Leakage Fraction Research Articles

Model of Gas Flow Through Porous Refractory Applied to an Upper Tundish Nozzle

Argon gas commonly is injected into the liquid metal stream through the porous refractory walls in many metallurgical processes. In this work, a new model has been developed to investigate gas diffusion through heated porous refractory, including the effects of refractory geometry, the thermal expansion of the gas, temperature-dependent gas viscosity, and possible leakage into unsealed joints. A novel one-way-flow pressure boundary condition has been formulated and implemented to prevent unrealistic flow into the refractory. The complete model is validated with both analytical solutions of 1D test problems and observations of a water bubbling experiment. Then, to demonstrate practical application of this general model, argon gas flow is simulated through a double-slitted upper tundish nozzle during continuous steel casting with a slide-gate system. Realistic liquid steel pressure distributions with the bubbling threshold condition are applied on the inner surface. Parametric studies are conducted to investigate the effects of joint gas leakage, refractory conductivity, permeability, and injection pressure on the resulting gas distributions, gas mass flow rates, and leakage fraction. This new model of porous flow can serve as the first step of a comprehensive multiphase model system.

Key continual-leakage resilient broadcast cryptosystem from dual system in broadcast networks

In order to tolerate possible leakage of secret keys, leakage-resilient cryptosystem models a class of attractive leakage output by allowing an adversary to provide any computable leakage function and learning the partial keys or other possible internal states from the output of function. In this work, we present an adaptively secure broadcast encryption resilient to key continual leakage in the standard model. Our scheme provides the tolerance of continual leakage, in which any user can generate multiple private keys per user by periodically updating the key. We use the dual system encryption mechanism to implement the leakage resilience and adaptive security, and intrinsically set an algorithm to refresh a key and produce a same distributed new key. We also give the evaluation of the leakage bound and leakage fraction, and the simulations show that our scheme can tolerate about 71% leakage fraction with 3.34 × 10−52 failure probability in standard 80-bit security level when we adjust the leakage factor to allow the private key to be 100 Kb.

Bubble formation at a central orifice in a gas–solid fluidized bed predicted by three-dimensional two-fluid model simulations

We apply a recently developed two-fluid continuum model (TFM) based on kinetic theory of granular flow (KTGF) in three dimensional cylindrical coordinates, to investigate bubble formation through a single central orifice in a gas–solid fluidized bed. A comprehensive study for Geldart D type particles, revealing the influence of particle diameter, jet injection flow rate, and bed size on bubble characteristics have been investigated. At a given gas injection flow rate, the bubble diameter continuously increases while gas leakage from the bubble to the emulsion phase decreases with time. With increasing particle diameter, leakage fraction increases and hence a smaller bubble diameter is predicted. These results are consistent with DPM simulations, experimental results and approximate bubble formation models reported previously in the literature.

Demonstration of surface electron rejection with interleaved germanium detectors for dark matter searches

The SuperCDMS experiment in the Soudan Underground Laboratory searches for dark matter with a 9-kg array of cryogenic germanium detectors. Symmetric sensors on opposite sides measure both charge and phonons from each particle interaction, providing excellent discrimination between electron and nuclear recoils, and between surface and interior events. Surface event rejection capabilities were tested with two 210Pb sources producing ∼130 beta decays/hr. In ∼800 live hours, no events leaked into the 8–115 keV signal region, giving upper limit leakage fraction 1.7 × 10−5 at 90% C.L., corresponding to < 0.6 surface event background in the future 200-kg SuperCDMS SNOLAB experiment.

Aerothermal Impact of the Interaction Between Hub Leakage and Mainstream Flows in Highly-Loaded High Pressure Turbine Blades

This paper describes experimental and numerical investigations of a highly-loaded rotor blade with leakage (purge) flow injection through an upstream overlapping seal. The effects of both leakage mass flow rates and swirl have been studied to examine their effects on the aerothermal performance. As the leakage mass flow rate was increased, the loss generally increased. The increase in the losses was found to be nonlinear with the three distinct regimes of leakage-mainstream interaction being identified. The varying sensitivity of the losses to the leakage fraction was linked to the effects of the upstream potential field of the blade on a vortical structure originating from the outer part of the seal. This vortical structure affected the interaction between the leakage and mainstream flows as it grew to become the hub passage vortex. Very limited cooling was provided by the leakage flows. The coolant was mainly concentrated close to the suction surface in the front part of the rotor platform and on the blade suction surface in the path of the passage vortex. However, the regions benefiting from cooling were also characterized by higher values of the heat transfer coefficient. As a consequence, the net heat flux reduction was small, and the leakage injection was thus deemed thermally neutral.

