Production Readiness Review at PNPI
Chamber assembling: Main production and quality control steps
30 January 2004 – G.Alkhazov
1. Input components control and preparation for assembling2. Gluing the bars, control of the results of gluing3. Wire winding, pitch control4. Wire gluing, pitch control5. Wire soldering, tension control6. Soldering the HV capacitors, resistors, connectors7. Assembling with O-rings, HV tests with air and working gas8. Gluing the assembled chamber9. Gas leakage test10. Final HV test (current and noise)11. γ-Ray test (checking gas gain uniformity)
Input components control and preparation
1. Panel thickness and flatness control is done by the production firm. Provisional thickness control at PNPI: d = 9.0+/-0.2 mm
2. Processing the anode panel edges for precise comb fixation. 2 man* hours per chamber
3. Panel surfaces polishing with special abrasive sponges 4 man*hours per chamber
Panels
Input components control and preparation
• Control of the bar thickness d=2.40 μm (+50, -150) μm This thickness should be provided after processing the
bars (2.4±0.15 mm ) at CERN2. Processing the “noses” on the anode bars at PNPI It is a big job – about 5000 bars in total. It would be better to do this work at the company3. Polishing the side bars and cathode bars for better
sticking to the suction cups. It is a time consuming work. Would it be possible to receive the bars with better surface quality?
Bars
4. Washing bars with alcohol
5.Visual checking of the printed circuit boards on the bars
All these operations take about 4 m*hours per chamber ( not including bar “noses”)
Input components control and preparation
Bars
Gluing the bars
Precision gluing with a gluing machine
First, the bars are put on the gluing machine frame and fixed with the suction cups.Then, the glue (Adekit 140A) is put on the bars, and the frame is turned over and pressed to the panel. The distance between the surface of the frame and the panel is determined by the spacers. Glue polymerization requires at least 6 hours
1 chamber needs gluing of 8 panel sides.With 4 gluing tables operating in two shifts one can prepare 4 anode panels per day needed to produce one chamber per day. It takes 8 m*hours in 2 shifts. Critical procedure in the production scheme
Check of the thickness of the glued bars
The distance H from the surface of the bar to the surface of the panel is measured in 24 points along the panel perimeter with a micrometer. The values <H>, Hmin, Hmax are presented in the following table. This measurement takes about 2 m*hours per chamber.
MCPanelGapResult
UP UP UP UP Down Down Down Down
Panel_ID Panel Position Hmean,[mm] Hmin,[mm] Hmax,[mm] delta_H,[mkm] Hmean,[mm] Hmin,[mm] Hmax,[mm] delta_H,[mkm] Date_Tested .
13 Pos 1/5 2.48 2.39 2.55 160 24-DEC-2003
16 Pos 1/5 2.44 2.4 2.5 100 19-DEC-2003
18 Pos 2 2.44 2.4 2.5 100 2.46 2.41 2.51 100 19-DEC-2003
22 Pos 4 2.45 2.43 2.48 50 2.46 2.4 2.49 90 21-DEC-2003
72 Pos 3 2.47 2.43 2.5 70 2.46 2.4 2.51 110 24-DEC-2003
77 Pos 1/5 2.48 2.4 2.56 160 24-DEC-2003
15 Pos 1/5 2.49 2.41 2.58 170 24-DEC-2003
55 Pos 2 2.47 2.42 2.51 90 2.46 2.43 2.49 60 26-DEC-2003
69 Pos 2 2.47 2.41 2.52 110 2.45 2.37 2.51 140 26-DEC-2003
76 Pos 1/5 2.47 2.35 2.54 190 29-DEC-2003
59 Pos 2 2.47 2.41 2.52 110 2.45 2.37 2.51 140 30-DEC-2003
71 Pos 3 2.48 2.4 2.57 170 2.47 2.41 2.57 160 05-JAN-2004
83 Pos 1/5 2.47 2.41 2.52 110 05-JAN-2004
93 Pos 1/5 2.49 2.4 2.61 210 05-JAN-2004
21 Pos 1/5 2.48 2.4 2.54 140 08-JAN-2004
58 Pos 2 2.48 2.39 2.53 140 2.47 2.44 2.5 60 12-JAN-2004
73 Pos 3 2.47 2.43 2.53 100 2.48 2.4 2.52 120 12-JAN-2004
90 Pos 3 2.48 2.41 2.52 110 2.48 2.43 2.52 90 13-JAN-2004
60 Pos 2 2.47 2.41 2.52 110 2.48 2.45 2.51 60 13-JAN-2004
61 Pos 2 2.5 2.41 2.54 130 2.5 2.45 2.52 70 15-JAN-2004
http://oraweb03.cern.ch:9000/pls/muchdb_dev/WebMCPanelGapResult
Gap Measurement Results
All panels Last panels
Wire winding
Wire winding is performed with the winding machine.Proper fixation of the combs is crucial for obtaining small pitch jitter and for putting the wires in the right positions on the printed circuit boards. The combs are adjusted in the longitudinal and transverse directions For the moment this procedure takes about 2 m*hours/panelBefore winding the wires (of 30 μm diam.), the anode panels are washed with alcohol, and the guard wires of 100 μm are soldered at the edges of the winding zone.Winding is performed with the speed of 5 turns per minute.It takes about 2 hours for winding one panelBy now, 9 anode panels have been wound.
