We already reported about the status of NC gains. It does seem
critical for applications Q2> 0.1 GeV2
That estimate and the NC energy calibration is not finalized
until we have a better understanding of why the most part of the
L-wall pedestals correspon to positive charge in the adc.
The cause for the offset was thought to be in the cable, the input
patch, the ADC itself. All those combinations were checked but not found
to be true. Instead on friday we realized the following
_ took run 3337 (reference run, all thing being as they "should")
_ took run 3331 when the ADC cables were swapped at the adc, so that
the L-wall was digitized in the right crate and vice-versa.
The R-wall pedestals are about 200-300 smaller when digitize in the Left
crate adc.
The L-wall pedestals (which are typically cut off) turn out to be mostly
positive when digitized in right crate, slot 7. The comparison is attached.
It also turned out that
_ there is a 10 mV DC offset between pin 17 in any L-crate ADC (slots 5,7)
vs pin 17 for any R-crate adc. A similar offsets appears between the two
backplane units.
_ this seems to be significantly higher than similar offsets measured
across crates, racks.
- There is only up to 3 mV dc offsets between pins 17 for ADC's in the
*same* crate.
It is also true that L,R sector crate ADCs digitize the same thing when
nothing is connected to them (some skimpy FB pedestal around 400, with a
width of 3-4 channels max).
As said, swapping a left sector ADC with a new one does not help and,
unfortunately, swapping backplane boards also does not happen.
conclusion:
We could investigate further to find out if this is due to the 2nd level
trigger boards. But that will not provide a solution. It seems that
our problem is internal to L-crate backplane and or power supply, *in
combination* with a small pedestal to begin with for the L-wall
(in fact, for the R-wall pedestals we see a systematic shift but not
big enough to be a problem)
I propose, for the L-wall neutron counter, to use an ADC in the right
crate and break the L/R cabling symmetry. If we want to preserve a spare
adc module, we could swap L,R neutron wall adc cables and change the e-
map as well (that will need some strong labeling language)
-- ________________________________________________________________________________ Tancredi Botto, phone: +1-617-253-9204 mobile: +1-978-490-4124 research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:30 EST