Bounded Leakage-Resilient Functional Encryption with Hidden Vector Predicate

Recent research shows that many public-key or identity-based encryption schemes are vulnerable to side-channel attacks on the keys by the interaction of an adversary with a physical device. To tolerate the possible key leakage, leakage-resilient cryptography models a class of leakage output by allowing the adversary to be able to specify a computable leakage function and obtaining the partial keys or other possibly internal states from the output of function. In this article, we propose a leakage-resilient hidden-vector encryption (HVE) scheme that supports the predicate operators such as conjunction, disjunction, comparison, range query and subset query, etc. The proposed scheme is leakage-resilient attribute-hiding secure in the sense that the adversary cannot only obtain the tokens of non-match vectors but also learn amount of key information of the vector that matches the challenge vector. To the best of our knowledge, this is the first HVE that supports token-leakage resilience. We prove the security with a series of computationally indistinguishable games that uses the dual system encryption mechanism. We also analyze and discuss the performance of leakage bound parameters and leakage fraction in the practical security level. Finally, we also give an extensive scheme to achieve the security of both attribute-hiding and payload-hiding, and analyze the performance in larger alphabets.

Analysis of Hold Time Models for Total Flooding Clean Extinguishing Agents

This study documents the experimental results of a research program designed to evaluate the validity of the widely published hold time prediction models found in NFPA 2001, Annex C and ISO 14520-1, Annex E. The models discussed in these standards obtain a measure of the equivalent leakage area, which, when coupled with ‘worst case’ assumptions, can be used to determine the minimum hold time. Three hold time prediction theories are adopted from these standards for validation; a wide descending interface model as implemented in ISO 14520-1 and two sharp descending interface models from the 2004 and 2008 publications of NFPA 2001. The experimental program is comprised of 15 tests conducted in a 103 m3 test enclosure. Three commercially available clean agents are selected to span a wide range of agent vapor densities including FK-5-1-12, HFC-125, and IG-541. A series of holes were drilled through enclosure boundaries at upper and lower elevations which were opened or closed as a means of regulating the amount of leakage area for any given test. Vertical profiles of agent concentration and ambient pressure are used to evaluate the agent concentration distribution, rates of agent draining, and the effective lower leakage fraction. A non-dimensional hold time is used to compare experimental results involving differing agent types and leakage areas. Results show that empirical values of the hold time are up to 50% longer than the theoretical hold time predictions when evaluated as the time to reduce the agent concentration to half its initial value. When evaluated as a 15% drop in concentration the model validity is significantly reduced. Under this condition, empirical hold time values are up to 50% shorter than the predictions of the sharp descending interface models and up to 100% longer than the wide descending interface model.

Incontinentia Lactis: Physiology and Anatomy Conducive to Milk Leakage in Dairy Cows

Incontinentia lactis is a possible predisposing factor for an elevated level of intramammary infection. The goal of the present study was to investigate possible causes of incontinentia lactis in dairy cows. Two farms that differed in breed composition, but that had similar average milk yields were studied: herd A, 28 kg/d, 31 Red Holstein cows; and herd B, 26 kg/d, 16 Brown Swiss cows. Herd A was classified into 2 groups: incontinentia lactis (ILA group) and control, whereas herd B was exclusively a control herd. Milk samples that represented foremilk and the main milk fraction were collected during 4 milking sessions. In addition, milk leakage samples from the ILA group were collected at different time intervals from 0 to 5h before milking. Measurements of the teat, milk flow, fractions of cisternal and alveolar milk, intramammary pressure, and blood oxytocin pattern also were obtained. The ILA cows did not have differences in fat content between milk leakage and cisternal milk fraction. Milk fat content, however, increased during milking in response to continuous milk ejection (1.95, 1.99, and 4.61% for milk leakage, cisternal, and main milk samples, respectively). Teat canals were 9% shorter in the ILA cows, which showed greater milk yield, peak, and average flow rates. Quarter cisternal milk yield of ILA cows tended to be greater (0.50 vs. 0.23 and 0.28kg for ILA and controls from herds A and B, respectively), whereas percentages of cistern milk and alveolar milk did not differ from controls. The greater pressure in the ILA group, both before and after manual udder stimulation (ILA: 4.0 and 6.4 kPa; control: 2.0 and 5.0 kPa, respectively), could be an important cause for the leakage. Nevertheless, the increase in IMP that occurred after udder preparation affirms that milk ejection occurred in response to the tactile teat stimulation, but not before the onset of leakage. Blood oxytocin concentration in ILA cows was low until the start of udder preparation and increased in response to the milking stimulus (reaffirming the hypothesis that milk leakage occurred in the absence of milk ejection). In conclusion, milk losses by leakage are likely due to the large amount of cisternal milk, which creates pressure and causes leakage, in the absence of milk ejection.

