Transcript Leakage current measurements of GEM foils

Helsinki T2 status
Risto Orava/ on behalf of Kari Kurvinen 8.11.2005
• the 1st production GEM under tests
(operational tests)
• the 2nd GEM not accepted due to
persistent leakage current problems in
two foils (all other parts finished) - wait
for components for replacement (and
for further GEMs)
• spare parts needed in future to avoid
delays in the future
Characterisation of GEM foils:
A. leakage current measurements
B. visual inspection
C. optical scanning
Leakage current measurements of GEM foils
• the four segments (A,B,C & D, see Fig.) of the
GEM foils are measured separately by Picoamp
/Voltage source device (Keithley 487).
•leakage currents are measured three times
during the assembly (the foil before and after
framing, then after gluing into a stack).
3 x 3 foils x 4
36 measurements / detector
• current limited by 100 MW resistor (corresponding to the max current of 5 mA at 500V).
• a foil is approved if the current stays at < 0.5 nA for 5+ min @500V
• so far 2 /12 foils found to be bad. These were observed to be temporarily
short circuited during the second measurement! Both exhibited instabilities
during the first measurement, still passed the criteria.
• due to these observations, a graphical display (Labview) was added to the
measurement system to detect & document instabilities as early as possible
(the display was not yet available for the first production GEM).
A bad GEM foil
noise
0.5 nA
A-segment is short
circuited (60 – 150 kW)
B-segment
0 min
TIF
10 min
HV ramp up
B,C & D segments OK
0.5 nA
(Note: the time scale in the plots must
be multiplied by two)
C-segment
noise
6 min
0.5 nA
noise
D-segment
6 min
A ”good” GEM foil ?
HV ramp up
N2 flush
started
in air
0.5 nA
0.5 nA
A-segment
14 min
in N2
A-segment
spark!
> 10 min
N2 flush
stopped
20 min
spark!
in N2
> 5 min
C-segment
B-segment
14 min
12 min
in N2
(Note: the time scale in the plots must be multiplied by 2)
D-segment
noise
10 min
defect in the mask
 0.5 mm
Visual inspection
• only the largest visible defects are observable
• coordinates are recorded approximately
• serious defects due to dust, stains or scratches are
easy to see
number of recorded defects in the six latest foils:
GEM 1 18, 12, 10 (top, middle, bottom)
GEM 2 10, 31, 40
General observation:
the quality of GEM foils is good
(i.e. is significantly improved from the past)
defect in the mask
Some improvements to the quality
checking of the GEM manufacturing
1. Handling procedures of the separate detector
components will be registered into a database
2. Optical scanning of the GEM foils
• scanning of the entire foil with a
resolution of 2400 (3200) dpi feasible
• 0.5 (2) h scanning time per foil
• in a reflective or in a transparent mode
a) the most common defects recorded for the
future reference (joint holes, stains, scratches)
b) blocked holes seen easily in the transparent
mode (also partially blocked ones)
c) tracking of certain defects by software from
the picture file possible (e.g. by measuring the
hole sizes)
Optical scanning of GEM foils
• scanning in mixed mode (transparent +
reflective)
•blue diffuser produces a colour
contrast between the holes and copper
surfaces
1.8 million GEM holes
scanned pictures can be analysed
by using fitting algorithms to
count anomalies and to register
their coordinates
AIM:
a quantitative measure of the
quality of GEM foils
The worst foil so far: Blue curve shows all the
seen objects; the red one only the ”roughly round
ones” (e.g.scratches are filtered out). The number
of defects (red ones > 80) was 240, while only 40
were found under visual inspection (see previous
slide).
defects, dust, scratches,etc