BRIL BHM Detector Commissioning (draft)

This page summarises the status of the activities, ongoing and planned, for the commssioning of the BHM detector.

Installation Plan, cabling, naming

Units positions are numbered from the top down, according to the hole number in the structure (1 to 28, skipping 2 usually). Phi is just a guess based on 120/9~=13 degrees, it must be corrected by looking at the drawing

PLUS NEAR

Position	Approx phi	Unit	Signal cable	Signal cable old label	HV cable	Fiber
1	87	18	1	24	1	E11
4	74	1	2	21	2	E9
7	61	2	3	17	3	E10
10	48	8	4	20	4	E7
13	35	3	5	19	5	E4
16	22	6	6	18	6	E5
19	9	7	7	16	7	E8
22	356	26	8	15	8	E6
25	343	41	9	13	9	E12
28	330	23	10	14	10	C1
Spare			11	22
Spare			12	23

Note: Fiber bundle No. 2, Source C3 (NEAR 1), return C2 (NEARAR 2)

PLUS FAR

Position	Approx phi	Unit	Signal cable	Signal cable old label	HV cable	Fiber
1	93	4	1	2	1	C2
4	106	5	2	3	2	E4
7	119	19	3	7	3	E12
10	132	21	4	6	4	E6
13	145	33	5	5	5	E7
16	158	37	6	8	6	E10
19	171	16	7	12	7	E11
22	184	13	8	10	8	E8
25	197	11	9	11	9	E5
28	210	27	10	9	10	E9
Spare			11	1
Spare			12	4

Note: Fiber bundle No. 3, Source C1 (FAR 1), return C3 (FAR 2)

MINUS NEAR

Position	Approx phi	Unit	Signal cable	Signal cable old label	HV cable	Fiber
1	87	17	1	11	1	E5
4	74	25	2	12	2	E10
7	61	36	3	10	3	E6
10	48	34	4	9	4	E8
13	35	30	5	8	5	E12
16	22	15	6	7	6	C3
19	9	9	7	4	7	E7
22	356	14	8	3	8	E4
25	343	10	9	2	9	E11
28	330	44	10	1	10	E9
Spare				5
Spare				6

Note: Fiber bundle No. 4, Source C2 (NEAR 1), return C1 (NEAR 2)

MINUS FAR

Position	Approx phi	Unit	Signal cable	Signal cable old label	HV cable	Fiber
1	93	22	1	12	1	E9
4	106	20	2	11	2	E11
7	119	39	3	10	3	C1
10	132	42	4	9	4	E10
13	145	43	5	8	5	E8
16	158	32	6	7	6	E7
19	171	12	7	6	7	E12
22	184	28	8	5	8	E5
25	197	45	9	3	9	E4
28	210	24	10	1	10	E6
Spare			11	2
Spare			12	4

Note: Fiber bundle No. 1, Source C2 (FAR 2), return C3 (FAR 1)

Friday 13/2/2015

1)Structure

A quarter of the structure has been assembled and installed by Mimmo + 2 people from central team. 8 units have been installed.

2)Units

20 units are at Point5, all set painted (see data from Kelly below for characterization). We will install for the moment only 18 units till we make sure that we have 40 perfectly coupled units.

3) Cabling

HV :

- HV multicore cable: Last HV connector is at the moment connectorized by Svetlana in S1.

- SHV patch cables for S1: 40* 1m cables for PP in S1 to CAEN are stored in S1 cupboard.

- PP in UXC Z+ will be installed by noon. (currently installed in S1 cupboard).

- PP in UXC Z- is already installed.

- The two PP in USC were not installed in the final position in S1F07 because of PLT cables preventing it. I hope somebody will take care of it.

- 1 out of the 2 A1535SN CAEN modules. the second one is in the lab. the LV module is also in the lab.

- SHV cables for PP to detector are being brought to point5 from Meyrin.

Signal:

- All signal cables are connectorized and have pigtail in the external braid in both ends to allow different grounding schemes.

- BNC Patch cables of 30m (for UXC) are stored in S1 cupboard.

- N to BNC adapters (50) are stored in S1 cupboard.

Fibers:

- Nicolo fixed the connectors in the fiber of Z-.

- Z+ is laid.

- Octopi have been characterized by Kelly.

4) Electronics

-uTCA crate, MCH, AMC13, 1*uHTR installed.

