CERN Cryomodule
Summary
CERN (Conseil Européen pour la Recherche Nucléaire) is undergoing an high luminosity (Hi-Lumi) expansion of their Large Hadron Collider (LHC). As part of this expansion, they are adding 4 more cryomodules to their collider; TRIUMF being a leader in this space was commissioned to build these cryomodules. (A cryomodule is a linear particle accelerator operating at cryogenic temperatures so that it may be superconductive.)
After CERN designed the cryomodule, they gave TRIUMF some STEP files and unreleased drawings, with a promise that not much would change; since things were subject to change, and the drawings were unreleased and in ISO standards, TRIUMF decided to remodel the parts in SolidWorks (as to have editable features) and then from these, make their own drawing package in ASME standards (so it can be more easily manufactured in North America).
Scope
The scope of my responsibilities was the vacuum vessel - the main structural component of the cryomodule. The goal of the project was to re-model each part of the vacuum vessel in SolidWorks, referencing the incomplete ISO drawings provided in conjunction with the STEP files, then to create a drawing package in ASME standards that falls within the tolerances specified by the ISO drawing package provided by CERN.
Technical Details
The project was broken down into 4 steps:
Model each part
To model each part, the incomplete drawings were used as the starting point, and only deviated from where they contained incomplete information
If information was missing the STEP files provided were referenced in order to match the geometry, and this was recorded in a formal inconsistencies document to be sent to CERN (any points of confusion or questions about the design were also noted here)
Verify each part
To verify each part, the 'compare bodies' tool in SolidWorks was used, deviations were cross-referenced to the drawings; if they were mistakes they were corrected, if they were inconsistencies between the drawings and STEP files, a reasonable assumption about which was correct was made, and the situation was logged in the inconsistencies document
Assemble the parts into assembly
Starting with the base plate, each part was mated into the assembly carefully following the assembly drawings and verifying against the STEP file assembly
Upon mating each new part, the part interfaces were verified at every contact point to be touching, and every fastener point to have concentric holes; this revealed several parts where things didn't line up
At each point of concern, the STEP file assembly was checked for the same issue, if it was present, the issue was logged to be given to CERN, if it was not present, the problem was identified and fixed.
The final assembly was kept separated in two parts: the hermetic string, and the lower assembly. This was done since the entire cryomodule will be assembled into these two parts and then during installation the hermetic string will be lowered into the lower assembly.
Create a drawing package
In order to create a drawing package, ISO drawings had to be interpreted and reproduced in ASME standards; for this, a competency in both standards was required
After reviewing and understanding both standards and the differences between them (including first vs third angle projection and many GD&T differences) drawings were made side by side with those from CERN
After drafts of a few drawings were complete, the decision was made to no longer try to make the drawings similar in appearance to those from CERN (which had been done initially to make side by side comparison easier)
Each drawing was then arranged in the most intuitive way possible (as decided with feedback from the team) and then simplified as much as possible, notably:
To reduce the drawings to a single frame of reference for GD&T where possible to save manufacturing costs (as long as it wouldn't result in any unreasonable tolerances)
And, to adjust all profiles to use profile of a surface tolerances for ease of reading, and consistency
Once a few drawings had undergone successful review, the team decided to wait until a package was ready before reviewing the rest
The package was completed and sent out as a PDF for three internal review cycles before being sent off for a final review from every interested party at TRIUMF
Results
At the end of my time at TRIUMF, the drawing package was under final review before being sent off for manufacturers to bid on.