A Simple Representation of the Impact of a Loop Heat Pipe on a Space System Perry Ramsey Andrew Hensley ITT Industries Space Systems Division Fort Wayne, IN TFAWS '04 31 August 2004
Abstract Abstract � As two phase thermal control devices like loop heat pipes become more common, it is necessary to develop a method to simply represent their effects in a system-level model. Thermo-hydraulically simple models are available, but they still are too slow to include in system models that will be used routinely during concept development. This discusses an attempt to create a model that will represent the gross impact of an LHP on an orbiting space vehicle. � The approach taken is to model the LHP as a node connected to the radiator through an array of variable conductors. The value of each conductor is dependent on the source temperature. By varying the shape of the conductance function, a broad range of potential responses is possible. � The thermo-hydraulic model of the loop heat pipe was run at various input powers over its range of viability. The variable conductor model was correlated against the results using the Sinda solver. The variable conductor model was then embedded in a spacecraft system model. � Though the variable conductance model is applicable only over a fairly limited range of operating conditions, it has proven very useful the development of the system and has provided excellent insight into the behavior of the operating space vehicle. TFAWS '04 31 August 2004 2
An Geostationary Instrument Deals With An Geostationary Instrument Deals With Widely Varying Heat Loads Widely Varying Heat Loads � We are developing concepts for a geostationary instrument � GOES experience shows that varying heat loads must be managed • Diurnal • Annual TFAWS '04 31 August 2004 3
One Way to Deal With the Load is a One Way to Deal With the Load is a Loop Heat Pipe Loop Heat Pipe � LHP is a viable design option • Can transport internal loads to external radiator • Can control evaporation temperature � Need some way to model it at system level � Thermo-Hydraulic Model Simulates Two-Phase Physics (Reference 1) � Converted from SinAps to Thermal Desktop/FloCad • Changed variables to reflect our design, and added heated compensation chamber TFAWS '04 31 August 2004 4
Top Level LHP Behavior Inspires Top Level LHP Behavior Inspires Reduced Representation Reduced Representation LHP Behavior Reduced � Liquid evaporates � Load removed � Condensation in � High conductance radiator until all from evaporator input heat load is to radiator rejected Outlet Inlet � Beyond, radiator � Low or zero decoupled from conductance from load evaporator to radiator � At increasing � More radiator is load, full connected condensation takes longer TFAWS '04 31 August 2004 5
LHP Integrated with Radiator in Orbit LHP Integrated with Radiator in Orbit Produces Credible Results Produces Credible Results � Controlled compensation chamber temperature � Ramp up of evaporator 310 temperature 300 • Drop ~ 40 W/K 290 Evaporator T em p eratu re (K ) � Radiator opens as load 280 increases 270 Inlet • Eventually becomes fully 260 open 250 Outlet � Fluid Model used as “True” 240 behavior in future 230 calculations 0 100 200 300 400 500 Input Load (W) TFAWS '04 31 August 2004 6
Model is ‘Expensive’ to Run Model is ‘Expensive’ to Run � A thermohydraulic model, even simplified, is difficult to incorporate • Takes a lot of calculations per time step • Needs small time step to resolve fluid physics – Fluid variations are not particularly important to external thermal performance with slowly varying boundary conditions – Spending a lot of time on things that don’t affect the results much • Transient instabilities can stop the entire analysis run � We need a representation that is • Fast enough for parametric analysis of the system • Robust • Accurate enough that we don’t make big mistakes in architecture TFAWS '04 31 August 2004 7
Try to Capture the Essence of Behavior Try to Capture the Essence of Behavior Without Modeling All the Physics Without Modeling All the Physics � In simplest terms, a controlled LHP looks like a variable resistor • Resistance is dependent on source temperature � We’re most interested in the temperature drop from the evaporator to the radiator � Secondary goal is temperature variation across the radiator TFAWS '04 31 August 2004 8
