Stefano Belforte's SVT docs
Still unofficial,
informations in here may change without warning,
use at your onw risk
HOME : back to Stefano Belforte's
home page
documents for, and
wrap up
from, the
SVT Workshop
October 27-28 1998,
also summary
given at the CDF Collaboratin meeting on Nov 5, 1998
AMS : Associative Memory Sequencer
-
AMS PRODUCTION CHIP list of what is needed to
send from Pisa to UniGE and what from Geneve to Pisa.
-
VME addresses from the Geneve web site.
- VME addresses in more compact Stefano's form.
- list of modifications for next version of AMS (bug fixes, our
improvements, CDF
pcb review outcome) from Annie.
- An
AMS board block diagram
showing all the big chips and the main
data path, made with the unix free-domain xfig program is
availble as
gif or encapsulated
postscript, or even as the xfig
source file
-
Clocks, distribution and roboclock settings.
-
The AMS internal microsequencer code is stored in RAM. The RAM content
is created by a C program. Get here the latest verion
of the
source code and of the needed
header file. Also the
state diagram.
- AMS schematics as A3-size gzipped postscript files:
AMS(top level),
Clock_Gen,
Convme_5rows,
Front_Panel,
Glueams_Hard,
In_Bus,
Input_Control,
Micro_Sequencer,
Out_Bus,
Output,
P3,
P3conn,
Prom_Switch,
Vme_Bus,
Capa
- Also the scheatics of how the Xilinx elemental bodies have been
wrapped into *_xepld bodies with nice shape and big busses:
hspy_ct_xepld,
ospy_ct_xepld,
inpreg_xepld,
input_xepld,
usq_oreg_xepld,
usq_sreg_xepld,
out_ctr_xepld,
vme_ctr_xepld
- Xilinx chips programming. All the details
of what is inside the various Xilinx EPLDs.
- AMSglue (aka GLUE3) programming. The
internals of the AMSGLUE chip.
AMB: Associative Memory Board
Modifications to 1998 prototypes planned for production:
Franco's list shown at pcb review and more
extended
Stefano's wish list (includes the previous)
Schematics as
gzipped postscript files (A3 format).
If you have problems with the postscript file, see the note about
my postscript files in the top page.
top level,
vme bus,
Xilinx EPLD that implemets the VME slave
Summary of talk submitted to the
1998 IEEE Nuclear Science
and Medical Imaging conference
(or more precisely to the
1998 IEEE Nuclear Science Symposium)
in
postscript or
pdf format. Also the
abstract, the LaTex
source file and the required
latex style.
The paper status can be viewed clicking
here, paper numer is 1124, but you will need a password (palle1).
HB: Hit Buffer
MRG: Merger
Old from CAEN
AM-FPGA project
AMfpga Board
- OGLUE chip: bottom level output glue, ties 64 AMfpga chips from 8 busses
and talk to TGLUE on top side.
Other stuff
FIFO pitfalls
Note on Postscript files
Stefano Belforte (
stefano.belforte@ts.infn.it)