iTOP readout firmwarefirmware development
Not shown: Joshua Sopher (firmware)Lili Zhang (DSP coding)Lili Zhang (DSP coding)
PNNL: David AsnerDavid Asner, Mitchell Myjak, Scott Morris, L W d
1K. Nishimura and G. Varner 25-MAR-2011 l-DAQ meetingLynn Wood
Overview• Status of various components• Status of various components
• Immediate deadline cosmic test in JapanE ti l t i t CERN b t t• Essential gate prior to CERN beam test
T d ’ f• Today’s focus:• ASICs: BLAB3A or IRS2+amps
P th 128 h l d t d l• Progress on the 128-channel readout module
Fi d l t i• Firmware development issues
2
iTOP Readout Overview64 DAQ fiber
transceivers
8k channels
32 FINESSE8 COPPER
1k 8-channel waveform ASICs
64 SRM64 SRM
Precision timing requires 64 channels high-precision clock distribution (<~ 10ps)
3
distribution (<~ 10ps) Approximately 30m runs
Major milestone:1/16 system test
Third generation waveform
sampling ASIC p g
Clock jitter cleaners
4
Current incarnation of Subdetector Readout Module (iTOP)Subdetector Readout Module (iTOP)
5
A very crowded location!8k vs. 14k (CDC channels) << 10% of space!
6
BLAB3 Specifications32768 samples/chan (>5us trig latency)32768 samples/chan (>5us trig latency)
8 channels/BLAB3 ASIC8 Trigger channels9 bits resolution (12[10] bits logging)~9 bits resolution (12[10]-bits logging)
64 samples convert window (~16ns)4 GSa/s1 word (RAM) chan, sample readout
1+n*0.02 us to read n samples (of same 64)30 kHz sustained readout (multibuffer)( )
• Time alignment critical– Synchronize sampling to accelerator RF clockSy c o e sa p g to acce e ato c oc
– >5us a must for trigger, since single photon rates high
• Needs Gain!
BLAB3/IRS (amp/no-amp)( p p)
5.82mm
7 62mm7.62mm
8x RF inputs(die upside down)
32k storage cells per channel(512 f 64)(512 groups of 64)
8
BLAB3/IRS Single Channel 9
• Sampling: 128 (2x 64 separate transfer lanesp
Recording in one set 64, transferring other (“ping-pong”)
• Storage: 64 x 512 (512 = 8 * 64)
Wilkinson (32x2): • Wilkinson (32x2): 64 conv/channel
Sampling speed
Sampling Simulation with full parasitic Extraction
4.000
4.500
5.000
Extracted
2.000
2.500
3.000
3.500
Sam
plin
g R
ate
[GSa
/s]
0.000
0.500
1.000
1.500
0 0.5 1 1.5 2 2.5
RCObias [V]
10
IRS2 DC Linearity Calibration
ARA Digitizer - 12-MAR-2011 11
IRS2 Noise Measurement
<1mV<1mV
12
Measurement via RF sine
Analog BW~1GHz
BLAB3A testing (carrier board)
23mm x 50mm
Plan to submit soon BLAB3A
14
Should work mechanically, if can really fit components…
15
mechanical mockup
16
SCROD block diagram
17
SCROD Fabricated
Rest of board stack needed: Firmware!!
18
Beam test: a 1/16 system test
Third generation waveform
sampling ASIC p g
Clock jitter cleaners
19
Beamtest back-end (cPCI_DSP)
For freference
when di idiscussing issues b lbelow
cPCI Interface
20
(PCI bus)
Front-End Board Stack Hardware Status21
Board Name Design* Fab & Assembly
Testing** Comments
SCROD Complete ~Complete Not started 2x SCROD assembled, in-hand.Si l f ti lit t t b i t kSimple functionality tests begin next week.Full testing requires complete board stack (except HV).
InterConnect ~Complete Not started Not started Fabrication waiting on ASIC carrier card design (in case any changes).
ASIC Carrier Not started Not started Not started Design waiting on ASIC daughter cards
BLAB3A In Process Not started Not started Focusing on design now.Daughter
IRS2 Daughter Not started Not started Not started Only necessary if problems w/ BLAB3A.
HV Not started Not started Not started Waiting on someone with time to design.
Personnel: • Design: primarily Louis Ridley
• Supported by Gary Varner, Matt Andrew, Kurtis NishimuraF b & A bl i il M tt A d
*Designs are NOT the same as final Belle II system.
• Fab & Assembly: primarily Matt Andrew• Testing: primarily Kurtis Nishimura
• Supported by Matt Andrew, Gary Varner
**”Testing” here means checking board connectivity / power / etc.
Front-End FirmwareFront End Firmware DevelopmentTask Code Status Testing
StatusComments
General ASIC Control
Partially Complete
Partially Complete
BLAB3A/IRS interface, timing.Currently in debugging/testing using IRS_eval.Will need to be adapted to new board stack.
Feedback loops Rewrite? Restart? Feeds back to maintain sampling rate / ADCFeedback loops Rewrite? Restart? Feeds back to maintain sampling rate / ADC conversion rate, etc.Old version available, but needs rework.Must be adapted to new DACs.Pending front-end design completion
DAC control Not started Not started
Pending front end design completion.
Front-end data link Not started Not started Data links available from other projects / Belle2Link, need to be adapted to board stack.(Trigger not strictly necessary for beam test.)Front-end trigger
linkNot started Not started
Trigger time encoding
Not started Not started (Trigger not strictly necessary for beam test.)
Personnel: P i K ti Ni hi J h S h ( til• Primary: Kurtis Nishimura, Joshua Sopher (until now)
• Support: Gary Varner 22
Back-end / Auxiliary Boards23
• Not many changes to back-end hard arehardware.