-A rack mounted PC was installed underneath the uTCA crate to allow follow the HCAL rule temporarily. It will act like the ControlHubServer but will not be controlled centrally. The final solution will be decided after the uTCA integration workshop in Cruzet.

-GLIB from 904 will have to be moved to P5.

-QIE card, ngCCM are available. Fiber from QIE to uTCA will be given to us from HCAL (Tote).

-Once everything is configured in the uTCA crate (AMC13, MCH, GLIB) we will have also the backplane.

For the connection of the signal cables to QIE

Plan A (final) : Patch panel with N, splitter to 3 lemos.., 2*Winchester cables cut in half and reconnectorized to LEMO (ongoing)

Plan B: N to MMCX adapters (2 available, waiting 7 more in a few days), MMCX to WMPO pigtail borrowed by HCAL (Tote), 1* Winchester cable.

Detector Unit Data

PMT #	Tube #	Characterization at 1800V (B3E5)	Best low Bias	Deduced gain(10^5)	Best high Bias	Deduced gain(10^6)
BA3270	44	94.2	1640	5.8	2100	~5
BA3285	25	190.8	1480	~6	1900	4.95
BA3261	27	257.8	1570	~6	2000	4.8
BA3294	18	110.3	1620	6	2050	~5
BA3263	6	273.1	1500	4.64	1970	~5
BA3283	4	67.1	1650	5.95	2100	~5
BA3281	26	261.5	1550	6.81	1950	5.08
BA3265	2	168.7	1520	~5	1980	~5
BA3280	37	128.3	1550	6.27	1970	~5
BA3297	33	156.4	1520	~6	1890	~5
BA3287	8	88.4	1600	6.39	2040	~5
BA3272	1	136.2	1520	~6	1940	~5
BA3277	23	142.3	1600	6.01	2040	~5)
BA3298	7	345.8	1490	~6	1920	~5
BA3309	11	262.5	1480	~6	1850	4.4
BA3299	41	329.6	1440	~6	1830	~5
PMT #	Tube #	Characterization at 1800V (B2C2)	Best low Bias	deduced gain(10^5)	Best high Bias	deduced gain(10^6)
BA3251	16	200.3	1580	6.8	2000	4.81
BA3259	12	377.3	1550	5.92	1980	~5
BA3266	13	367.4	1530	~6	1950	5.03
BA3278	3	471.5	1470	~6	1870	~5
PMT #	Tube #	Characterization at 1800V (B1E11)	Best low Bias	deduced gain(10^5)	Best high Bias	deduced gain(10^6)
BA3260	5	118.6	1560	~6	1990	~5
BA3289	19	67.3	1650	6.14	2100	~5
BA3262	36	125.7	1590	6	2020	~5
BA3307	17	82.4	1500	6.11	1910	5
BA3310	9	242.8	1440	~6	1840	~5
BA3282	21	148.2	1480	~6	1890	~5
BA3269	28	132.2	1590	~6	2020	~5
BA3288	24	140.4	1550	6.08	1980	~5
BA3230	42	215.2	1410	~6	1790	~5
BA3273	34	129.6	1540	~6	1970	~5
BA3284	10	155.9	1540	~6	1990	~5
BA3295	15	191.9	1520	~6	1900	~5
BA3293	14	155.1	1550	4.89	2050	~5
BA3312	30	152.7	1520	~6	1950	~5
BA3253	45	102.0	1650	~6	2100	~5
BA3291	39	67.8	1510	~6	1930	~5
BA3271	32	119.2	1550	6.28	1970	~5
BA3144	20	40.3	1620	no caracterization	2020	no caracterization
BA3292	22	152.4	1540	~6	1950	4.92
BA3311	43	152.7	1450	6.09	1850	4.96

Prototype Tests at P5

Prototypes are measured with the full calibration system setup. The gain of each prototype is porportional to the ratio of the charge of the respective channel with the reference channel

B Field (kA)	Raw Ch1	Raw Ch2	Raw Ch3	Raw Ch4	P1	P2	P3	Relative P1	Relative P2	Relative P3	Notes
0 (0)	864.4	70.23	32.31	6.47	133.6	10.85	4.99	100%	100%	100%
1 (4.5)	864.1	69.56	32.16	6.55	131.9	10.61	4.91	98.7%	97.7%	98.4%	No change within uncertainty

PMT characterisation tests

BHM uses 40 R2059 Hamamatsu PMTs. The PMTs have been delivered to Cern and are undergoing characterization. The characterization involves measuring the dark count rate and the gain over the relevant bias voltage range. The quantum efficiency is also measured.

