|
|
Michigan B184 Update April 29, 2004 |
Back to B184 updates * Previous * Next
Here is the weekly report from B184, prepared by Ed as Zhengguo is in
the States this week.
EMS4C16 - Top ML failed leak test (2.9 ml/day, failure if > 2
ml/day). Epoxy applied, undergoing leak test.
EMS5A02 - Has had all endplug ends painted with epoxy, allowed to
cure, and leak tested. The leak rate (with He) has changed from (ML
top, ML bot) = (11,14) to (14,10) mb/day. These numbers are raw,
without any He/Ar correction factor, so the Ar rates are obviously
smaller by a factor of 2 or 3. At any rate it is clear that the
painting has had no obvious effect, and the leak rate is perhaps 5-10X
ATLAS. It has been decided to move on and finish integrating this
chamber.
Note that the pressure drop data looks fairly ragged even with
temperature compensation due in part to the unstable temperature
situation in building 184. Therefore, the errors on the leak rate
are probably high unless you measure for several days. We could get
a large blanket and throw it over the chamber while testing to
stabilize the temperature.
One strange observation on this chamber is that bubbles have formed
in the epoxy painted on to the endplugs at some locations. However,
there are contradictory inclusive observations on this so it is hard
to reach a firm conclusion about what is going on. Initially there
was a small amount of residual pressure in the chamber (<1psi).
After the first bubbles were seen, this pressure was discovered and
released. However, bubbles continued for form. Epoxy was applied
in some locations by hand, and in other locations with the epoxy
pump. Neither method was observed to form bubbles in other
applications. However, endplugs both hand and machine applied
locations have some bubbles. Some bubbles formed a 2nd time after a
2nd epoxy coating, others not. The location being painted is the
junction between endplug and FC. Possibly air could come from
behind the FC and make bubbles (though why the air would flow like
this is not clear). So it is hard to come up with an explanation
which explains all observations.
EML3A07 - Just opened, needs epoxy. Survey mounts are installed but
lack targets. Called Jim Bensinger on this and he said he'll send
out some targets (if he has any left, which was uncertain) and we'll
have to attach and measure their positions with respect to the edge
of the platform (250 micron precision required). There are probably
other EML3 chambers with this problem (likely EML3A01, 03, and 05).
EML3A15 - Epoxy applied. Installing HH and T sensor cabling.
EML3C07 - Passed long term leak test; Hedgehogs and T sensor cabling
being installed.
EML4A01 - Epoxied at Michigan. Upper ML failed LT leak test with 2.6
mb/day. The upper ML was found to have many small holes in the
epoxy applied to the half jumpers - 84 holes in all in both RO and
HV sides. The lower ML had a few and was leak-tight. The upper ML
was filled with He, and the Q196 found leaks at a number of these
holes. The chamber was depressurized and the holes repainted with
epoxy, rechecked with Q196 and found to have no leaks. Then the
both ML were recertified at about 1.5 mb/day.
I am a bit confused about what has happened to this chamber. One
might conclude that the upper ML had many half-jumper leaks, and
then was painted with epoxy while under pressure, causing holes to
form in the epoxy. Both of these features seem rather strange,
however. I thought gas bars were leak certified before applying
epoxy. The epoxy is not to fix leaks, but to provide a backup in
case of leaks. Also, if a leaky gasbar was painted, why was it
painted under pressure?
EML4A07 - This chamber was found to have a broken wire (ML1 layer 2,
tube 38, tube number 239824). Resistance between wire and ground
was measured and found to be Signal End = 203 Ohm; HV end = 0 Ohm.
The total wire resistance is about 205 Ohm so these measurements are
consistent with a wire broken near the HV end and touching the tube.
The wire crimp pins did not show any wire slippage, i.e. there was
short length of wire visible at the end of the crimp pin. The wire
was intact at Michigan.
This chamber also has a pinhole-leak tube (ML which had been previously
disconnected from the tubelets.
This chamber is now undergoing electronics testing. One anomalously
"hot" tube has been found and is being debugged. Mezz card has been
swapped to no avail. The tube has been treated to -HV for 1/2 hour
which did not fix it. Now it is getting a long term treatment of
+HV.
General Comments:
o GAS TEST. We had an external company take some gas samples for
examination. They took samples directly from the mixer, and from the
exit after flowing through a chamber. Here are the results:
Mixer Chamber
====================== ================
CO2 6.84 +- 2% 6.92 % +- 2%
O2 3.5 ppm +- 10% 6.02 ppm +- 5%
H2O 3 ppm +- 40% 25 ppm +- 10%
THC <0.3 ppm 0.7 ppm +- 35%
(THC = total hydrocarbon). The errors above are relative error, not
absolute. So the CO2 error about 0.15% in absolute. We asked them
to test for silicon. They are looking for a testing company which
can do this.
