[PPT] - Timeline for FEMB Development, System Tests and protoDUNE Run 2 PowerPoint Presentation

SLIDE 1

Timeline for FEMB Development, System Tests and protoDUNE Run 2

Marco Verzocchi - Fermilab Cold Electronics Workshop – Brookhaven National Laboratory 17 July 2018

SLIDE 2

Content

Different systems available
40% APA at BNL
protoDUNE APA in CERN cold box
new small TPC at Fermilab
DUNE prototype APA
protoDUNE Run 2
Timeline for FEMB development and goals

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

2

|

SLIDE 3

40% APA @ BNL (i)

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

3

|

SLIDE 4

40% APA @ BNL (ii)

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

4

|

SLIDE 5

40% APA at BNL (iii)

Requires 8 FEMB
Test in LN2 (it actually requires a lot of LN2)
Not a TPC
Always available
Can be used for fast feedback
Will be in our testing plan for at least all of 2019

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

5

SLIDE 6

protoDUNE APA #7 in cold box @ CERN (i)

New protoDUNE APA will be built in the UK with minor

modifications

Use to commission new winder
Use new system for the ground mesh (this will not be the final mesh,

it will be less transparent for photons)

Regular discussions with APA consortium on timeline
We will again have access to the cold box on Nov. 12 (beginning of

SPS shutdown)

Some work needed to put back rail support system (and remove

some of the muon detectors ?)

Not clear we will have photon detectors installed
Plan is to fully characterize the APA in the cold box using

protoDUNE FEMBs, ideally before the CERN Christmas closure

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

6

SLIDE 7

protoDUNE APA #7 in cold box @ CERN (ii)

Requires 20 FEMB
We may prefer to keep 10 protoDUNE FEMBs, and replace the
ther 10 with new ones to allow direct comparison
Test in cold gaseous LN2
Known system, but remote
Availability depends on CERN support (conflicts with continued

protoDUNE cosmics run ?)

Can be used for fast feedback
Will be in our testing plan for at least all of 2019

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

7

SLIDE 8

New small TPC at Fermilab

Discussed by Dave yesterday
Plans is to fully commission in the Fall with protoDUNE FEMBs
Fast / easy turnaround (2-3 weeks) when inserting new

components (FEMBs, photon detectors)

It’s a real TPC (cosmics)
Plan to use it until beginning of production, the cryostat will then

become a user facility for testing new designs in LAr

It’s not an inexpensive object (~$0.5M to build it, $5k of LAr

every time we empty it and fill it again)

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

8

SLIDE 9

DUNE prototype APA at CERN (i)

The APA consortium is planning on building 3 DUNE prototype

APAs in 2019 using new 4”x4” tubing for the frame

First two will be purely mechanical (no wires) and will be used to

check the APA stacking and the cable routing in a test at Ash River

We are interested in this for the cable routine part of the test
Will be further discussed after lunch in Manhong’s presentation
The third APA will be a complete prototype
PD consortium informed only this past week, will need to make sure that

PD cables and connectors will be available before winding starts

Will be available at CERN toward the end of 2019 / beginning of 2020
We will want to use this APA instead of protoDUNE APA for further tests
f cold electronics / new photon detectors
We will need 20 new FEMBs to fully populate it (if possible we will also

do some tests with older FEMBs for reference)

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

9

SLIDE 10

DUNE prototype APA at CERN (ii)

We need to understand whether we are going to have changes

to the CR boards, CR-FEMB connection, FEMB form factor (cannot make it bigger, but no push at this point to make it smaller)

Ideally we want to maintain compatibility (connector) of current

FEMB with FEMB for DUNE

But we (as entire Collaboration) should make an effort and

understand the few channel failures observed in protoDUNE and decide whether this requires a serious investigation of the reliability

f the connections in the APA board stack/ CR boards / FEMBs
If we change connectors / form factors we may need adapter

cards to continue using 40% APA at BNL, small TPC at Fermilab

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

10

SLIDE 11

protoDUNE Run 2 (i)

APA plans
2018/2019 modifications to the APA design, new winders
2019 first set of prototypes
2020 set up factories
2021 begin production
Some APAs could also become available in 2020
Bring 2 or 3 DUNE APA (depends on difference with 1st

prototype) to CERN

Fully populate with final (pre-production) FEMBs
Characterize in cold box
Install in protoDUNE (beam side), take data again in 2021/2022

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

11

SLIDE 12

protoDUNE Run 2 (ii)

TCO closure done in such a way that cryostat can be reopened

and resealed a few times

Second run of protoDUNE will be a stress on the Collaboration

resources

Calibration data with the final detector
Need to present a proposal to CERN SPSC in October
Approaching quickly, have not yet seen real discussion of this in the

DUNE Collaboration

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

12

SLIDE 13

Timeline (i)

May 2018: test FEMBs with commercial ADCs in 40% prototype at

BNL (SBND Collaboration)

Fall 2018: commission new small TPC at Fermilab and 7th protoDUNE

APA at CERN with protoDUNE FEMBs

Question #1: do we (does SBND) want to test SBND FEMBs with the

small TPC at Fermilab (SBND doing tests with LArIAT) and/or with the 7th protoDUNE APA at CERN ? Require 10 FEMBs