Quantitative Imaging of Microvascular Permeability in a Rat Model of Lipopolysaccharide-Induced Sepsis

The aim of the present study was to evaluate the magnitude of endothelial defects by micro-computed tomography (CT) quantitation of contrast agent diffusion across the vascular endothelium in a rat model of vascular permeability caused by lipopolysaccharide (LPS)-induced sepsis. LPS was administered intraperitoneally (i.p.) at a dose of 10 mg/kg body weight in male Wistar rats (n = 18). Vascular leakage and vascular volume were quantified, by micro-CT and cryostatic micro-CT. The contrast agents used were Fenestra ( approximately 70-nm particle diameter), Microfil (large polymer), and iopamidol (Isovue, MW 777 Dalton). Micro-CT revealed an increase in endothelial permeability as indicated by entry of contrast agent (Fenestra) into the extravascular space after LPS administration (P < 0.01). Endotoxin exposure also induced a decrease of vascular luminal volume in the myocardium, liver, kidney, and colonic wall determined by micro-CT (P < 0.01). Vascular leakage, expressed as the ratio of extravascular to intravascular gray scale intensity (IE/II) after injection of contrast agent, increased significantly in the myocardium, liver, kidney, and colonic wall (P < 0.001). The elimination of iopamidol from the intravascular compartment in LPS-challenged rats was decreased compared with control rats. The endothelial defect size was estimated to be >70 nm and <1 microm. Contrast agents are useful to characterize vascular leakage and vascular volume fraction in an animal model of endotoxin priming.

The Interaction of Turbine Inter-Platform Leakage Flow With the Mainstream Flow

Individual nozzle guide vanes (NGV’s) in modern aeroengines are often cast as a single piece with integral hub and casing endwalls. When in operation, there is a leakage flow through the chord-wise interplatform gaps. An investigation into the effect of this leakage flow on turbine performance is presented. Efficiency measurements and NGV exit area traverse data from a low-speed research turbine are reported. Tests show that this leakage flow can have a significant impact on turbine performance, but that below a threshold leakage fraction this penalty does not rise with increasing leakage flow rate. The effect of various seal clearances are also investigated. Results from steady-state simulations using a three-dimensional multiblock Reynolds-averaged Navier-Stokes solver are presented with particular emphasis paid to the physics of the mainstream/leakage interaction and the loss generation.

Effect of Entrainment and Overflow Occurrences on Concentration Profile in PUREX Flow Sheet

A deviation in the operational condition of a mixer settler and a centrifugal contactor causes an entrainment or an overflow, which affects the concentration profile. Although there has been no quantitative study about the effect of such abnormal flows on the concentration profile, the occurrence of such abnormal flows has been severely restricted for a PUREX flow sheet. However, the restriction of abnormal flows can be relaxed when the effect of such flows is limited within the allowable range such that the concentration of the product does not deviate from its specification. This relaxation could serve to benefit a continuous operation under a certain degree of deviation from the operational condition and a smaller design load of a solvent extractor. From this viewpoint, the relationship between the magnitude of abnormal flows and the effect of them on the process was studied quantitatively using a specially developed code in a wide range of PUREX flow sheet conditions, and the possibility of this relaxation was investigated. The results showed that the effect of the abnormal flow on the concentration in the organic outflow or aqueous raffinate was dominated by the leakage fraction under normal conditions regardless of each specific flow sheet condition. The common correlations were found between the leakage fraction of uranium and plutonium under the occurrence of abnormal flows and that under no abnormal flow for the stripping and extracting conditions, respectively. Comparing the given correlations and the usual specification of the leakage fraction of uranium and plutonium suggested that the restriction of the abnormal flows could be relaxed for a usual PUREX flow sheet.