Model LHP as Array of Variable Model LHP as Array of Variable Conductors Conductors � Break radiator into segments � Connect each to evaporator through a variable conductor � When cold, all are off � As the heat input increases, conductors turn on in sequence based on rising evaporator temperature TFAWS '04 31 August 2004 9
Conductor Behavior Tailored to Mimic Conductor Behavior Tailored to Mimic Reality Reality � Model it in Sinda � Derive values of conductors Inlet Outlet TFAWS '04 31 August 2004 10
In Sinda, this is a SIVA In Sinda, this is a SIVA � A SIVA is similar to more familiar SIV conductor • Interpolation is based only on the first node temperature Evaporator Radiator Value Scaling node node array factor SIVA 101,HUB.10, RAD.101, hp.A101,fac . . . SIVA 125, HUB.10, RAD.125, hp.A125,fac � Data in the array drives the network behavior TFAWS '04 31 August 2004 11
Rising Ramp Allows Smooth Turn On Rising Ramp Allows Smooth Turn On � Each conductor 120% ramps from 0 to Conductor 2 T incr Conductor 3 100% Conductor 4 100% F ra c tio n o f F u ll C o n d u c ta n c e � Some tweaks 80% needed for stability 60% 40% T LHP On 20% G Near On G Base 0% 274.5 275.0 275.5 276.0 276.5 Temperature [K] TFAWS '04 31 August 2004 12
Use Optimizer to Match Simplified to Use Optimizer to Match Simplified to Detailed Model Detailed Model � Run the thermo-hydraulic model for range of loads • 50 to 350 W input in 50 W steps • Varying beta angles • NOT trying to handle anything but normal operation � Fit simplified to detailed model using the Sinda optimizer TFAWS '04 31 August 2004 13
Two Systems Have Comparable Two Systems Have Comparable Performance Performance � Evaporator and radiator segment temperatures � Some differences are apparent, but evaporator response is similar 290 290 280 280 270 270 Temp [K] 260 260 Temp [K] 250 250 240 240 230 230 220 220 50 100 150 200 250 300 350 400 50 100 150 200 250 300 350 400 LHPload [W] LHPload [W] LHP Fluid Model LHP Simplified Representation TFAWS '04 31 August 2004 14
Installed in System Model, Performance Installed in System Model, Performance Met Program Needs Met Program Needs � Able to run system model with fast turnaround � Not driven by simulated LHP performance � User has to be careful to keep the simulation within applicable bounds • Exceeding shutdown limits • Overrunning maximum heat input TFAWS '04 31 August 2004 15
Variation in LHP Heatload Variation in LHP Heatload � Load input to LHP – up to 250 W diurnal variation 400 350 300 LHP Input (W) 250 Beta 0 200 Beta 8.7 Beta 23 150 100 50 0 12 PM 3 PM 6 PM 9 PM 12 AM 3 AM 6 AM 9 AM 12 PM Time of Day TFAWS '04 31 August 2004 16
Stable Temperature Over Large Stable Temperature Over Large Variation in Heat Load Variation in Heat Load � Temperatures at LHP baseplate vary ~ 10°C 25 20 Temperature (°C) 15 beta 0 beta 8.7 beta 23 10 5 0 12 PM 3 PM 6 PM 9 PM 12 AM 3 AM 6 AM 9 AM 12 PM Local Time of Day TFAWS '04 31 August 2004 17
Opportunities for Improvement Opportunities for Improvement � We’re correlating a model to a model • Test data would be helpful � Impact of external load variations require examination • Varying loads due to, beta angle, property variation � Gradient in radiator too high � Are we optimizing on the right thing � Simulation of return line environment TFAWS '04 31 August 2004 18
Simplified Representation Functional, Simplified Representation Functional, Useful for Conceptual Trades Useful for Conceptual Trades � A representation of LHP behavior has been created • Suitable for concept development � Within the limits of its capability, the representation has been useful • Fast enough for parametric analysis of the system • Robust • Provides an accurate enough result to guide architecture development TFAWS '04 31 August 2004 19
References References � 1) Cullimore, B. and J. Baumann, “Steady State and Transient Loop Heat Pipe Modeling”, SAE Paper Number 2000-ICES-105 TFAWS '04 31 August 2004 20
Acknowledgements Acknowledgements � This work performed under contract NAS5-01119, ABI Formulation Phase TFAWS '04 31 August 2004 21
Recommend
More recommend
