Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “power electronics cooling”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Investigation of radiated-power for low aspect ratio fusion plasmas

Investigating radiated power in fusion plasmas is of utmost importance to understand the effect of undesired impurities, such as metals, in present devices, or also desired impurities, such as noble gasses, to purposefully radiate a large fraction of power in future devices. These studies are especially important for high Z impurities which will play a crucial role in future generation fusion pilot plants. In this work, we have developed a power radiation analysis module, which is used to investigate 2D distributions of impurity densities and radiated power asymmetries caused due to both plasma rotation and the cooling rate dependence on temperature profile for the cases of experimental NSTX plasmas and designed scenarios for the spherical tokamak advanced reactor (STAR), both being low aspect ratio tokamaks. Two different atomic databases have been tested during this work to study their impact on the radiated power distribution, especially for high Z impurities. Also in PRAM, self-consistent calculations of two-dimensional electron, main ion, and impurity ion densities are carried out using one-dimensional input density, temperature, and rotation profiles. In the case of NSTX, discharges with high rotation of ∼ 170 km s, measured with charge exchange recombination spectroscopy, have been investigated. Rotation-induced charge separation, leading to an electrostatic potential, is calculated iteratively to a self-consistent solution while testing high Z impurities to observe any 2D asymmetry in the core radiated power due to centrifugal forces. The STAR design, being much larger (R = 4 m), is projected to have a much lower rotation, and is shown to have low rotation-induced asymmetries, on the order of ten percent or less, between the low field and high field sides. However, another effect not due to rotation but to the dependence of impurity cooling rates on temperature can lead to radiation peaking off-axis, near the plasma edge. This effect is noticeable for argon in NSTX, for example, but can also be enhanced for certain impurities at much higher temperatures projected for STAR (T e0 ~ 32 keV), for example for undesired tungsten or possibly desired xenon.

NSTX↗
NASA NTRS1977

Portable reflectance spectrometer

A portable reflectance spectrometer is disclosed. The spectrometer essentially includes an optical unit and an electronic recording unit. The optical unit includes a pair of thermoelectrically-cooled detectors, for detecting total radiance and selected radiance projected through a circular variable filter wheel, and is capable of operating to provide spectral data in the range 0.4 to 2.5 micrometers without requiring coventional substitution of filter elements. The electronic recording unit includes power supplies, amplifiers, and digital recording electronics designed to permit recordation of data on tape casettes. Both the optical unit and electronic recording unit are packaged to be manually portable.

Goetz, A. F. H.↗
NASA NTRS1998

Gamma-Ray Spectra and Variability of Cygnus Z-1 Observed by BATSE

We present new BATSE earth occultation observations of the 25 keV-1.8 MeV spectrum and variability of Cygnus X-1 made between August 1993 and May 1994. We observed that the normal soft gamma ray spectrum (gamma2) of Cygnus X-1 has two components: a Comptonized part seen below 300 keV, and a high-energy tail in the 0.3 - 2 MeV range. We interpret it in terms of a two-layer region, consisting of a high-energy core (with an equivalent electron temperature of approximately 210-250 keV) near the event horizon, embedded in an about 50 keV corona. In this scenario, the observed 25-300 keV photons were produced by Compton scattering of soft photons (about 0.5 keV) by the hot electrons in the outer corona. These same hard x rays were further up-scattered by a population of energetic electrons in the inner core, producing the spectral tail above 300 keV. Cygnus X-1 went through an extended sequence of transitions between August 1993 and May 1994, when the 45-140 keV flux first decreased steadily from approximately gamma2 to roughly one-quarter of its intensity over a period of about 140 days. The flux remained at this low level for about 40 days before returning, swiftly (approximately 20 days) to approximately the initial gamma2 level. During the transition, the spectrum evolved to a shape consistent with either a power law with photon index of about 2.6 or a single temperature Compton model with electron temperature kT = 110 +/- 11 keV, and optical depth t = 0.40 +/- 0.06, and then returned essentially to the original gamma2 spectrum at the end of the active period. The overall cooling of the system during the low flux period may be due to an increase in the soft photon population which effectively quenched the hot electrons in these regions through Compton scattering.

