BV_FinalReport

Report
Final Report
Viren Bhanot
Work Done
•
•
•
•
•
•
•
•
Rotary compressor – Its working and calculations
Small Experiment Line
Bypass Heater Calibration
Heat Balance Report
Accumulator Dry-Out testing
Accumulator Cooling Power measurements
CMS Pixel Upgrade
User’s Manual for Building 158
Rotary Compressor
Sanyo Compressor
• Model no.: C-C140L5
• Dedicated for CO2
• Two-Stage
compression
• Pressure rating of 90
bar (outlet, 2nd stage)
Matlab Compressor Calculations
Small Experiment Line
• 250W Cartridge Heater
• Swagelok Fittings
• Concentric Heat Exchanger after Mass Flow
Meter
• Actuators for Metering Valves
Heater Selection
•
•
•
•
250 W heater required
Cartridge, insertion heater for direct heating
Inlet to heater perpendicular to its length
Flow development analysed for selection
▫ Hydrodynamic and Thermodynamic flow
development
• Low watt density required
Heater Selection
Correlations Used:
• Hausen
• Stephan
• Shah/London
• Flow Development
• Fully Developed
• Thermodynamically
developing,
Hydrodynamically
developed
• Simultaneously
developing
Heater Selection
Specifications:
• Watlow Firerod Cartridge Heater
• Ф -3/8”, Length – 7”
• Power - 250 W
• Watt Density – 5 W/cm2
• 4” No-heat zone
• Epoxy Seals to protect from moisture
Swagelok Fittings
•
•
•
•
Heater mounted on reducer union
Inlet through Welding Tee fitting
Tube inner dia = 1/2”, heater dia = 3/8”
Line between mass flow meter and metering
valve in a concentric internal heat exchanger.
• Whole Assembly Welded Together
CAD Model
Heater Assembly
Actuator
• Electrical Actuator for
Swagelok metering
valves
Two companies
discovered:
• Hanbay inc. (Canada)
• Grotec (Germany)
• Actuator ordered from
Hanbay (cost ~ $1500)
Bypass Heater Calibration
• Bypass Heater (2 kW) performance not
satisfactory
▫ Virtually no heating at low powers, then sudden
overheating at medium-high powers
▫ Output power not equal to power requested
through PVSS
• Heater controlled through phase angle controller
(inexpensive way of controlling heater)
• “Span” setting of heater too narrow. Corrected.
Bypass Heater Calibration
• Tests performed to measure heater power and
compare it against input power (through PVSS)
• System run in single phase to measure enthalpy
using Pressure and Temperature.
• Power o/p = Enthalpy Change x Massflow Rate
• Pressure, temperature measured across internal
heat exchanger.
• Output power found to be not equal to input
power
Bypass Heater Calibration
Bypass Heater Calibration
• Phase Angle Controller chops up the sine-wave
signal (4-20 mA) linear with time, instead of
linear with output power
• Mustapha prepared MATLAB and PVSS
programs to correctly calculate output power to
match input power.
• New logic incorporated into PVSS by Lukasz.
Works perfectly, and has been tested.
Heat Balance
• Heater Calibration tests
expanded to give
overview of heat balance
for entire system.
• Heat addition/extraction
measured to get an idea
of system performance
• Compressor data also
included
Parameters measured
• Preq
• Qpump
• Qheater
• Qin
• Qcond
• Qcomp
Heat Balance
Heat Balance
Heat Balance
Conclusions drawn:
• Up until 1100W, the readings are reliable.
• Readings above 1100W are unreliable due to premature
boiling of CO2 inside the tubes.
• Offset between requested power and power measured is due
to heat added by the pump and surroundings.
• Compressor cooling capacity is lower than expected (this data
is already 2 months old)
Accumulator Dry-Out Testing
• Accumulator heater in
thermo-syphon
configuration
• During start-up,
accumulator heated for
long time
• At higher vapor
pressure, higher vapor
density.
• At higher density, lower
convective currents
• Risk of dry-out, heater
melting.
Accumulator Dry-Out Testing
• A parameter called Thermal Resistance used to
calculate dry-out thresholds
ℎ − _
ℎ =
 
• Accumulator heated for long periods with 250,
500, 750, 900 and 1000W power
• Rth calculated for all data points, and plotted
against accumulator saturation temperature.
• Combined graph for all readings plotted
Accumulator Dry-Out Testing
Accumulator Dry-Out Testing
S. No.
Heater Power
Dry-Out
Saturation
Temperature
(°C)
1
250
N/A
2
500
27
3
750
22
4
900
19
5
1000
17
Conclusions drawn:
• At 250W Heater Power, dryout is not witnessed.
• Higher the heater power,
lower the saturation
temperature at which dry-out
occurs.
• Some unexplained bumps are
observed at higher powers,
through sudden, steep rises
and falls in the values of
Thermal Resistance
Accumulator Cooling Power
• Tests done to measure cooling power in
accumulator
• It was expected that cooling at 100% valve
opening should match heating at 100% heater
power (1 kW)
• This was not the case
• Some cooling power lost because cooling spiral
passes through liquid CO2.
Accumulator Cooling Power
• Accumulator cooled and then heated at specific
rates. For example, 50% CV1105 valve opening
corresponds to 50% heater power, 500W
• Slopes of cooling and heating measured.
• Cooling slope observed to be less than heating
slope.
• Since slope unequal, this method not enough to
determine cooling power
Accumulator Cooling Power
Accumulator Cooling Power
• Rough estimate obtained by plotting Heater’s
power versus the value ‘dp/dt’ (change in
pressure per unit time)
• dp/dt values of cooling spiral superimposed on
heating graph.
• This gives rough but useful estimate of cooling
capacity.
Accumulator Cooling Power
Cooling Power Measured:
Valve Opening
Cooling Power
25%
60
50%
225
75%
575
100%
740
Accumulator Cooling Power
Conclusions:
• Cooling power does not correspond to its respective
heating power.
• The maximum cooling power available is only 740W.
• The cooling power at 25% valve opening is not a quarter
of the full cooling power. This is due to the inertia of the
fluid.
• Cooling Power is not completely linear over the entire
range.
CMS Pixel Upgrade
• CMS Pixel layout being discussed with various
iterations proposed.
• Latest proposal (at that time) was simulated to
measure the fluid temperature and pressure drop.
Results were compared with Bart’s results (with his
global calculator)
• Joao’s calculator was used in Matlab, and
simulations with Friedel and Chisholm correlations.
• The simulation results were similar to, but not
exactly the same as Bart’s own simulations.
CMS Pixel Upgrade
CMS Pixel Upgrade
User’s Manual
• The eventual aim of project was to prepare User
Manual to allow external researchers (Belle,
SLAC, IBL) to use the system without distracting
Bart, Lukasz or Joao!
• The manual is about 60 pages long and will
hopefully be used by someone in the future.
Thank You

similar documents