PMT Data Summary

A summary of the most important data for each PMT

Dark Count Rate

The dark count has been measured using a setup based on VME discriminators and scalers. Measurements have been taken with different discrimination thresholds, increasing the bias voltage in steps of 50 from 1400 to 2000. A point at 2500 has been measured to compare with the reference value from Hamamatsu.

Measurement Completed for 20

Data storage (temporary)

Gain Measurement

The gain measurement has been performed with a setup similar to the one that will be used for the calibration system. The relative light output from each leg of the octopus has been measured prior to the test and will be used to normalize the results.

Measurement Completed for 20

Raw files are very large, will be processed before uploading

Quantum Efficiency

Measurement Completed for 20

To Be uploaded

Mechanics

Part	Status	Extra info	Production Drawing
PMT	40 ready	Actual length 169.3 - 171.7	Length is not uniform within the same PMT (0.5 mm)
Silicon Disk	40 ready plus spares		-
Quartz bar	40 ready	Actual length to be measured	-
PMT Socket	40 ready	Actual length 82.2 - 82.4 mm	-
Inner spacers	pending decision
Front cap (fiber holder)	pending decision
Inner mu metal	40 available	To be cut to correct length
Intermediate spacers	pending decision
Intermediate mu metal	pending decision
Outer spacers	pending decision
Steel tube	pending decision
Front Endcap	pending decision
Rear Endcap	pending decision

Preparation of the detector units

- 20 R2059 Hamamatsu PMTs (first batch, characterized), 20 optical disks of DC93-500 1mm thick, 20 quartz bars of synthetically fused quartz were assembled in the POLYMER LABORATORY (Building 101/1/1014, Sebastien Clement) by Eric Albert. Twenty assembled units are now in the 596 lab. At least half of the assembly will need to be redone.

- Next step is to find a way to blackpaint the front face of the quartz bar except for 1 mm hole for the light from the fiber at the center. A first thought is to use something to stamp a trace on top of the quartz bar but based on the variance of the diameters of the tube there will need to be something that can accomodate the different diameters and allow to center correctly the 1 mm hole.

Calibration System

• The four 12-fiber bundles are available, three need to be characterized.
• The long Y cables will have to be connectorized (waiting for parts) and characterized.
• Waiting for the final mechanics for how to mount the female connector on the detector unit.

Patch Panels and cabling

Cabling by Mimmo

HV Patch Panel USC and UXC

• Z+: Two Patch panels need to be soldered by Pierpaolo (ready in Stella's office), two HV cables have to be connectorized (second week of Dec).
• Progress at Point5:
• UXC: Two cables are installed at -Z with patch panel.
• USC: The two cables have been crimped and connectorized at S1F07. On Monday (20/10) Svetlana will take care of the connectorization at UXC and the mapping of the cables since they lack of labelling.
• For the mapping of the cables the reflectometer is not necessary. A multimeter can do the job. In any case, a reflectometer is available for any measurements at SX5 - hcal region (contact person: Alexander Kaminskyi).
• The HV cable is a multicore cable with 12 twisted pairs. The radiall connector has 23 pins that will be used as following: 1: channel 1 HV, 2: channel 1 return, 3: channel 2 HV,....., 22: channel 11 return, 23: GND. The braid around the twisted pairs will be connected to the chassis of the connector with 6mm lugs.
• When all this work is finalized with (Monday?Tuesday?) a document will be uploaded with all the info for the channels at -Z.

HVPpp.png:

Signal Patch Panel USC

• Pending decision on signal coupling/splitting

Signal Patch Cable in UXC

• Ordered 5 for this test. BNC-male to N type female. Thick cable (not the best). Connectors will need to be crimped. A crimping tool is available.

Channel naming and mapping

Signal

HV

Calibration

Naming of Y cables

• BHMCAL Z- NEAR 1 (source)
• BHMCAL Z- NEAR 2 (feedback)
• BHMCAL Z- FAR 1 (feedback)
• BHMCAL Z- FAR 2 (source)

• BHMCAL Z+ NEAR 1 (source)
• BHMCAL Z+ NEAR 2 (feedback)
• BHMCAL Z+ FAR 1 (source)
• BHMCAL Z+ FAR 2 (feedback)