Ling, J. C.↗
NASA NTRSDate not supplied

Electronic Speed Controller Thermal Management

This experiment was coupled with an acoustic test for NASA’s QUiet Electric ENgine (QUEEN). A cold plate with water was used to cool the electronic speed controller of the motor under different conditions (shown on right). Part 1: Fixed motor RPM, translating to fixed power/heat dissipation (4W per MOSFET) of electrical components, with variable cooling capacity measured by flow rate of cold plate water in gpm. Part 2: Fixed cooling power (1 gpm) with variable RPM.

thermal↗
DOE OSTI2023

Evolution and feedback of AGN jets of different cosmic ray composition

Jet feedback from active galactic nuclei (AGN) is one of the most promising mechanisms for suppressing cooling flows in cool-core clusters. However, the composition of AGN jets and bubbles remains uncertain; they could be thermally dominated, or dominated by cosmic ray proton (CRp), cosmic ray electron (CRe), or magnetic energy. In this work, we investigate the evolution and feedback effects of CRp and CRe dominated jets by conducting 3D magnetohydrodynamic simulations of AGN jet-inflated bubbles in the intracluster medium using the FLASH code. We present the evolution of their energies, dynamics, and heating, and model their expected cavity-power versus radio-luminosity relation (P cav –L R ). We find that bubbles inflated by CRe dominated jets follow a very similar dynamical evolution to CRp dominated bubbles even though CRe within bubbles suffer significantly stronger synchrotron and inverse-Compton cooling. This is because, as CRe lose their energy, the jet-inflated bubbles quickly become thermally dominated within ∼30 Myr. Their total energy stops decreasing with CR energy and evolves similarly to CRp dominated bubbles. The ability of CRe and CRp dominated bubbles to heat the intracluster medium is also comparable; the cold gas formed via local thermal instabilities is well suppressed in both cases. The CRp and CRe bubbles follow different evolutionary trajectories on the P cav –L R plane, but the values are broadly consistent with observed ranges for FR-I sources. We also discuss observational techniques that have potential for constraining the composition of AGN jets and bubbles.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Heterogeneous Integration Technologies for High-temperature, High-density, Low-profile Power Modules of Wide Bandgap Devices in Electric Drive Applications (Final Technical Report)

The goal of this project is to develop packaging technologies for making high-temperature, high-density, and low-profile wide-bandgap (WBG) power electronics modules for electric drives. These modules are aimed at enabling the DOE’s University Consortium to reach its 2025 inverter targets of ≥ 100 kW/L and ≤ 2.7 $/kW. The specific objectives are to: design and fabricate SiC half-bridge power modules with double-sided cooling and parasitic inductances < 5 nH, heat flux density > 400 W/cm 2 , and working junction temperature of 200 o C; design, fabricate, and deliver a gate driver with double-sided cooled modules for the construction of a 100 kW/L inverter at Oak Ridge National Lab; and design and prototype intelligent gate drivers with integrated current sensor and a low-profile DC-DC power supply with air-core transformer for testing power modules at 200 o C junction temperature. We followed an iterative technical approach of design, simulation, fabrication, and testing of various versions of modules, current sensors, and power supply. The state-of-the-art silicon carbide devices rated at 1.2 kV and 149 A were packaged by sintered-silver bonding on an aluminum nitride direct-bond-copper substrate for high thermal conductivity, high working temperature, and high joint reliability. Porous silver posts were used to interconnect the device’s source pads to the other direct-bond-copper substrate for low mechanical stresses, ease of manufacturing, and double-sided cooling. A current sensor based on package parasitic inductance was developed to measure switching current. A dynamic feedback scheme was developed to compensate the effect of parasitic resistance and temperature variation. A constant-current class-E dc-dc converter with air-core transformer was developed. Air-core transformer was used due to the unavailability of magnetic core at high temperatures. Gate driver and power supply were integrated with the double-side cooled, high temperature SiC power modules for testing the modules at 200 o C junction temperature. Double-pulse and continuous testing of the integrated technologies validated the design and fabrication of the three component technologies. Throughout the project, we overcame the challenge for design verification caused by low prototyping yield, which then helped train the graduate students, the future workforce, to learn the engineering know-how for low-cost manufacturing of reliable products. Below is a summary of the major accomplishments of this project: development of a prototyping process for fabricating double-side cooled (1200 V, 149 A) SiC phase-leg modules capable of working to 200 o C Tj; simulation and experimental verification of the improvement of thermo-mechanical reliability of the double-side cooled SiC phase-leg module by using rigid encapsulant; design and experimental validation of a current sensor based on package parasitic inductance and a compensation solution for eliminating the effect of parasitic resistance; design and experimental validation of a low-profile power supply with six-output air-core transformer for gate driver; functional demonstration of a SiC phase-leg module integrated with its gate driver, current sensor, and an air-core power supply at 200 o C Tj in a double-pulse switching test setup and Buck converter continuous test setup; successful completion of six PhD and two MS students who are or will work at Apple Inc., Tesla Inc., Wolfspeed Inc., Microchip Inc., Monolithic Power Systems Inc., and LG Magna Inc.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS2013

Doppler Lidar System Design via Interdisciplinary Design Concept at NASA Langley Research Center - Part I

