Silicon rotation


1 - How to rotate?

ps

In the figure above, A is the assembly we start with. From the top to the bottom the Silicon (with the strip direction indicated), the Kapton cable, the MCM and the output cable are drawn. Note that the width of an assembly is 53mm, and the height of the (inner) silicon is 52mm.

In B is the same assembly, rotated so that the strips are now oriented horizontally.

In C, the Kapton cable is folded over the top of the Silicon to the left. The location of the fold is indicated by the arrows in B and C. We are looking at the back of the Kapton, MCM and output cable.

In D, the cable is folded at 45° such that the cable points down again, and is again face up. The assembly is now 5 cm shorter than in A.

2 - Which Silicon in r-phi?

The only Silicon under consideration are the 'Inner-Middle' detectors in the MVD, because:
ps

3 - Where in z?

Mostly in the middle, to match the e-arm acceptance. The number is limited by availability of components (inner silicon and inner-top cables), and unoccupied inner-middle cage positions, see this message from John, and messages referenced therein. The upshot is that we can make 2x5 such assemblies (out of a total of 2x12 positions) before we have to start discarding already-made units. The maximum is ~2x7 assemblies.

4 - Technical hurdles

4.1) The hard part is to fold the kapton cable close to the silicon. The layout is sketched below. The dimension in z of the rotated silicon is 1 mm less than the normal units. Therefore, in order to stay out of the neighboring silicon's space, the first 180° fold has to occur within 1 mm from the edge of the silicon (as indicated by the dashed line). Kapton and silicon are aligned on specialized mounting fixtures, and are transferred between positions by means of vacuum pickup fixtures. Therefore the kapton has to be flat and open when it is glued to the silicon, and the silicon has to be unobstructed when it is mounted onto the rohacell support. Therefore the cable must be folded after mounting. Forces needed for folding are large, and a few mm of foam does not lend a lot of support to the brittle silicon.

We have glued up some silicon with kapton, and are looking at the folding and mounting procedures.



4.2) Is the cable long enough? To first order, the length of cable 'used up' by the folding process is a square of 52 mm. Recall that we start with an 'inner-top' detector assembly, (4) in the figure below, and want to do the folding and then place the silicon in the 'inner-middle' position, (3) in the figure. However, there are at least 2 additional length penalties: the first is the gentle bend that the IT cable takes between (3) and (2), and the fact that the IT MCM position is the rightmost one in the plenum, whereas the IM sits 6mm further to the left.

In Jan's 3-D model the sum of these effects is that the IT cable is 2 mm too short if the MCM is in the IM slot. However, if the IT position is not occupied, then there would be no problem. The modeling also showed that in the location where the two folds meet, a notch needs to be cut out of the cable's edge, as can be seen in the renderings.
This is what happens if you carefully try to fold the cable close to the silicon. The silicon does not survive this, and also delaminates from the kapton.

Toshi Shiina took the problem as a challenge, and with his intimate knowledge of our silicon assembly systems, many of which he helped design and build, and his superb craftsmanship, has come up with specialized jigs and tools with which we can, in principle, so far, do the job.

This jig has 1/2mm-wide jaws to hold the kapton close to the silicon. The silicon itself is protected in a well behind the jaws. This allows one to fold the kapton 90° within 1 mm of the silicon edge, without putting any stress on the silicon itself.

This is a test assembly where the kapton was folded using this jig. Note the silicon is unbroken.

This shows the mounting fixture, with heavy blocks holding a C-cage. Various vacuum pickup fixtures need to be modified to handle the rotated silicon, with the folded cable (90°) attached.








Last update 4 Sep 2002 - HvH
back