• Sufficient quantities already in h dhand. Primarily testing / firmware / DSP / software work remains.
Back-End / Auxiliary Hardware Statusy
Board Name Design Fab & Testing* CommentsAssembly
DSP_cPCI Complete Complete Partially Complete
DSP programming still needs testing (coupled with firmware).
TRIG FIN C l C l C lTRIG_FIN Complete Complete Complete
FIN_cPCI Complete Complete Complete Simple board for powering FINESSEmodules without COPPER.
CLK FIN C l t C l t P ti ll Cl k di t ib ti f lCLK_FIN Complete Complete Partially Complete
Clock distribution successful.FTSW module could fill in here.
*”Testing” here means checking board connectivity /
Personnel: • Testing: Matt Andrew, Kurtis Nishimura
board connectivity / power / etc.
24
Back-End / Auxiliary Firmware/DSP/Software StatusTask Code Status Testing Status Comments
Back-End Data-Link Partially Complete In Process Tested stand-alone, not integrated w/ other firmware.
Back-End Fiber / PCI Bridge
Not Started Not Started Interface between fiber link FPGA and PCI FPGA.g
DSP Interface Partially Complete In Process (Not strictly necessary for beam test.)
DSP Algorithms Partially Complete In Process (Not strictly necessary for beam test.)
PCI Interface (Firmware) Partially Complete In Process Stable version available @ 3 MB/s WorkingPCI Interface (Firmware) Partially Complete In Process Stable version available @ 3 MB/s. Working toward DMA version (60+MB/s).
PCI Kernel Driver Partially Complete In Process
PCI Software ~Not Started ~Not Started Only test software exists to check driver/firmware statusdriver/firmware status.
Trigger Data-Link Complete Complete Fairly mature.
Trigger Combiner / Logic ~Complete ~Complete Fairly mature, may need small changes for beam test.
CAMAC Interface ~Not Started Not Started No code yet… can adapt from other projects.
Personnel: • DSP code: Lili Zhang• DSP interface: Matt Andrew, Serge Negrashov• PCI interface / driver: Xin Gao, Serge Negrashov, Andrew Wong• PCI software : Serge Negrashov, Andrew Wong• Trigger: Xin Gao
25
Necessary Hardware (Beam Test)Hardware Needed In-Hand Comment
Front-end board stack 4 <1 8 SL10 / stackFront end board stack 4 1 8 SL10 / stack1 data fiber / stack1 trig fiber / stack
CLK_FIN 1 2 Or FTSWFIN_cPCI 1 1TRIG FIN 1 2_DSP_cPCI 2 4 1 for data
1 for triggercPCI Crate 1 1cPCI Crate 1 1CAMAC Crate Controller
1(?) 1
CAMAC Crate 1 0 Assuming NagoyaCAMAC Crate 1 0 Assuming Nagoya brings crate & modules. 26
27
Summary/Open issuesM h fi k d d• Much firmware work needed• Help from PNNL; write system documentation
• Hardware – confirm items previous slideC l A 3A i i b d• Complete BLAB3A carrier, routing boards
• Interface board done, submit 3x designs soon
• Confirm performance of integrated module, i l di i h MCP PMTincluding with MCP-PMTs
28• Development manpower resource limited (where PNNL can help)
Back-up slides
29
Photo-detector: Hamamatsu SL-10• Micro channel Plate:• Micro-channel Plate:
– Operates in 1.5T B-field– <50ps single photon timing
• Multi-pixel (4x4 anode pads)• Enhanced Lifetime (Al protection layer)
Interesting mechanical challenges• Interesting mechanical challenges (PMT case at HV)
30Approximately 1” x 1”
BLAB3 status and scheduleSpecification BLAB3 BLAB3A BLAB3B FINAL
A l 175 MH 400MH 500MH ? 400AnalogBandwidth
175 MHz 400MHz 500MHz? 400-600MHz
Gain [50 ref] 34-36x 60x 100x ? SNR>=50
Sampling speed [Giga-sample/s]
3.6 3.8 4.0 4.0
Usable li d
~1.4 3.8 4.0 >= 3sampling speedInternal DACs no no yes yes
Design Sept. Sept. January Autumncompletion 2009 2010 2011 2011Delivery Jan. 2010 Nov. 2010 May 2011 Winter 2011
quantity 120 240* 120? 1000q y
* = not for Belle2, but will learn from design
SL-10 Timing PerformanceNagoya Hawai’iNagoya Hawai i
σ ~ 38.37
N f i di i i CAMAC• Nagoya = constant fraction discriminator + CAMAC ADC/TDC H i’i f li + f t t ti
32
• Hawai’i = waveform sampling + feature extraction
High speed Waveform sampling“oscilloscope on a chip”osc oscope o a c p
• Comparable performance to best
CH1
CFD + HPTDC• MUCH lower power, no
d f h bl CH2need for huge cable plant!
• Using full samples
CH2
Using full samples reduces the impact of noise
6 4 RMS• Photodetector limited 6.4 psRMS
NIM A602 (2009) 43833
Advanced Detector Research award
NIM A602 (2009) 438
Belle2 barrel PID upgrade: iTOP
34
references and further inforeferences and further info
references: references: http://b2comp.kek.jp/~twiki/pub/Organization/B2
TDR/B2TDR pdfTDR/B2TDR.pdf http://www.phys.hawaii.edu/~idlab/taskAndSched
ule/ICBMS.pdfule/ICBMS.pdf latest info:
http://idlab phys hawaii edu/pcb designs/scrod http://idlab.phys.hawaii.edu/pcb-designs/scrod
35
brainstorming SCRODbrainstorming SCROD
36