Optimized designs of the Navigation Doppler Lidar (NDL) instrument for Autonomous Landing Hazard Avoidance Technology (ALHAT) were accomplished via Interdisciplinary Design Concept (IDEC) at NASA Langley Research Center during the summer of 2013. Three branches in the Engineering Directorate and three students were involved in this joint task through the NASA Langley Aerospace Research Summer Scholars (LARSS) Program. The Laser Remote Sensing Branch (LRSB), Mechanical Systems Branch (MSB), and Structural and Thermal Systems Branch (STSB) were engaged to achieve optimal designs through iterative and interactive collaborative design processes. A preliminary design iteration was able to reduce the power consumption, mass, and footprint by removing redundant components and replacing inefficient components with more efficient ones. A second design iteration reduced volume and mass by replacing bulky components with excessive performance with smaller components custom-designed for the power system. Mechanical placement collaboration reduced potential electromagnetic interference (EMI). Through application of newly selected electrical components and thermal analysis data, a total electronic chassis redesign was accomplished. Use of an innovative forced convection tunnel heat sink was employed to meet and exceed project requirements for cooling, mass reduction, and volume reduction. Functionality was a key concern to make efficient use of airflow, and accessibility was also imperative to allow for servicing of chassis internals. The collaborative process provided for accelerated design maturation with substantiated function.

Boyer, Charles M.↗
DOE OSTI2025

Window Cooling Studies and Disk Vibration Testing on a Subset of Mo-100 Disks

Production of metastable Technetium-99 (Tc-99m), a radioactive tracer that emits gamma rays, is vital to the medical imaging community. Tc-99m is extracted from the decay of Molybdenum-99 (Mo-99) which has a half-life of about 2-3 days. The work presented in this report is part of the NNSA’s mission to produce Mo-99 commercially, within the US, without the use of highly enriched uranium (HEU) in support of nonproliferation and global security. Los Alamos National Laboratory (LANL) is working with NorthStar Medical Radioisotopes (NMR) on their efforts to produce Mo-99 through the irradiation of Mo-100 targets using an electron beam. The NMR target comprises a stack of approximately 60-70 Mo-100 disks with diameter 24 mm and thickness 0.74 mm held in stainless steel laminations, each separated using 0.25 mm thick stainless-steel spacers. The symmetric target stack is housed in an Inconel vessel with two Inconel windows on either side. Two electron accelerators are used to produce 40 MeV, 3.16 µA electron beams each that penetrate the Inconel windows and irradiate the Mo-100 disks. Approximately 90% of the total 250 kW beam power is deposited in the NMR target during the irradiation process, with a smaller percentage adding up to 2.2 kW of heat deposited on the Inconel window. During irradiation, pressurized helium gas flows through thin gaps between the disks cooling the beam window, target disks, disk laminations and spacers. Both NMR and LANL have found during cold testing of the target system (no heat deposition) that the Mo 100 disks undergo significant mass loss and disk breakage due to vibrations induced by the flowing helium gas. The mass loss is not only undesirable due to monetary loss from reduced final quantities of Mo-99, but also due to the hazards associated with radioactive material trapped in the cooling lines and particle filters. The effect of flow rate and target geometry on the flow induced vibrations need to be quantified, and recommendations provided to minimize this mass loss. LANL has previously also tested NMR’s Inconel beam window by heating the window, while flowing pressurized helium, using the average heat deposited on the window. However, the NMR beam is pulsed with a duty cycle of 12.5%, which introduces oscillation in temperature around the nominal 600 °C steady state value with each pulse. Available fatigue curves for Inconel are few, established for room temperature, and they are based on mechanical strain cycles not thermally induced strain as in the NMR target. The effect of pulsed beam heating on the Inconel window therefore needs to be quantified. This report details the experiments conducted to assess the factors that lead to mass loss in the NMR target disks as well as to understand the effect of a pulsed beam on NMR’s Inconel window. This work describes LANL’s experimental characterization of the flow induced vibrations and disk mass loss in a reduced scale set-up containing 5 to 10 Mo-100 disks. We use high speed imaging, displacement measurements and microphone measurements combined with signal processing to estimate the vibration frequency of each disk. The effect of disk thickness, target fit and duration of testing on the mass loss is described. We find that in the current configuration of NMR targets, the vibrations and mass loss on the first disk are minimized, while those in the adjacent disks are highest. The microphone and high-speed image data show that increased flow rates and increased duration of testing increases vibration frequency and mass loss. The mass loss is due to both disk rotation and back and forth motion. There are visible wear marks on the disks with the highest mass loss. We also note that the current NMR window gap reduces flow induced vibrations compared to the previous smaller gaps. Improved target holders significantly reduce disk mass loss to almost negligible quantities. This work finds that the larger window to first disk gap and improved target holder geometry should allow NMR to successfully conduct irradiations with minimal mass loss. The window tests were conducted to understand the effect of a pulsed beam on both the window longevity and to estimate the window temperature and displacement during pulsing. The experiments presented here were performed at significantly low power, due to the limitations of the induction heating system. The window temperature rose to approximately 73 °C with a significantly reduced power of 45 W without beam pulsing. With a 5 Hz pulse rate, 12.5% duty cycle, the window temperature remained constant at 26 °C. These experiments will be repeated with improved coil geometry and reported in upcoming journal papers.