Electronics

Rack layout in P5

Front End

• The test stand in 904 is operational with a new QIE card prototyep card (SN2).
  • The previous card, SN3, is broken, will be shipped to Fermilab for investigation.
• Status and instructions on how to operate the slow control are here.
• It is likely, but not confirmed, that one out of the four boards produced in the US will be sent to Cern
• The production in Brazil of the QIE10 cards has not yet started. Production of the QIE engineering run should be ready by the 15th October, then a few days to test them and assemble on prototype cards in Fermilab
• The backplane of the frontend crate needs to be ordered (Gilvan contact). There is documentation available in cms db for the HF production backplane.
• ngCCM Baseboard (S. Goadhouse):
  • a number of (minor) changes have been identified as necessary before moving to preproduction.
  • Some more changes have to be discussed after measurements (ongoing at FNAL)
  • 5 preprod boards will be assembled with different (pin compatible) parts due to shortages. One such board will have non radhard parts-good for us
• Mezzanine (T. Grassi):
  • layout is essentially complete.
  • Expected Assembly completion by end of Oct. 2014 (tbc)

Pulser

• Need to decide on which board we want or make our own.

Signal splitting

• The input signal for the QIE cannot be split or attenuated unless it's also AC coupled.
• A Nim module can be used to do splitting, which has an AC coupled output: LeCroy 428A

Back End

While more mature than the Front End, the Back End still needs some work. The main issues are that:

• the uHTR is low on available resources in terms of embedded RAM (especially on the Back FPGA), which prevents us from implementing all the histograms that BHM requires.
• the Ipbus protocol, that is used to readout the histograms was designed for slow control operations, and it lacks the requirements of an efficient readout protocol (too many roundtrips mean large latency).

Tasks that need some work, in orderd of importance:

• Modify the link receiver to work with the asynchronous link and the new data format. This is being done by J. Mans.
• The histogram module needs to be rewritten to increase speed and memory usage efficiency. I have alreay proof of concept code.
• An expert of IPbus could try to figure out if there is a more efficient way of reading out histograms. If this is done the read latency could be reduced and therefore the buffering requirements could be relaxed, freeing more RAM for active histograms. This is a nontrivial task that would require some cooperation with the software side of IPbus.
• Code needs to be written to interpret the raw data and produce a binary answer to fill the histogram bin. This would require a (simple) algorithm that checks the signal amplitude against some threshold and looks at the pulse position in the BX. This algorithm should be written in a well "insulated" way to support evolution into a more sophisticated form in the future. This task is suitable for people somewhat less experienced with verilog.
• Provided we save enough memory with all the above, a few histograms for a combination of channels could be implemented (all-OR, beam1-OR, etc...) This is quite trivial.
• For the future: we had been asked if we could provide a technical trigger. The physical interconnection still has to be decided, and some code will also be required. Also trivial.

uTCA Crate

• One crate is available: MicroTCAstandardmodules
• A second crate will be ordered to serve as development platform and spare.

PVSS

BHM_PVSS_March2015.xlsx: BHM_PVSS_March2015.xlsx

CMSSW

Roberval is helping us with the digitization of the hits. Need to follow up on this.

DAQ and database

A first version of the software has been released.

SVN

Online Monitoring

Magnetic field measurement UXC

Mimmo is investigating with Benoit Cure on installing a 3D magnetic field probe on the rotating shieding. To be followed up; Cabling and readout plans for this ...

Magnet test at MNP 22

The steel tube of 80 mm inner diameter, 100 mm outer diameter, 35 cm will be used for the tests in MNP22. (SCEM 39.20.05.220.7)

The available mu metals for the intermediate layer are all of 1 mm thickness. On hand at the moment the available lengths: 35.0 cm, 21.0 cm, endcap with no hole.

The available hamamatsu permalloy pieces are all 0.8 mm thick and length: 16.4 cm, 19.4 cm, 12.0 cm, 27.0 cm with endcap and drilled hole, 37.0 cm with endcap and drilled hole, 37.0 cm with endcap w/o hole.

The magnet test plan will be announced via elog entries and will be kept up to date as much as possible. The magnetic probe is experiencing some calibration issues so from now on we will proceed only with light measurements (there are new mechanics that allow secure measurements with the LED always positioned stably.

Should there be any recommendations on the magnet test plan please contact stella.orfanelli@cernNOSPAMPLEASE.ch or add directly a configuration in the priority list of the measurement plan at elog

Status on 10/10/2014:

• Elog

A picture of the most succesfull measurement up to now is attached below:

• BHM_PVSS_March2015.xlsx: BHM_PVSS_March2015.xlsx