|NOTE: Old contents of BU CMS wiki. These (green) pages cannot be edited|
See www.hcal.info for more links to active TWikis
It works! Using 2c34 firmware from 16h 2/19/09. Added NIM TTC output to TTS test gadget. Currently it fires once per L1A at a separately-specified BcN.
Incantation for setup_tts (cms2.bu.edu:src/TTS_tester):
./setup_tts -r 1000 -n 0 -d 3516 -t 1 -w 1000 -g
-w 1000 means the TTS command happens at BcN 1000 -g means allow L1A in Orbit Gap
Here is the DCC script which sets up the TTCvi:
ttc/write 0x82 0xf000 # reset BGO fifos
ttc/write 0x80 0xff64 # enable external orbit, disable triggers ttc/write 0x92 10 # inhibit 0 delay (250ns) ttc/write 0x94 10 # inhibit 0 duration (250ns) ttc/write BData0 0x00800000 # write one word (BCR, cmd=01) to fifo 0
ttc/write 0xa2 10 # inhibit 2 delay (250ns) ttc/write 0xa4 10 # inhibit 2 duration (250ns)
ttc/write BData2 0x20000000 # write one word to FIFO 2 (cmd = 40 laser)
- ttc/write BData2 0x30000000 # write one word to FIFO 2 (cmd = 60 pedestal)ttc/write 0xa0 0xa # enable BGO channel 2 (works for external input) ttc/write 0x90 0xd # enable BG0 channel 0
After running this script, one can immediately type "dcc/read" and see:>dcc/read DCC is in run mode Spy buffer has 0x009c words da201b08 52004cf0 19e00110 05000000 00004802 00340003 4000f082 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 004c06f0 90018e40 00094da2 00520d85 05ff05ff 05ff05ff 05ff05ff 05ff05ff 05ff05ff 0dff0dff 0dff0dff 0dff0dff 0dff0dff 0dff0dff 15ff15ff 15ff15ff .... f5fff5ff f5fff5ff f5fff5ff f5fff5ff 9fed9fed 9fed9fed 9fed9fed 9fed9fed 8f4150f0 f0000082 b0330080 a000004e >
The '52' in the 2nd 32-bit word indicates a calibration type (2). The '05' in the 4th 32-bit word indicates laser setting 5 (selected in the DCC script above).
Tried 2c34... still always get bunch count errors. Install DCC #38 in slot 19 with firmware 2c32 for reference. Reprogram #151 (slot 11) with 2c34.
Latest 2c34 version seems more or less OK, though at initialization one often sees error counts of 1-4 (different from 2c32).
Back to orbit gap stuff: Gap-Set-Laser (TTC cmd 0x80) seems to work, if generated by hand. Bits 15-12 at 0x24 count correctly.
Now working on generating the other commands with L1A in the gap. This is not at all trivial. Finally get BGO-2 on TTCvi working (apparently) with external trigger. Here is the recipe:ttc/write 0x82 0xf000 # reset BGO fifos ttc/write 0x80 0xff64 # enable external orbit, disable triggersttc/write 0xa2 10 # inhibit 2 delay (250ns) ttc/write 0xa4 10 # inhibit 2 duration (250ns) ttc/write BData2 0x40000000 # write one word to FIFO 2 (cmd = 80)ttc/write 0xa0 0xa # enable BGO channel 2 (works for external input)
Add another output to TTS test gadget to fire BGO-2
Jeremy tried 2c33 briefly and reports that BC0 was is apparently being processed correctly by the DCC.
Setup to test orbit gap stuff on DCC. DCC 151 in slot 11, with HTR 16 (top) to spigot 0. Update firmware:** Firmware Revisions: LOG1: 0x000f (15) LOG2: 0x000e (14) LOG3: 0x0015 (21) MIP1: 0x0025 (37) MIP2: 0x0028 (40) MIP3: 0x0028 (40) MIP4: 0x0028 (40) MIP5: 0x0022 (34) XILINX: 0x2c32 (11314) CPLD: 0x02 (status bits=0x00) serial no: 151
Start Xilinx trigger gadget at 120Hz triggers. Observe as expected no orbit length errors. Change the orbit length by one, see errors as expected.
Try above test with DCC v2c33. See orbit length errors no matter what the orbit length setting. Report to Wu...