config summary file, provide the reading and writing software, and provide an example.
Dariusz
-
Attachment: nic.txt
(Size: 5.36KB, Downloaded 878 times)
|
|
|
|