LtStruct output data and petals qualification

1
LtStruct output data andpetals qualification
Before reporting a problem
LtStruct manual (continuous update)
. http//www-ekp.physik.uni-karlsruhe.de/zhukov/C
ms/LtStructman/ltstructsoft.html
Questions about software? zhukov_at_ekp.physik.uni-k
arlsruhe.de wim.beaumont_at_ua.ac.be militaru
_at_fynu.ucl.ac.be (SlowCtrl)
About setup? Same guido.dirkes_at_iekp.fzk.de
2
Structure of the ROOT file
3 Records are taged(more Records stored)
SECLTFIRST, SECLTCOLD, SECLTLAST
One Record 70Mb (28 modules)
Record
Record
Record
Header TNcdb....
Record4
SecLtSummary TNstructid....
Environment TVtempbox...
Module1
Module1
Up to 28 modules
Module1
Module1
Module1
Summary Tnmodid Tibadch TIdaqerr...
Environment
TimeTune
PeakInvOff
CommonMode
PeakInvOn
AOHTune
Test folders are created if such a test is done
Pedestal
DecInvOn
Noise CMSubtractedNoise...
DecInvOff
File structure and names are compatible with ARC
and module Lt
PulseShape PulseHeight...
Depending on number of Records file can be up
to1Gb.
.......
3
Root file more details_1
Detailed description of all data can be found in
HowTo manual
Which are important? TimeTune/THDelay_0,1,2
tick marks TH1F per AOH ch.(should be
above 100 ADC) Summary/TVvtickamp
vector of tick amplitudes for each apv (
see amplitude) Summary/TVfinedelay
PLL delay (ns) should be 30-70 (default in
Petal.xml) AOHTune/THTAOHBase
baseline offset versus AOH base (bug in
ver.28!) Summary/TVvaohbias
vector of aoh bias for each AOH
channel Summary/TVvaohgain
vector of aoh gains (default 2) fro each AOH
ch) Pedestal/Frame
Frames(shot) for all apvs (watch
baseline!) Noise/CMSubtractedNoise noise
TH1F CommonMode/THCmn_0,1,2 TH1F common
mode distribution (depends on modcmn) PulseShape/
PulseHeight cal amplitude TH1F
(positive, 20-50adc) PulseShape/Profiles/TPcalprof
cal pulse shape ( see shape, check
saturation!) PulseShape/RiseTime
rise time TH1F (should be /- 5 (1.25ns) unless
saturation) Calibration/CalAmp
cal amplitude TH1F (can be negative for inv
on) Pipeline/THpipe_0,1,2,3,4 pipeline
addresses scan (should be scared up to 192)
4
Root filemore details_2
Summary/TGndaqerrt TGraph Ndaq errors (no
frames) versus time (should be flat) Summary/TGni2
cerrt TGraph Ni2c errors(hybrid,aoh)
versus time Summary/TGnapverrt TGraph
Napv (error bit, address,header) versus
time Summary/TLval status flag o
if OK, -1 if module was excluded due to errors
or low
tick marks (check optofibers) Module/Environmen
t/TGtemphybdcu1t,2,3 Tgraph T measured by DCU
versus time (sec)

1-internal dcu, 2-ext. hybrid, 3- ext sensor
Module/Environment/TGvt, it
TGraph HV and current (from HV supply) versus
time Module/Environment/TGiv
TGraph IV curve (Ileak calibration in
settings.cml) Environment/TGtempboxt
TGraph Temp in the box (as measured by
TRHX) Environment/TGtempcoldt
Temp in the Cold plant Environment/TGtemptagt
TGraph tagged temperature
Environment/TGhumboxt TGraph
humidity box (exteranl measurement) Header/TNcdb
tag for Db
(SECLT)
Qualification is not implemented online. i.e
badchannelflags are filled according to relative
cuts ( if in settings.xml all absolute cuts are
equal to 0) For final qualification use
defectAnalyzer
5
Root file examples
Small apv baseline (large Vpsp35) causes
saturation of CalProfile in invON and bad
RiseTime. -gt decrease calamp and decrease Vpsps.

Frames
Profiles
RiseTime
APV errors but good pipeline scan! Corrupted
frames? Error bit? Under investigation
Bad pipeline scan, TLval-1 excluded Napverrgt1000(
set in xml)
6
Quality of data
See summary for all petals in G.Dirkes talk
1. Saturation affect RiseTime and to less
extent PulseHeigtht Petal80 badly set baseline
(Vpsp)
Petal71 (Strasbourg) perfect
7
Petals qualification proposal
Proposal 28.01.2005 http//www-ekp.physik.uni-kar
lsruhe.de/zhukov/Cms/petalaccept.ps
Bad channels (Ntot) and modules grading A
Ntotlt1 lt50 grade B B Ntotlt2 no grade
C C Ntotgt2 lt50 of grade C D the rest
HV performance Ileak(HVtest) at Twarm Ileak
from Db or if not accessible 1.5 uA/sensor. A
Ipetaltot(temp,petal type) lt Imax1 Nsens
Imax11uA B IpetalltImax2 Nsens Imax25uA C
others
Nerrors Napverrors (pipelineaddress)
ltNmax NdaqerrorltNmax Ni2cerrortotltNmax
Final grading offline in Defect Analyzer see
J.Fernandes talk WORKS FINE... TryIt!
Cooling dT masured by DCU should be in the
range (50) defined by the T map, see Thomas
talk.
8
Petals qualification comments
Bad channel and module qualification. -use
normalized(per apv) parameters (noise, calamp,
risetime) and absolute cuts. -need a cut on
absolute pedestal to avoid saturation now if
module is excluded (Tlval-1) it is graded as C,
need to degrade the petal
HV -qualify by measurement at fixed voltages
(450V) -no temperature corrections , have to use
tempdcu3 to calculate Tsensor -Ileak measurement.
Need a map of the sharing of Ileak channels - IV
run no qualification on it so far.
Errors and tunning -data quality tune of AOH at
different temperature -reliability?(see Wim's
talk) -no real selection on Ndaq,i2c,apv
errors-gttricky! (setup and petal problems) -use a
cut on absolute tickmark amplitude (gt100ADC)
Cooling(important!) -temperature map for each
cold sink is still missing, has to qualify on it
(dT) - stabilty check (Tgtempdcu) probably is
needed
9
Petals qualification comments
What to do if 1. One qualification(first,cold,la
st) test(ped,pulseshape) is corrupt for some
modules. check other records, can be
tunning/settings problem-gt repeat the tests at
the same conditions with a short scenario. In
DefAna need possibility to combine two file. 2.
hardware problem (interlock, humidity,etc) as
above , repeat the missing part of scenario and
save in another root file to be combined with the
first one. 3. Identified problem (bad
module,AOH, bad termocontact) -gtfix the problem,
exhange, repeat the whole test.
General Need publicly available for each petal
initial seclt rootfiles, logfile, scenario,
README(test related), root file processed by
defana , summary txt from defana Can we have
(apart of elog) the standard HTML(php) script for
each PIC?