The gas bottles which feed the mixer are have purity ratings of:
Purity O2 H20
Ar 99.996% < 5 ppm < 10 ppm
CO2 99.9% < 100 ppm < 100 ppm
So the measured impurities are consistant with the input gas
Rachel looked up the effect of H20 on the drift time.
The effects is +0.0645 ns /ppm, negligible for our gas. We are
looking for some O2 info.
Anyway, I conclude that the gas mixture is nearly correct with no
substantial contamiants. I don't know enough about the mixer to know
if we can tweak it to exactly 7%, but certainly it is close enough for
chamber integration.
o SURVEY TARGETS. We are now recording the offsets of the survey
targets from nominal positions. Unfortunately many targets are
fractions of a mm off, and some even 1 or 2 mm. The accuracy required
is 250 microns. Apparently, the gluing/positioning technique was not
adequate. I suspect the platforms may have shifted before the glue
dried (even though we used 5 minute epoxy).
o B184 DELIVERIES. When getting parts delivered to B184 make sure to
specify ATLAS Michigan group. Otherwise, you may never see your
package, as I have found. B184 is large with many nooks and crannies
and there is no telling where you package might end up.
o B SENSOR CABLES. I have finally gotten all the parts for B sensor
cable making. I have found remarkably poor documentation on making
these cables. I have basically come up with a way that should work,
but am seeking confirmation from Nikhef engineers (who designed the
system).
o TUBELET PAINTING. We need to make a consistent policy about which
tubelets need to have epoxy applied, since there has been confusion on
this issue. After discussing this with Bing, we decided that all HEIM
hard brass need to be painted. However, we used a variety of other
tubelets which do not need to be painted. Here is the (probable)
situation for each chamber series (but each chamber must be
double-checked for tubelet type since not everything will be uniform).
EMS5 - Have HEIM hard brass. Need to be painted.
EMS4 - Early EMS4 have ATM stainless and US "dirty brass" tubelets
which do not need painting. At some point we switched to HEIM. For
example the EMS4C16 chamber currently open has HEIM.
EML3 - Have HEIM hard brass tubelets. These chambers seem to have had
their half jumpers painted at Michigan. They need tubelet painting.
EML4 - Some early chamber have HEIM hard brass which was painted at
Michigan so no painting is needed in 184. At some point we switched
to HEIM soft brass which doe not need painting.
EML5 - Should not need painting.
Note: Half jumpers ALWAYS need to be painted.
---------------------------------------------------------------------
NEWS FROM Muon Week.
---------------------------------------------------------------------
o BUILDING 180 - SECTOR BUILDING SPACE. The toroid magnet people say
the welding is under control and that they have finished 2 (of 8)
toroids. However, they will occupy all of the main space of 180 until
at least Jan 2005. The other end of 180 will have some space made
available earlier, but it is not clear that it is enough space to set
up sector building in this location. Apparently part of building 185
will be available in the fall which could also be used, but it too
small for vertical sector building (door too low), though it could be
used for the horizontal table for the initial mechanical construction
of sectors.
However, there is also the issue of how and when the sector tooling
will be made (things like the horizontal table and vertical stand).
The plan now is to have this tooling made in Pakistan where apparently
the price is good, but the delivery schedule may be tight.
Anyway, there are still many uncertainties on the availability of space
and tooling for sector building.
Also TGC wheels will be made in parallel in the same space and with
similiar tooling as MDTs. The TGC groups are certifying their
chambers at CERN and have 3% done so far.
o BARREL CHAMBERS. The commissioning of barrel chambers is being
slowed up because of the lack of RPC electronics. The RPC
construction itself seems reasonably on track with 1/2 the chambers
built and the remainder done by the end of the year. However, the big
holdup is the lack of electronics which won't arrive until 2005.
There are a handful of sets for H8 and the barrel CR test stands. TTC
(Technical coordination) would like to advance the barrel installation
schedule to try to make up time for the toroid magnet delays by
installing some BOF and BOG chambers starting in Oct 2004. These
chambers would have to have final RPC electronics installed in the
pit.
o ANOTHER NOTE ON GAS. There was another "aging" test done at X5 this
year on a test chamber. The radiation exposure was 5 years of ATLAS
(including a 5X safety factor). Only 3 tubes showed aging,
i.e. decreased gain, and only for a distance of 10 cm from the gas
inlet. That is, the aging did not affect the whole tube. The tubes
will now be dismantled and examined. Last year the aging test
accidently used a gas system with silicone grease. In that case all
tubes were affected for 40 cm of their length from the gas inlet.
Examination revealed silicon oxide "hairs" on the wire in the aged
region of the tube.
It has been suggested to flow gas in the Big Wheel from different
sides for L and S chambers so if there is aging it won't occur
simultenously in the overlap region of L and S chambers.
Edward Diehl