March
14, 2001
Tom
reported on his trip to Italy. He said
that he and Sandro added a lot of detail to the tray and tower assembly
schedule and built in a 4.5 month reserve between when towers are finished and
when they are needed for I&T. In
this new schedule, the carbon-fiber panels and the electronics are the pacing
items. Tom also worked with Bari on the
planning for environmental testing.
Bari will write up the test procedures for the PDR. Tom noted that the current plan is to do
acceptance testing of the towers in the U.S., as the Italian institutes do not
have thermal-vacuum chambers and shake tables big enough for complete
towers. There was also a lot of
discussion of assembly methods and still some outstanding issues. Sandro has an idea on removable clips to
possibly replace the corner-pin and space concept of Hytec.
It
was decided that Erik, Steve, Tom, and Robert should travel to Italy the week
following Easter.
Sandro
reported that he visited Plyform. They
have mounted tungsten on the first tray, with good results. The planarity was better than 100
microns. It will be better after
bonding of the bias circuit. They
thermal cycled the tray from 20C to 50C with minimal deformation (less than 20
or 30 microns).
Sandro
also visited the factory for ladder assembly.
They have produced 3 tools. The
alignment results are on the web. The
maximum error is 10 microns, with a planarity of 20 microns. They have started to produce dummy
ladders. Soon the production will also
start at the second vendor (Mipot, near Trieste).
John
reported that he tried out his glue application machine and found that it
worked well, giving much better uniformity of application that was achieved
manually. He will ship it to Italy for
them to try out.
Sandro
has talked to Plyform about testing of tray panels before mounting of the
tungsten. They are interested in the TV
holography system of Hytec. They would
like to try it out on a panel at Hytec.
Ossie
asked about the grounding of the tungsten foils. Sandro noted that it will be buried in glue, and to make an
explicit, reliable ground connection would be difficult. Robert suggested that we study if a
connection really is needed in this case.
Tom will look into it.
Erik
reported that he released the closeout and insert drawings, which are on the
web. He sent them out for quotes. Steve’s document on tower tolerances was
also release on the web: #1020600007.
Steve
reported on his tolerance study. The
critical tolerances are as follows:
1.
the
profile of the closeout frame, which determines the height and tilt of the
tower. It has a 25 micron tolerance.
2.
The
location of the corner-post holes, for which a systematic error could produce a
winding of the tower. These have a 12.5
micron projection tolerance and a 25 micron position tolerance.
This
is explained in more detail in an e-mail from Erik, which I have included at
the end of these minutes (see below).
Ossie
noted that the sidewalls generally should act to improve the alignment of the
tower, since they can be machined to greater precision than the stackup of the
tower. He didn’t think that the
corner-post pins would have any significant effect once the walls were bolted
on. Sandro said that for that reason he
would like to have removable corner posts (the clips mentioned above).
Tom
said that he wants to set up a web-based meeting to study drawings on a weekly
basis. He will organize it.
Sandro
expressed concern about aluminum inserts.
He wondered whether the threads would be able to withstand repeated
installation and removal of the screws.
Sandro
also expressed concern about the tight tolerances on the Hytec closeout
design. Erik replied that a vendor in
Phoenix is testing machining of the Alcomp material and so far is comfortable
with the specified machining tolerances.
He doesn’t think he would be able to do it with the SGL material
however.
Sandro
said that SGL could machine the closeouts as well as provide the material, but
they could only achieve 50 micron tolerances.
He noted that the time for the engineering model is too tight for Pisa
to arrange for material delivery, machining, and coating at separate
vendors. He particularly wants to avoid
any post-machining of the closeouts (e.g. the corner-post holes).
Tom
expressed concern about the carbon-fiber material availability. The panels are in the critical path, so a 3
or 4 week delay in material deliver would be a disaster.
Gwelen
reported on the meeting at UCSC with Q-Flex, the vendor who can do the readout
cables in single pieces (no splicing).
The meeting was encouraging. We
need to layout one cable for prototyping.
He noted that we have decided to put the thermistors on the cable rather
than MCM, in order to save money on the parts while preserving
interchangeability of the MCMs.
Gwelen
also reported that the MCM is almost done.
He is still working out some interface issues between the GTFE64 chips
and the detectors.
Action
Items:
1.
John:
ship the glue-application machine to Pisa.
2.
Tom:
check on the requirements for grounding of the tungsten foils.
3.
Tom:
organize a web-based engineering meeting.
Here
are a few notes from the March 16 meeting between Robert, Tom, and Erik:
1.
Engineering
model CC material needs to be ordered immediately after PDR. We should plan for a 4-month lead time,
since that is consistent with what we have been getting.
2.
For
flight-material we should get Alcomp on contract November 1, 2001, with an
April delivery specified (the stuff actually will be needed in June).
3.
Tom
and Erik will talk to Alcomp about the production volume, schedule, and cost.
4.
It
was decided that for the upcoming tower prototype we will develop removable
clips in parallel while proceeding with the existing Hytec design. Having holes in the corner posts will not
cause any problem with trying out the removable clips.
Comments
from Erik on tolerances:
There are three primary tolerances that we
are concerned with to ensure the tower alignment is met. These tolerances are the profile tolerance
of the tray, which defines the relative height difference from left to
right. The tray height can vary
considerably, however a slight difference on one side will cause a rotation of
a single tray, which will be amplified by the tray stackup. Machining tolerance of 0.0005" have
been specified, however if we cannot achieve them with machining tolerances,
there are alternatives to simply achieve the tower alignment.
The second tolerance is the positional
tolerance of the corner post holes. This tolerance is not extremely critical
because variability in the hole position only causes slight misalignment, and
will not disrupt the tower alignment.
However, these tolerances need to be somewhat tight to ensure that the
trays are easily stacked without causing damage in the holes themselves. A positional tolerance of 0.001" has
been specified.
The third tolerance is the projection
tolerance. The projection tolerance is
the tolerance that controls the relative position of the corner-post holes on
the top of the trays relative to the bottom of the trays. The corner-post holes can be off by
0.010" if the projection from top to bottom is very near zero, and this is
repeated from tray-to-tray. For this
reason, the projection tolerance becomes critical and must be less than
0.0005". The projection tolerance is not easily corrected, therefore we
must test this machining to ensure it can be achieved. If this dimension is missed on the trays, an
entire tray may be lost. To correct the
problem, we would need to enlarge the holes and bond in inserts to position the
holes. This is more costly and less
desireable, unless absolutely required.
The tight tolerances of the closeout wall
machining include only the profile tolerance described above. Additional tolerances that must be tight are
the alignment features necessary for assembling the trays. All of these tolerances can be achieved when
machining aluminum (for example).
However, this carbon fiber reinforced carbon will be more difficult
because fibers and voids will cause the machine tools to drift during
machining. For this reason, we have
sent a sample of this C-C to our vendor in Phoenix. He has been playing with the material and believes that these
tolerances can be achieved. He has
ranked the quality of this material to be in the top 5% of all C-C materials
that he has worked with. Also, I should
note that they are an SGL representative and have worked with SGL materials
quite a bit.
He does not believe that he would be able to
achieve these tolerances with SGL materials.
We are hoping to have his final assessment this week.