Integral data for most important materials, namely lithium, beryllium, lead and uranium in infinite medium per fusion neutron

The physical behavior of integral data in an infinite medium has been evaluated for incident fusion neutrons with the help of the 3-D Monte Carlo Code. In a fusion reactor blanket with finite dimension, the integral quantities will be more or less different from the infinite medium results, depending on the neutron leakage fraction. Design studies foresee reduction of the neutron leakage out of the blanket to very low levels in order to prevent nuclear heating in superconducting fusion magnets and to keep all neutrons primarily in the coolant. The most important materials in fusion technology, namely tritium, beryllium, lead, and uranium have been investigated in an infinitive medium. The main purpose of this work is to calculate integral neutronic data for incident 14.1-MeV (D,T) fusion source neutrons in an infinite medium for the most important materials in fusion technology, namely lithium, beryllium, lead and uranium. The study concludes the calculation of the integral tritium breeding ratio, 239Pu breeding rate, neutron multiplication ratio through (n, x) and fission reactions and that of the heat release in those mixtures that are composed, first when natural uranium and 238U, are mixed with natural lithium or 6Li for a volume fraction from 0% to 100%, and then, the variable uranium–lithium composition is mixed with Be or Pb for a volume fraction from 0% to 100%.

Decreasing peak flow rate with a new bag-valve-mask device: effects on respiratory mechanics, and gas distribution in a bench model of an unprotected airway

Reducing inspiratory flow rate and peak airway pressure may be important in order to minimise the risk of stomach inflation when ventilating an unprotected airway with positive pressure ventilation. The purpose of this study was to assess the effects of a newly developed bag-valve-mask device (SMART BAG ®, O-Two Systems International, Ont., Canada) that limits peak inspiratory flow. A bench model simulating a patient with an unintubated airway was used consisting of a face mask, manikin head, training lung (lung compliance, 100 ml/cm H 2O, airway resistance 4 cm H 2O/l/s, lower oesophageal sphincter pressure 20 cm H 2O and simulated stomach). Twenty nurses were randomised to each ventilate the manikin using a standard single person technique for 1 min (respiratory rate, 12/min) with either a standard adult self-inflating bag, or the SMART BAG ®. The volunteers were blinded to the experimental design of the model until completion of the experimental protocol. The SMART BAG ® vs. standard self-inflating bag resulted in significantly ( P<0.05) lower mean±S.D. peak inspiratory flow rates (32±2 vs. 61±13 l/min), peak inspiratory pressure (12±2 vs. 17±2 cm H 2O), lung tidal volumes (525±111 vs. 680±154 ml) and stomach tidal volumes (0±0 vs. 17±36 ml), longer inspiratory times (1.9±0.3 vs. 1.5±0.3 s), but significantly higher mask leakage (26±13 vs. 14±8%); mask tidal volumes (700±104 vs. 785±172 ml) were comparable. The mask leakage observed is not an uncommon factor in bag-valve-mask ventilation with leakage fractions of 25–40% having been previously reported. The differences observed between the standard BVM and the SMART BAG ® are due more to the anatomical design of the mask and the non-anatomical shape of the manikin face than the function of the device. Future studies should remove the mask to manikin interface and should introduce a standardized mask leakage fraction. The use of a two-person technique may have removed the problem of mask leakage. In conclusion, using the SMART BAG ® during simulated ventilation of an unintubated patient in respiratory arrest significantly decreased inspiratory flow rate, peak inspiratory pressure, stomach tidal volume, and resulted in a significantly longer inspiratory time when compared to a standard self-inflating bag.

Analysis of the neutronic data in infinite medium using fusion neutron source for various material compositions

The main purpose of this study is to calculate the integral neutronic data, which are integral heat release, 233 U and 239 Pu fissile fuels breeding rate, tritium breeding ratio, integral neutron multiplication and net neutron gain through (n,x) and fission (when applicable) reactions in an infinite medium using a fusion neutron source for different mixtures. The most important materials in fusion technology, namely tritium, beryllium, lead, thorium and uranium, have been investigated in an infinite medium. Fusion neutrons, which have a high first wall flux value of 2.22×10 14 (14.1 MeV) neutron/cm 2 s, are released in an inertial confinement fusion (ICF). ThO 2 and UC, composed with Be and Pb, are mixed with natural lithium or 6 Li for volume fractions from 0% to 100%. However, the variable ThO 2– 6 Li (ThO 2 mixed with 6 Li), UC– 6 Li (UC mixed with 6 Li), ThO 2–Nat.Li (ThO 2 mixed with Nat.Li) and UC–Nat.Li (UC mixed with Nat.Li) compositions will be mixed with Be and Pb for the volume fractions mentioned above. Integral neutron data calculations have been conducted with the help of the 3-D Monte Carlo Code. In a fusion reactor blanket with finite dimension, the integral quantities will be more or less different from the infinite medium results, depending on the neutron leakage fraction. Design studies foresee reduction of the neutron leakage out of the blanket to very low levels in order to prevent nuclear heating in the superconducting fusion magnets and to keep all neutrons primarily in the coolant.