Wire gluing
Wire winding has been done, the wire pitch measuring is performed (with a wire pitch measuring machine). At this stage the wire positions may be corrected manually. If the results are O.K., then the wires are glued to the anode bars (Adekit 145A). This operation is done manually: first, the glue is put on a mylar tape, then this tape with the glue is put on the anode bar with the wires.
The hardening of the glue takes about 1 day. After hardening the glue, the tape can be easily removed The pitch is checked again.
Wire soldering
Before soldering the wires, the wires (excepting the zoneof soldering) are covered with a protective shielding.
Soldering the wires is performed with a soldering machine.Including the time for preliminary soldering machineadjustment, it takes about 3 hours per panel (2 sides).
The quality of soldering is checked by visual control andby measuring pad capacities.
Wires on 2 panels for 1 chamber are soldered in 1 day.
Measuring of the wire pitch
Soldering of the anode wires
Wire pitch distribution
Anode plane 22, side UP
Wire pitch distribution
Anode plane 22, side DOWN
Wire pitch distributionAnode 22 UP
ANODE 22
Wire pitch distributionAnode 22, Down
Results of pitch measurements
For the first 12 wound planes,98.8% of the pitches are around 2.0 mm within +/- 50 μm;1.2% of the pitches are within +/- 80 μm.In 2 planes from 12, all pitches are within +/- 50 μm.Pitch jitter may be further decreased.Pitch measurement is rather quick: about 15 min. per plane.
Wire tension measurements
Wire tension measurements
Wire tension measurement is performed with the tensionmeasuring machine. From the stability consideration, thewire tension should be >50 g. Elasticity limit: 130 g.Average wire tension is from 63 to 67 g. (for 99% of wires).For 90-95% of wires the tensions are within +/- 8% (5 g.)from the average value.For about 1% of the wires (at the comb ends) in the woundpanels the wire tension is smaller than the average value down to 20%.This effect should be eliminated in future.It takes 6-8 hours to measure wire tensions in 4 planes (!).We have 6 pairs of combs, it is desirable to have more.
Preparation of the panels for assembling
The wire ends (protruding outside of the soldering spots)are removed.A careful check is performed to be surethat no wire ends have been left.The wires are covered with a protective shielding.Then the HV capacitors, the resistors, the connecters, andthe HV and ground wires are soldered. (This proceduretakes about 3 hours per panel). Before the chamber assembling, the wires are cleaned withionized nitrogen. For these operations we need 2 operators to produce1 chamber per day.
Gas leakage tests
Chamber assembling and HV tests
First, the chamber is assembled with O-rings.A HV test is performed, the chamber being filled with air.We had current of 20-30 nA at U=2-2.4 kV.(The air humidity was 25%).
If the HV test is O.K., the chamber is filled with the working gas (Ar/CO2/CF4; 40/50/10%).
After 1 day of operation of the chamber kept at HV, we had current of 10-15 nA at U=2.9 kV (air hum. 25%).The HV test having been done, the O-rings are removed, and the chamber is glued (Araldite 2011).
Final gluing of the chamber
Final HV test (current and noise)
Chamber 1. Gaps 3 and 4, pad 12. HV=2.75kV.
Dark and 137Cs (1mCi) currents versus HV.
1,00
10,00
100,00
1000,00
2,2 2,3 2,4 2,5 2,6 2,7 2,8 2,9 3,0
High Voltage, kV
Cu
rren
t, n
A
I(Cs137)+ DC=12nA I(Cs137) without DC dark current
Current from one double gap vs HV with Gamma source
HV nom = 2.7 kV
Scan with gamma source
Chamber 1. Gaps 3 and 4. HV=2.75kV.
Dark and 137Cs (1mCi) currents versus number of pads.
240
260
280
300
320
340
360
380
400
0,0 5,0 10,0 15,0 20,0 25,0Number of pad
Cu
rren
t, n
A
left side, DC=12nA center, DC=12nA right side, DC=12nA
middle
left
right
Scan with gamma source
Chamber 1. Dark and 137Cs (1mCi) currents versus number of pads. HV=2.6kV
45
55
65
75
85
95
105
0,0 5,0 10,0 15,0 20,0 25,0
Number of pad
Cu
rre
nt,
nA
Gap 1, DC=3nA, bottom Gap 2, DC=3nA
Gap 3, DC=3nA Gap 4, DC=3nA, top
Separate pads
Scan with gamma source
Chamber 1. Dark and 137Cs (1mCi) currents versus number of pads. HV=2.6kV
45
55
65
75
85
95
105
0,0 5,0 10,0 15,0 20,0 25,0
Number of pad
Cu
rre
nt,
nA
Gap 1, DC=3nA, bottom Gap 2, DC=3nA
Gap 3, DC=3nA Gap 4, DC=3nA, top
Separate pads
Scan with gamma source
Chamber 2. Gaps 3 and 4. HV=2.75kV.
Dark and 137Cs (1mCi) currents versus number of pads.
240
290
340
390
440
490
0,0 5,0 10,0 15,0 20,0 25,0
Number of pad
Cu
rre
nt,
nA
left side, DC=12nA center, DC=12nA right side, DC=12nA
Some conclusions
All assembling technology has been tested.
The required production rate would be possible after some additional actions:
◊ Tuning of the wire winding and wire tension machines
◊ Fixing the final gluing procedure
◊ Multiplying some tooling :
* 6 sets of comb fixtures ( in addition to 6 available)
* 5 sets of vices for O- ring assembling ( one is available)
* more HV supplier lines etc.
This tooling can be provided continuously during Feb-March
The full production rate of one chamber per day can be planned for
the end of April.