Proposal for the Barrel Alignment Upgrade

Proposal for the Barrel Alignment Upgrade

Bencze György, Béni Noémi, Szillási Zoltán, Ujvári Balázs

(ATOMKI, RMKI )

This project is supported by Hungarian Research Fund (OTKA) NK 81447 .

the Hardware Barrel Alignment Group

Outline

Two bigger topics:

• Reinforcement of measurements in Sector 13 & 14

• Measurement of inter-wheel distances within Barrel

Two minor items:

• Additional PG targets/redesign PG targets on MABs

• Programming extension to BoardPCs

Barrel Alignment – Upgrade proposal 22011. 03. 29.

And a wish…

• Incorporate DT temperatures into the Ali reco

Reinforcement of alignment measurements in Sector 13 &

14


Schematic view of the Barrel Alignment

- Rigid structures (MABs, z-bars)- Video-camera boxes (on the MABs)- LED holders (called forks, on the chambers)- Diagonal and Z-LED holders

(on the MABs and Z-bars)- Board computers (one for each MAB)

muon chambers

light sources (LED)

CCD’s

MABPassive plane (6)

Active plane (6)

Muon chamberoptical line(8 lines)

Link line

MAB (36 pieces)

Z bar

Diagonal connection


Sectors 13-14 for MB4 chambers

Sector 13

Sector 14

Sector 4

Sector 10

Sectors 4 and 10 are measured at the –phi side only (by 4 MABs)

Chambers in sectors 13 and 14 are measured at the +phi side only (by 2 MABs)

X


Results – what is expected

What is measured (sect 13,14):

MAB

DT

Transversal view Longitudinal view

Measurement point

MAB

x

y

zYB+2 YB-2

Fair results in X and Z at MB+/-1,2, weak in rot_Z, therefore also in Y (depending on the chamber orientation), very weak in general for YB0.

What is expected:

MAB


Results – COCOA-error study

Mean and Max errors (COCOA-errors only but real measurement data):

Mean Max

In X (all chambers, micrometer) 26 61

In Z (YB0 excluded, micrometer) 23 34

In rot_Z (YB0 excluded, mrad) 1.9 2.0

Other rotations (all chambers): <0.3 mrad

For YB0:

In X: 61 micrometer; In Z: 2770 micrometer(!) In rot_Z: 13 mrad (!)

As expected but here we have the numbers


The proposal

Put a MiniMAB with 2 cameras at the joint region looking inside the wheel.The cameras can observe the LEDs (already installed but not used so far).The MiniMABs can fix the DTs wrt. each other.

YB+2 YB+1 YB0 YB-1 YB-2

Sector 4Proposed MiniMABarrangement:

MAB

MiniMAB

Sector 13

Sector 10

Sector 14


MiniMABs

Cameras identical to those used for chimney-chamber regions

d

Camera distance „d”:Sector 13 Sector 14 (top) (bottom)

368 mm 457.4 mm

Two types:

top and bottom

Carbon-fiber plate (5mm thickness)


Results of simulations, sectors 4 and 13

Position error for YB0_sect13:

w/out MM with MMsmicrometer micrometer

phi 27.4 20.4R 1423.8 166.2Z 1511.2 90.1

rot_Z error:

Sect 13 Sect 13 Sect 4 Sect 4Wheel w/out MM with MMs w/out MM with MMs

mrad mrad mrad mrad2 0.13 0.046 0.11 0.0461 0.11 0.058 0.14 0.0580 0.48 0.056 0.48 0.055-1 0.11 0.046 0.14 0.046-2 0.13 0.046 0.11 0.046

Worstcase


Results of simulations, sectors 10 and 14

Position error for YB0_sect14:

w/out MM with MMsmicrometer micrometer

phi 27.4 7.5R 1423.8 135.2Z 1511.2 70.1

rot_Z error:

Sect 14 Sect 14 Sect 10 Sect 10Wheel w/out MM with MMs w/out MM with MMs

mrad mrad mrad mrad2 0.13 0.046 0.11 0.0461 0.11 0.046 0.14 0.0460 0.48 0.045 0.48 0.045-1 0.11 0.046 0.14 0.046-2 0.13 0.046 0.11 0.046

Worstcase


What is needed18 carbon-fiber plates to be bought (cheap)36 video-sensors available from spares36 lenses from spares (f=20mm)36 camera-houses to be fabricated 4 board-PCs from spares* 2 CAEN A3006 LV modules to be bought (~12kCHF) 2 power distrib.+eth.switch parts (switch, connectors, box) to be bought, assembly needed

In addition:Camera-box assembly, calibrationMiniMAB calibrationCabling, installation

Installation is possible during the next long shut-down when the barrel is opened.Free spaces for LV modules and 1U distrib. modules are OK


* New BPC are to be installed only on YB±1. Otherwise use free ports of existing ones.