The Comparison of Respiratory Mechanics after Laryngeal Mask Airway and Endotracheal Tube Insertion during Positive Pressure Ventilation

Background: The laryngeal mask airway (LMA) is relatively easy to use and may have advantages over an endotracheal tube (ETT). This randomized cross over study was performed to compare the respiratory parameters of a LMA and ETT in anesthetized mechanically ventilated patients. Methods: Thirty ASA physical status I and II adult patients undergoing peripheral limb surgery were studied. Anesthesia was induced with fentanyl 1.5 /kg and thiopental sodium 5 mg/kg. Anesthesia was maintained with 1 L/min - 1 L/min - enflurane (end tidal concentration 1.0 - 1.5 vol %) and muscle relaxation was accomplished using vecuronium 0.12 mg/kg. A LMA and ETT were placed in a random sequence in each patient and connected to a ventilator. After 15 minutes of each insertion at the same ventilator setting, ventilatory variables and arterial partial pressure were measured and physiological dead space/tidal volume ratio was calculated by a SMO plus! respiratory profile monitor. Results: Physiological dead space/tidal volume ratio and airway dead space in the LMA group were significantly higher compared with those in the ETT group. There were no statistically significant differences in dynamic and static compliance. Inspiratory and expiratory airway resistance were lower in the LMA group. There were no significant differences in air leakage fraction and total inspiratory work done by the ventilator on the respiratory system. Airway pressure, arterial, end-tidal and mixed expired partial pressure were similar in both groups. Conclusions: Efficiency of ventilation is similar with a LMA and ETT during positive ventilation in patients without pulmonary disease. We found no differences of clinical importance in either group of patients in the range of measured parameters.

Investigation of the infinite medium integral neutronic data for incident fusion source neutrons

In the present work, the physical behavior of integral data in infinite medium has been evaluated for incident fusion neutrons with the help to the 3-D Monte Carlo code. In a fusion reactor blanket with finite dimension, the integral quantities will be more or less different from the infinitive medium results, depending on the neutron leakage fraction. Design studies foresee the reduction of the neutron leakage out of the blanket as possible in order to prevent the nuclear heating in super conducting fusion magnets and to keep all neutrons primarily in the coolant. The most important materials in fusion technology, namely tritium, beryllium, lead, thorium, and uranium have been investigated in infinitive medium. The main purpose of this work is to calculate the integral tritium breeding ratio, 233U breeding rate, 239Pu breeding rate, heat release, neutron multiplication ratio through ( n, x) and fission (when applicable) reactions in those mixtures which are composed when first UO 2 and ThO 2 are mixed with natural lithium (Nat.Li) or 6Li for a volume fraction from 0 to 100%. Then the variable UO 2-Nat.Li (UO 2 mixed with Nat.Li) and UO 2- 6Li (UO 2 mixed with 6Li) compositions will be mixed with Beryllium (Be) and Lead (Pb) for a volume fraction from 0 to 100%. However, the variable TO 2-Nat.Li (ThO 2 mixed with Nat.Li) and ThO 2- 6Li (ThO 2 mixed with 6Li) compositions will be mixed with Be and Pb for a volume fraction mentioned above.

