When I removed all the phototube TDCs from the FASTBUS crates to add a
hardware jumper for buffering, I noticed that the left sector slot 11 TDC
(LADS 15cm bars) was missing the jumper for allowing hardware clears from
the backplane.
Since we started using the 2nd level trigger last fall (November 2003), we
have been issuing a clear pulse to the FASTBUS crate backplanes for all
level 1 triggers that fail the level 2 test. The 1881 ADCs and the 1877
drift chamber TDCs can be set to use the backplane clear with
software. The 1875 TDC (phototubes) requires a jumper to be soldered on
the board. This was done for all the TDCs in the crates last fall, but
evidently I missed the spare TDC in the cabinet. Michael installed the
spare TDC sometime in February 2004 when he noticed we were missing the
least significant bit on the existing TDC.
I don't believe this actually causes a problem with the data we've been
taking, for two reasons:
1) We have been running without using the TDC buffers. If a new START
comes in before the old one has been read out, I believe the old event gets
overwritten. The manual is not very explicit on this. It just warns that,
in non-buffered mode, you can't allow a START until you have read out the
previous start, because you will read corrupted data. I think this means
that you read something other than what you expect, NOT that the data words
are scrambled.
2) If there WERE a problem, then in recorded events involving L15, the ADC
would be from the bar that fired on the current event (bar 5, say), and the
TDC would be from a bar that fired during some previous rejected event (bar
9, say). Thus, plotting the ADC for bar 5, cut on TDC 5, would show mostly
pedestals. Michael looked at some data and doesn't see this.
Karen
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:31 EST