Measurement of inter-wheel distances within Barrel


Motivation

• measurement of the distance between the barrel wheels is desireable during closure and magnet ramp

• more sensor could also show wheel deformation (some hints from Barel Ali)

Easiest solution is to put distance meters on the wheels and read them out through the BoardPC. (since it would require the least cabling work)


Possible solutions

Potentiometer:

Sticks out -- vulnerable

Laser displacement sensor:

• quite expensive

• needs digital

• Voltage should be compatible with MAB, BPC


Possible solutions

• device used by EndCap Ali is an analog one (and no more produced)

• digital one needed (from different firm)(possible some serial comm, since PBC still has one free RS232 port)

• candidate: Micro Epsilon (RS422)

• firm has been contacted for technical & pricing details --- had an on-site meeting


≥ 1300 CHF/pc !!!

Where to install MAB volume (or close to it) is a good candidate

• only local cabling needed

• wheel ref holes are close and could be used

• Ref holes have very good surface and• used by Survey to define wheel possibility to follow wheel movements in survey approach


Conclusion


• enormous number of devices would be needed for accurate wheel position/deformation measurements

Min 50kCHF (or rather above)

• since Ali doesn’t have wheels (DTs only), these measurements would not improve

• these devices are not good for closing either…

We propose to abandon this project

To tell the truth, we plan to be involved in a similar closing sensor, but:

• less precise (~1-2 mm)• much less number• not radhard, not B tolerant• cheaper (HOPE!!!)

Additional PG targets/redesign PG targets on MABs


Motivation

• there is a discrepancy between the track and the hw alignments both in Z (2mm, scaling?) and phi-like torsion (3mm)

• so far we have only indirect ways to test where it comes from

• previous item wouldn’t solve this enigma, since Barrel Ali doesn’t have wheels.

• torsion has gone, but was not solved by us! We need a direct external measurement possibility.

This is a minor item, but having large impact…


What to do?It would be nice to measure all 6 dof of a MAB, but it is impossible measure at least one point of it:

These points are surely seen from outside

?


What to do with them? How?

• Survey group could measure Z coordinates of them (or at least Z distance)

• Antje says precision is around 70 microns (discrepancy, we have ~ 2mm!)

• Maybe they can measure phi position as well answer for torsion…

• If Z scaling found, its value can be used as a correction later.

• Solution is cheap

• Mimmo can design (probably the worn MAB definition targets are also to be changed at the same time)

• New targets are needed to be located inside the MABs frame needs access to them and probably to be done in ISR.


Programming extension to BoardPCs


Motivation and the solution

• In last May we lost a Custom Board since its software got frozen• Problem is known, it is due to a bootloader that is designed to be able to upload new firmware if needed• Now, on 30 (out of 36) Custom Boards, bootloader is disabled (May problem happened on one of the remaining 6)• This prevents us from uploading new firmware…

We would like to design and install a small device on BPC to solve it

Solution:

This item is very small, but also requires access to MABs…


And finally a wish…


Use TDT in Ali reco


DT Calibration in ISR…

• DT temperature varied between 19 and 24 degC

• We have this data for all DTs

Use TDT in Ali reco


Alignment measurements in UXC…

• Temperature on MABs varies between 19 and 26 degC

• This is the T of air and not DT

(αAl = 23.1 μm/K …might be 100 μm/K on MB4!!)

Tmeas ≠ Tcalib distance of alignment passages change APE gets worse

Use TDT in Ali reco


Latest APE calculations σXlocal ~ 500μm (wrt. PG)

We belive that APE could be reduced if actual chamber temperature could be used.

• This would require to measure rather the real DT temperature than the air (MAB).

• Measure the gas temperature? At the manifold on the chamber exit?

• Aachen has a proposal for that.

(αAl = 23.1 μm/K …might be 100 μm/K on MB4)

Summary


We have four items and would like to get the consent of the DT:

Institutes could be involved: • ATOMKI• Budapest• Univ of Debrecen• CERN ?• Torino (Mimmo)• Aachen (TDT)

Probably all of these modifications could be made in 2013.

Proposal for the Barrel Alignment Upgrade

Documents

z error

micrometer2661in z yb0

ali recokalibrcis eljrsok

zled holderson

mabs led holders

zbars videocamera boxes

results cocoaerror studymean

fair results