Changes considered for production tubes.
Remove resistors between the final dynode stage and the anodes,
R18 - R33.
This is to be compatible with QIE/SMQIE.
Charge-up problem?
SMQIE has to determine the ground level and thus the anode level.
Cause a difference from the HV ground?
Add a 300-Ohm resistor between HV shield and anode shield?
Anode pin output instead of cables.
Like one in
H7546 (64 ch), but with individual ground
pins.
H8711 drawings :
1 -
2 (HV is also a pin),
3 -
4 (HV coax cable)
Use the tapered voltage divider for a factor of 10 better
pulse linearity (dynamic range).
The ratios will be 1.5 : 1.5 : 1.5 : 1 : 1 ... 1 : 2 : 3.6.
This should ease the space-charge effects at the last synode stages,
though the overall gain will be reduced by roughly a factor of 5 (at -800 V).
Design proposal : Sept 2002
H8711A-10MOD drawings :
Electrical -
Mechanical
Oct 17, 2002. Order placed for ten H8711A-10MOD tubes.
To be shipped from Hamamatsu to Tsukuba on Dec 20, 2002.
Final drawings:
Mechanical -
Electrical
Delivered on Dec 24, 2002.
Tube data by Hamamatsu:
Data sheets -
Anode uniformity -
Mechanical drawing -
Electrical drawing
Pictures: (including the interface PC board
prototype)
Real size of the anode pixels
(Aug 30, 2002)
PMT box design as of
Nov 2002 (from Bob Miller, MSU)
And a matching HV connector candidate
11 SHV-50-3-1 on the PMT side.
Nov 14, 2002 :
A further modification of the pins is being asked
so that we can attach TSW-type connectors
(e.g. TSW-108-07-?-D)
on the PMT.
This is under consideration by Hamamatsu.
Feb 14, 2003 :
H8711A-10MOD (21B-030A) drawings :
pdf file
(both mechanical and electrical, though no change to electrical)