Present: M. Bosteels, H. Breuker, P. Fontaine, R. Gregory, S. Grohmann, M.
Hatch, G. Hallewell, J. Inigo-Goflin, A. Onella, R. Pintus, B. Pirollet, A.
Placci, W. J. Postema, P. Rosnet, A. Smith, W. Van Sprolant and M. Wilhelmsson.
 

1.      Review of Primary Cooling Installations
 Joaquin Inigo-Goflin summarised the state of the tendering for contracts for
cooling towers and other surface cooling installations. Most contracts had
already been awarded and the air conditioning contract for the underground
caverns was expected to be decided in July. He indicated the present state of
knowledge on the proposed layouts of the cooling equipment in the underground
service caverns. He presented the layout of USC55 dating from 1995 for CMS and
the equivalent drawing of USA15 dating from 1997 for ATLAS. It was evident
that updated information on these installations was badly needed since it is
hoped to go out to tender for them at the end of this year. Although ATLAS in
considering using only fluoroinert coolants, it was not obvious whether the
circulators could/would be installed in the service cavern or the experimental cavern.
Joaquin requested that each experiment furnish him with information on the
number of cooling circuits they expect to need, along with their cooling
power, flow rate, temperature requirements, pressure drops and fluid type.

2. ALICE Status
Ray Gregory said that cooling the coil of the L3 magnet would dominate the
power requirments of ALICE. In addition, the experiment will have a muon arm
dipole magnet which will also require cooling. It was pointed out that ALICE
planned to use only one of the two existing magnet coils as they would be
running at a lower field than L3 and hence they were unlikely to need the
whole 4 MW currently used by the L3 magnet. Ray believed that contacts with
ST-CV were good and ALICE expected to reuse the existing counting room cooling system.
The central tracking detector was currently planning to use water, but the
group was now starting to move towards using fluoroinert liquid. Polypropylene
glycol was being considered as the PHOS coolant which is to be operated at
-25C.  TIS has approved this liquid but it may have too high a viscosity for
use everywhere.

3. ATLAS Status
Greg Hallewell presented the results he had obtained from a survey he had
initiated on cooling needs within ATLAS. He gave a very detailed breakdown of
the Pixel and SCT cooling circuits and the SCT/TRT thermal enclosure. A series
of tests on evaporative cooling for the Pixel and SCT is currently in
progress, and several coolants are being studied. He continued with the TRT
itself, the liquid argon and tile calorimeters which may use underpressure
water or C6F14 as their coolant.  The muon spectrometer is expected to radiate
most of its power of which about 74 KW will have to be removed by the cavern
air conditioning system. Greg finished by providing details of the power
consumption of the trigger chambers and the trigger racks. ATLAS will use the
data Greg is assembling to define the quantities of fluids needed from ST-CV,
and expect to be able to produce their own schematic layouts for heat
exchangers within the experimental cavern.

4. CMS Status
Ron Pintus explained that CMS was in the process of setting up an Integration
Group to examine the layout and installation of equipment and pipework within
the experimental areas. This group should also rationalise the requirements of
the various subdetectors. He added that R. Principe of ST-CV was already
studying the requirements of the ECAL. A data base of CMS heat loads and
associated parameters has been maintained by Ron for some years. It was
initially assumed that the average rack heat load would be 8 KW but it is now
believed that a figure of 6 KW for counting room racks and 4 KW for cavern
racks would be more likely. He presented tables of heat loads expected in the
USC service cavern, the USC counting room and the experimental cavern, UXC. It
is assumed that most detectors will radiate only 5% of their power to the air,
but for the muon detector this will rise to 25%. He presented an overall
summary of the expected CMS heat loads which agreed surprisingly well with a
very early crude estimation. He finished by presenting the rack count for CMS
in which there is still a greater requested number of racks in UXC than space
will allow. It is considered that this is unlikely to be a real problem as
rack sharing by different systems should solve it. Ron thought that CMS would
not be able to give the information requested by ST-CV by the middle of the
year but that such information should be available towards the end of the year.
It was noted that ATLAS had already supplied information on rack cooling
requirements to ST-CV.

5. Results from tests with fluoroinert monophase cooling in an underpresssure
system near room temperature.
Phillipe Rosnet presented details of tests carried out on cooling ATLAS tile
calorimeter modules with C6F14 near room temperature. An underpressure system
was used and initial tests were done with water to provide a reference.
Various circuit boards have to be cooled including HT cards and electronics
near the PMT’s. The tests had initially suffered from a leak in the system
which caused the vacuum pump to be switched on automatically every hour, and
the pump’s position meant that fluid was drained from the system. This was not
a fundamental problem and would not happen with a different configuration of
the cooling system layout. Bubbles had also been detected which were formed at
a pipework T-junction which allowed expansion of the liquid. After the
dimension of the outlet pipe of the T was changed to eliminate possible
expansion, bubble production no longer appeared. Cooling with C6F14 produced a
temperature of the circuit boards about 0.5 C higher than with water cooling.
Increasing the flow rate of the C6F14 to try to further reduce the temperature
resulted in the production of bubbles and it was concluded that there was a
limited range of acceptable flow rates.

6. Materials  in contact with cooling fluids which should undergo irradiation tests
Triggered by a question from Antti about the use of O-rings in connectors,
Alasdair believed there was a need to produce a list of materials which are
expected to be in contact with cooling fluids and which ought to be tested
after irradiation in the presence of the fluid. Antti was looking into the
possible use of quickfit connectors which almost invariably used O-rings. He
had considered mineral, plastic or rubber O-rings and believed that NBR was
the most likely candidate to use with fluorocarbons. It can withstand a few
megaGrey and so should be OK regarding radiation damage.
Alasdair noted that the present list of materials to be irradiated in the
presence of coolants stood at Al, stainless steel, C-fibre epoxy and now NBR.
He asked each experiment to consider whether more materials should be added to
this list and to inform himself or Sorin Ilie of any such material. If
irradition tests of materials such as C-fibre epoxy were needed, the
experiment requesting the teats would have to provide the samples to be
irradiated and be prepared to define and participate in the tests on the
irradiated samples.
A general discussion on connectors ensued and was sufficiently lively to imply
that this should be a topic at a future meeting.

7. Next meeting
The next meeting will focus on connectors and fittings. Several persons have
expressed interest in this and the date of the meeting will be announced after
all speakers have been contacted.

A copy of the transparencies presented at the meeting will be sent to a
representative of each experiment plus ST-CV. Anyone else wishing to have a
copy can request one from A. Smith.

 
Back to Home Page.