COLD WP08 alignment

Monday Feb 9, 2026, 9:00 AM → 7:00 PM Europe/Paris

COLD WP08 
Alignment meeting

DMartin
C.Gonzalez Torres
T.Brochard

SLiuzzo

 

Summary from DMartin

 

Without proper fiducialisation, alignment becomes an exercise in uncertainty. When we lack visibility into how sensitive internal elements relate to external references, we are no longer measuring—we are speculating. It is like asking for “the length of a piece of string”: a deliberately vague question that exposes the absurdity of acting on partial knowledge. The result is that required micron-level precision collapses into centimetre-scale guesswork.

This highlights why fiducialisation is so critical for high-precision alignment. In this case, three key elements need to be clearly identified and characterised.

First, we must define the sensitive element. Here, this is the beam axis. A central challenge is finding a practical way to represent this inherently unobservable quantity in a form that can be measured and reproduced.

Second, this representation of the beam axis must be tied to an observable feature within the cryomodule that is rigidly and unambiguously linked to it. This internal reference provides the essential connection between the sensitive element and measurable coordinates, and its stability with respect to the beam axis is fundamental.

Third, this internal reference must then be related to an external fiducial system, so that the beam axis can be aligned within the wider reference frame.

A further complication is that the position of the sensitive element relative to the external fiducials is expected to change when the cryomodule is evacuated and cooled with liquid nitrogen. We therefore need to monitor the behaviour of the internal reference features, under the assumption that they track the motion of the sensitive element with respect to the external reference system.

This brings me to Frequency Scanning Interferometry (FSI), a commercially available system that is used extensively at CERN. FSI is a laser-based metrology technique for high-precision distance measurement and alignment monitoring, and is widely employed to track the relative positions of large accelerator and detector components with micron-level accuracy, especially under changing environmental and operational conditions. It is therefore well suited to our application, where internal sensitive elements move relative to external references as a result of vacuum and cryogenic operation.

At present, the CERN Survey and Alignment group is developing a low-cost system specifically targeted at this type of application. We therefore propose to collaborate, formally and/or informally, on the development of a system that could be deployed in the COLD project and, more broadly, in the future LINAC.

Such a system would provide dedicated feedthroughs into the cryomodule, linking the internal reference network to the external reference system used for alignment relative to the surrounding accelerator infrastructure.

The next step is to understand what our colleagues at SLAC National Accelerator Laboratory propose for the fiducialisation of the COLD cryomodule. They may already have relevant concepts in development, or be considering approaches along similar lines.

If this is not the case, we should then establish a closer collaboration with our colleagues at CERN and assess the expected cost and scope of such an effort. This will require familiarisation with the detailed design of the FSI system and its associated feedthrough interfaces.

In parallel, we should develop the internal COLD reference frame and define how it can be connected through the available feedthroughs in the existing cryomodule.

Finally, if this approach appears viable, it would be highly beneficial to design and construct a dedicated test system—such as a prototype tank—to validate the proposed concept and demonstrate proof of principle.

 

 

HLS in the linac tunnel in next shutdown

Drawings of installation under definition, waiting for lots of components

SLAC no definiton of feed through, how to connect QCM to rest of the world. ideas but unknow. definition in April, not before. 

start to populate linac and pinj just to define if there is enough space. 

then will precise each components one by one

no idea of fiducialization. if slac does it, else it is mm error.

fiducialization = where beam axis is respect to orher elements

need reference marks stable. 

what would you do. 

speak a lot on how to do proper fiducialization

best would be to use frequency scanning interferometry (reasonable inexpensive 100keuros)

used at ATLAS. to monitor movements. 

the proposal (from CERN) is to link the exterior of the cryomodule to the interior. 

What happens is:
access all the interior of the cryomodule. measure the beam-axis figure out where it is. Than measure with respect to a number of points inside the assembly, the cryomodule. The support hold spheares. ok for short ranges. then intall FSI system with feed through to exterior of the cryomodule. Measure fisical posision of measurement location. link interior of cryomodule to exterior. This way (if tested) 10s of microns. 
and allows to follow deformation in vacuum and in cold. 

needs fiducialization. 

at CERN with long magnets. used a MOLE. moving throught the vacuum chamber or magnet. This way they could follow the beam path. 

need also to follow deformations. 

if 2 structure roughly aligned. 
if see ligth entrance to exit. enough.

if last dipole laser that can align precisely. Align it trough the gun, this will be the beam path. What was shown at CERN CLEAR facility. 

DM: will not do that. The laser is very impressive, but discussed with Helene. lasers are wanderfull, but for alignment they provide problems. Japanese are leaders. made impressive presentation. problem is that laser is a blob. imaging this thing is didficult to get center of mass. Looking in litterature there is no convincing solution. For high precision it does not work.

If enough to align so much that ligth goes trough. 

May not do laser when cold or in vacuum.

ID02 and ID16 aligned trhough vacuum window, problem with refraction. alignment is mm range.

FSI system would be inside the interior of the cryomodule. one part inside, one outside. both are linked such taht we know precisely where they are repsect to each other. 

Looking for a system that would work for COLD and for the 6GeV linac later on. 

FSI system feed though rather large at CERN.

RF, LN2, electrical feed trhough, but for sure there are free windows. 

one feed trough hole on each end could be usefull for FSI.

could design something that fits on SLAC Cryomodule. 

in the tank we have 2 acceelerating structure. May be need 3, to check both accelerating strucutres?
Will take what we have.

Thank exists, certification on it. adding flanges will not be easy or certification fails.

minimum of 2 feed trhoughs. desing a system with internal measurement. inside the cryotank. 

30 keuros for 16-32 feed trouhgs. then controllers for the resolution-distance measurement. Can perhaps if 2 feedtrhoug dedicated to alignment can work.

need detail drawings to install interior newtworks to monitor the movement of the structures. 

can use a stick of metal inside the acc, structure with a centered head? 

a lot of work for CLIC project. tap into this knowledge instead of reinventing things.

BEST FSI. 

Thierry may be knows how to verify strigtheness of a given hole. 

scan interior of the hole and see the passage. shining collimated ligth can see a minimum passage of light. 

25th Februrary, meeting with SLAC.

2 separate problems. 
1) what comes from SLAC. (no holes possible)
2) can we modify the interior of it. for example with FSI system

Establish a working relation with SLAC colleagues to perform the alignment. 

Helen is happy to develop system. one for CERN and one for ESRF. 

Assemble one for both labs and test directly. 

Include alignment topics in EU grant.

Can we disassemble once at ESRF.

2 accelrating section. easy to measure holes at each end. done for vacuum chambers. 

if we could use FSI to measure reflectros, then we can connect feedtrhough toe xterio references.

wait see what they propose. 

speak with CERN about FSI. 

reasonably inexpensive 100-200 k euros. 

need to go to cern assembling and testing there, need to create our own simulated tank with copper inside to make sure things actually work in principle. 

ready in 2027 to install ancillary systems. 

proof of principle tank. 2 tanks, supports, 80k euros. 

think of a prototype, make a design and see if it is possible or not. piece of copper inside simple. 
proof of principle that it works.