My opinion is that this is something we need to do prior to the TDR

submission

CRYO ASIC:
Standalone tests Nov/Dec, mount on FEMBs in January, test FEMBs
Test with 40% prototype at BNL, small TPC at Fermilab, protoDUNE

APA#7 at CERN in February / March

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

13

SLIDE 14

Timeline (ii)

New BNL-FNAL-LBNL ADC
Standalone tests January/February, mount on FEMBs and test in

March (first version of FEMB, still using FPGA, 2nd version of FEMB with COLDATA will be available in May)

Test with 40% prototype at BNL, small TPC at Fermilab, protoDUNE

APA#7 at CERN in April/May (first version of FEMB, using FPGA, not COLDATA)

Test again (2nd version of FEMB, with COLDATA) in June/July
Delays in the ASICs submission mean that will NOT be having all the

system test results prior to the TDR submission

So far no indication that FNAL/LBNF/DUNE management want to

delay TDR submission

Will have additions to TDR for LBNC and DOE reviews
We do NOT have any room for further slippages of schedule

(absolutely need results for DOE review in October 2019)

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

14

SLIDE 15

First generation FEMBs

Maintain compatibility with protoDUNE
Minimize changes to WIB (just FPGA programming if possible,

more changes required for CRYO)

Connection to CR board
Number of power and signal/control cables
No change in functionality of FEMB / interface between WIB and

FEB (except when absolutely necessary, e.g. for CRYO)

Design of FEMBs for new cold ADC (first with FPGA, then with

COLDATA) to start soon

Minimal modifications relative to current design

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

15

SLIDE 16

Timeline (iii)

Second iteration of ASICs
2nd version of CRYO
LArASIC v9 + 2nd version of new cold ADC/COLDATA
Difficult to predict when these will be available, will depend on

the problems with the 1st version (and whether we want to add features / change specifications)

1 year after first version (i.e. roughly Spring 2020) ?
What do we use for the first DUNE prototype APA at CERN ?
Same boards used for first test ? Wait until 2nd generation of ASICs

is available ?

Do they need to be modified because of changes in connector(s) /

form factor

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

16

SLIDE 17

Second generation FEMBs (i)

Give up compatibility with current hardware
Connection to CR board should be revisited (if we understand that this

has been a source of dead channels), if needed use adapters for matching with protoDUNE APA / small TPC at Fermilab / 40% APA at BNL

Change dimensions of FEMB if needed / appropriate (we are allowed

to change only in one direction, if we need to grow in size there we will need to stack two boards)

Finalize dimensions as soon as possible (mechanical interface with APA,

contribute to overall size of object that needs to fit throughTCO)

Revisit cable connections and routing
Try to keep FEMBs for top/bottom APA identical, but if required this is

something we could sacrifice to gain space (route cables differently)

Reduce number of cables as much as possible
2.56 Gbit/s links for 2nd generation of COLDATA
Remove non-necessary signal and power links

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

17

SLIDE 18

Second generation FEMBs (ii)

Robustness
Should we revisit the cable connections ? This has been a source of problems for

protoDUNE (almost all failures of FEMBs)

Functionality
Faster connection (COLDATA)
In 2nd generation of COLDATA add e-fuses to burn address of ASIC / FEMB in

hardware (not using EPROM like for current FEMB, yield too low). Can we do something similar for CRYO ?

Any other addition ?
RTD on the FEMB ? Add 8-bit ADC inside COLDATA ? (this would be a SW monitor of the

temperature, cannot add RTD connected to interlock with increasing the cable plant)

Power consumption monitoring ?
Any other crazy ideas ?
This is the last time we can make changes ….

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

18

SLIDE 19

Certification of discrete components

In addition to the ASICs we should check the lifetime of all the

components on the FEMBs, if this has not already been done

Other discrete components (voltage regulators)
Resistors and capacitors
Is this being done also for the APA boards and the photon detector

components ?

Insist that this kind of tests be standardised for DUNE single phase

detector

Material checks (outgassing)
How many of these checks need to be repeated ?

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

19

SLIDE 20

ASIC Decision

We want to wait for test results from 2nd generation of ASICs to

take the final decision on the ASIC solution to be used in DUNE

Decision will be taken in Summer 2020, not sooner
Unclear whether this is acceptable to LBNC / DOE
LBNC seems to have the strange idea that we could be taking a

decision in early 2019

DOE will certainly require that we have at least one working solution

(that need to be improved upon) at the time of the CD2 review (October 2019)

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

20

SLIDE 21

Timeline (iv)

Engineering reviews prior to ASIC submission(s)

(preproduction) in Summer 2020 for both ASICs and FEMBs

Reviewed simultaneously
Preproduction FEMBs using final components available in early

2021 for final tests with small TPC at Fermilab, DUNE prototype APA

Use results from these tests for production readiness review
If more APAs are available at CERN we should have enough

components to fully populate them and prepare for 2nd run of protoDUNE

17 Jul 18

M. Verzocchi | Timeline for System Tests and protoDUNE Run 2

21