Date Created: Friday, April 22, 2005 7:07:17 PM CDT Date Saved: Friday, April 22, 2005 7:07:18 PM CDT Category - Topic - sequence number: Trigger/Log - Trigger_Log - 403976 Operator(s): Sabine Lammers Keyword(s): :Trigger: The commissioning team has just established communications from the TAB to the SBC through the VTM/VRB readout crate!!!! Here's a hex dump of the aaaa-bbbb pattern from the TAB for what I can tell to be one event. 0000000 007b 0000 002d 3b19 4567 0123 5678 1234 0000010 beef dead a5a5 a5a5 0001 0000 0000 0000 0000020 0068 0000 01a0 0000 0037 000a 0902 4908 0000030 83ee 0022 00c0 00c0 0000 0000 0000 0000 0000040 0000 0000 ffee 0150 aaaa 0103 aaaa aaaa 0000050 aaaa aaaa aaaa aaaa aaaa aaaa aaaa aaaa * 0000070 aaaa aaaa aaaa aaaa aaaa aaaa 6565 aaaa 0000080 aaaa 6565 aaaa aaaa aaaa aaaa aaaa aaaa 0000090 aaaa aaaa aaaa aaaa aaaa aaaa aaaa aaaa * 00000b0 6565 aaaa aaaa 6565 aaaa aaaa aaaa aaaa 00000c0 aaaa aaaa aaaa aaaa aaaa aaaa aaaa aaaa * 00000e0 aaaa aaaa aaaa aaaa 2d6f aaaa 0000 0000 00000f0 0000 0000 0000 0000 0000 0000 0000 0000 0000100 0000 0000 2d83 03ff bbbb 01aa bbbb bbbb 0000110 bbbb bbbb bbbb bbbb bbbb bbbb bbbb bbbb * 0000140 7676 bbbb 6565 7676 6565 6565 bbbb 6565 0000150 bbbb bbbb bbbb bbbb bbbb bbbb bbbb bbbb 0000160 bbbb bbbb 6565 bbbb 6565 6565 7676 6565 0000170 bbbb 7676 bbbb bbbb bbbb bbbb bbbb bbbb 0000180 bbbb bbbb bbbb bbbb bbbb bbbb bbbb bbbb * 00001a0 bbbb bbbb bbbb bbbb ff06 bbbb 0000 0000 00001b0 0000 0000 0000 0000 0000 0000 0000 0000 00001c0 0000 0000 007b 0000 002d 3b19 4567 0123 00001d0 5678 1234 beef dead a5a5 a5a5 0001 0000 00001e0 0000 0000 0000 0000 8297 76f3 The dump has been sent to Todd so he can have some data to play with using his unpacker. We need to still decode the dump and make sure the headers, trailers and data are exactly what and where we expect. Who knows what "dead" "beef" means? To get the communication over the VTM and VRB working, we had to do the following: - switch to 20 bit mode on VTM - switch to "D0Trigger application" on VRB - turn on both VTM channel 1 and channel 2 in VRB channel mask (for upper and lower bytes) - plug the TAB fiber into the right channel Date: Sat, 23 Apr 2005 19:47:21 -0500 (CDT) From: Sabine Lammers Subject: understanding VRB output I had the good fortune of being on shift with Dave Buchholz today and he helped me understand how to decode the data we managed to read from the TAB yesterday. It is in the spreadsheet attached (if someone opens the sheet in windows, please confirm that it worked, as I edited the original spreadsheet using OpenOffice). In columns B-I you will find the HEX dump in the form we wrote yesterday, for the first 2 (and a bit) events. You should read the data from left to right, like a book. In column J, the data set is just copied into one long column, but with every pair of words swapped. This is apparently a known feature of the either the VRB or SBC (I forget which one). So, the first two entries of column J are Row 1, Columns C and B. The second two entries are Row 1, Columns E and D, etc. Now jump to Columns N and O. These contain the first event (the second event is in Columns P and Q), read lengthwise. Column N has the SBC header, the VRB header, the TAB header, TAB data, TAB trailer, and the VRB trailer. I believe Column O is the SBC trailer, because the header and trailer are supposed to be identical. In Column T, I try to demarcate where I believe the boundaries between entries from SBC, VRB and TAB are located. Column U contains comments on what the data are represented in Row N. The aaaa and bbbb patterns make sense now (I was originally expecting aabb, since we have aa for EM bytes and bb for HAD bytes) because the VRB reads one channel at a time. The four MSB's of the 20-bit word are not accounted for. The upper byte of the remaining 16-bit word is sent along the first channel, and so you have all the EM data - aa's - together. The lower byte, containing the HAD data is sent along the second channel. The documentation of the VRB header is just a duplicate of what it was for the SMT VRB's, and may not be correct. For instance, the slot number for the VRB is 10 in the crate, but the data in row 22 of the spreadsheet claims it is 0. It's more likely to be the 0a in the row 21. The firmware version looks accurate, though. Have a look and if there are mistakes or you can provide more documentation for what the entries mean, please send it to me and I will update the spreadsheet and put it in some web location. Cheers, Sabine