Design of the KSTAR Divertor

The planned Korea Superconducting Tokamak Advanced Research (KSTAR) divertor has been designed to provide reliable power handling and particle control with enough shaping flexibility to accommodate a wide range of plasma operation. The physics basis for the current configuration of the KSTAR divertor through analyses of the heat flux at the target, particle control, and plasma-facing component is reported. A simple zero-dimensional model based on the power balance assumptions and two-dimensional codes is utilized to estimate the heat flux to the divertor plate. The limit for the peak heat flux on the divertor plate, 3.5 MW/m2, requires advanced operating modes such as the radiative divertor and radiative mantle, which are considered to overcome the weakness of a high-recycling divertor. A simple particle balance model could estimate the pumping rate with total leakage fraction assuming particle sources. A Monte Carlo neutral transport calculation determines the dimension of a gap between the center and outer divertor targets. It also determines the number and best position of the pumps, as well as the geometry for conductance. For the initial 20-s discharges, a bolted-tile carbon-fiber-composite design is relied upon for the upper and lower divertor targets. The design of the supporting structure for the divertors will allow for future modifications to accommodate thermal steady-state 300-s operation or to optimize divertor performance based on new understanding gained during initial tokamak operation.

The Control of Shroud Leakage Flows to Reduce Aerodynamic Losses in a Low Aspect Ratio, Shrouded Axial Flow Turbine

The losses generated by fluid leaking across the shrouds of turbine blade rows are known to form a significant proportion of the overall loss generated in low aspect ratio turbines. The use of shrouds to encase the tips of turbine blades has encouraged the development of many innovative sealing arrangements, all of which are intended to reduce the quantity of fluid (the leakage fraction) leaking across the shroud. Modern sealing arrangements have reduced leakage fractions considerably, meaning that further improvements can only be obtained by controlling the leakage flow in such a way so as to minimize the aerodynamic losses incurred by the extraction and re-injection of the leakage flow into the mainstream. There are few published experimental investigations on the interaction between mainstream and leakage flows to provide guidance on the best means of managing the leakage flows to do this. This paper describes the development and testing of a strategy to turn the fluid leaking over shrouded turbine rotor blade rows with the aim of reducing the aerodynamic losses associated with its re-injection into the mainstream flow. The intent was to extract work from the leakage flow in the process. A four stage research turbine was used to test in detail the sealing design resulting from this strategy. A reduction in brake efficiency of 3.5 percent was measured. Further investigation suggested that much of the increase in loss could be attributed to the presence of axial gaps upstream and downstream of the shroud cavity which facilitated the periodic ingress and egress of mainstream fluid into the shroud cavity under the influence of the rotor potential field. This process was exacerbated by reductions in the leakage fraction.

Respiratory mechanics, gastric insufflation pressure, and air leakage of the laryngeal mask airway.

A potential risk of the laryngeal mask airway (LMA) is incomplete mask seal, which causes air leakage or insufflation of air into the stomach. The objective of the present study was to assess respiratory mechanics, quantify air leakage, and measure gastric air insufflation in patients ventilated via the LMA. Thirty patients were studied after induction of anesthesia but prior to any surgical manipulations. After the insertion of the LMA, patients were ventilated with increasing tidal volumes until one of the three following end points were reached: 1) gastric air insufflation, 2) airway pressure > 40 cm H2O, or 3) limitation of further increase in tidal volume by air leakage. The following variables were determined:inspired volume (VI), expired volume (VE), maximum inspiratory pressure (Pmax), airway pressure at gastric inflation (Pinfl), respiratory time constant (RC), compliance (C), resistance (R), and leakage fraction (FL). Respiratory mechanics were in the physiological range. Gastric insufflation occurred in 27% of the patients at inspiratory pressures between 19 and 33 cm H2O. Air leakage of more than 10% was evident at inspiratory pressures between 25 and 34 cm H2O. The end point of 40 cm H2O airway pressure was reached in only three patients. We conclude that the LMA is not better in preventing airway pressure transmission to the esophagus than a conventional face mask. However, a high FL is associated with reduced gastric air insufflation.

Domestic ducted forced-air heating: Duct system leakage and heating energy use

A residential ducted forced-air heating system was investigated to assess the impact of duct system leakage. Ductwork pressurisation and tracer gas techniques were used to quantify leakage in the heating duct system, and added infiltration to the house caused by duct leaks. Determination of leakage fraction by pressurisation tests showed a distribution loss of 23% in supply ductwork and 3% for the return duct during normal operation of the air heater's circulation fan. Tracer gas tests showed that the house infiltration is affected by the duct system leakage: infiltration increased from 0.1 ac h-1 for no ducting, to 0.21 and 0.5 ac h-1 with heating ductwork when the circulation fan is off and on respectively. A direct calculation shows the heat loss due to duct leakage to be about 45% of the house peak winter load. Hourly simulation of the space heating requirements indicates an annual increase of 35% in heating load due to heating duct leakage and transmission losses.