42 ENGINEERING↗
DOE OSTI2025

Noise in Intense Electron Bunches (Final Technical Report)

The goal of this project is to investigate density fluctuations in electron bunches. Noise and density fluctuations in relativistic electron bunches, accelerated in a linac, are of critical importance to various Coherent Electron Cooling (CEC) concepts as well as to free-electron lasers (FELs). For CEC, the beam noise results in additional diffusion that counteracts cooling; and if this noise is not controlled at sufficiently low level, the noise heating effects can overcome cooling. There have been several proposals in the past to suppress the noise in the beam in the frequency range of interest in order to optimize the cooling effects. In SASE FELs a microwave instability starts from the initial noise in the beam and eventually leads to the beam microbunching yielding coherent radiation, and the initial noise in the FEL bandwidth plays a useful role. In seeded FELs, in contrast, such noise interferes with the seed signal, so that reducing noise at the initial seed wavelength would lower the seed laser power requirement. Our major goals were (1) to measure the electron beam density noise level in a 0.5 – 10 um wavelength range, (2) to predict the beam noise level in order to compare with the measurements, and (3) as a stretch goal, to find mechanisms that affect the beam noise to control its level in a predictable manner.

43 PARTICLE ACCELERATORS↗
NASA NTRSDate not supplied

Ceramic Heat Pipe for Thermal Management of a High Frequency Inductor

As the power density of electric machines and power electronics continues to increase to meet the demands of more electric aircraft, the thermal management of these devices becomes increasingly important. Ideally, the thermal management solution of the device should have high thermal conductivity and be electrically insulating so that it can be directly integrated with the device in question. Technical ceramics are a class of materials that meet these two requirements. Direct liquid cooling by submerging electric components in a dielectric fluid is an alternate option, but adds system complexity and may incur pumping loss. In this work, the thermal management solution for a high frequency inductor is presented. A conventional wicked heat pipe is designed that is made from a solid ceramic case with a sintered ceramic wick. This heat pipe can provide structural support for the core and windings and is electrically insulating, so it does not incur any additional eddy current losses from the high frequency magnetic fields. It does not require additional pumping power, because conventional wicked heat pipes are passive devices. In this paper, the manufacturing process of the ceramic heat pipe is also presented. It is shown that the sintering of large diameter spherical ceramic particles is difficult due to their high specific surface area and low surface free energy. To overcome this difficulty, a flux material was used to initiate liquid phase sintering. The results of the sintering trials show that for the 120 μm diameter α-Al2O3powder selected for the wick, the combination of 1650°C sintering temperature and the addition of 5 wt% talc sintered the porous matrix together sufficiently for use as a heat pipe wic

High-frequency Inductor↗
NASA NTRSDate not supplied

Ceramic Heat Pipe for Thermal Management of a High Frequency Inductor

As the power density of electric machines and power electronics continues to increase to meet the demands of more electric aircraft, the thermal management of these devices becomes increasingly important. Ideally, the thermal management solution of the device should have high thermal conductivity and be electrically insulating so that it can be directly integrated with the device in question. Technical ceramics are a class of materials that meet these two requirements. Direct liquid cooling by submerging electric components in a dielectric fluid is an alternate option, but adds system complexity and may incur pumping loss. In this work, the thermal management solution for a high frequency inductor is presented. A conventional wicked heat pipe is designed that is made from a solid ceramic case with a sintered ceramic wick. This heat pipe can provide structural support for the core and windings and is electrically insulating, so it does not incur any additional eddy current losses from the high frequency magnetic fields. It does not require additional pumping power, because conventional wicked heat pipes are passive devices. In this paper, the manufacturing process of the ceramic heat pipe is also presented. It is shown that the sintering of large diameter spherical ceramic particles is difficult due to their high specific surface area and low surface free energy. To overcome this difficulty, a flux material was used to initiate liquid phase sintering. The results of the sintering trials show that for the 120 μm diameter α-Al2O3 powder selected for the wick, the combination of 1650°C sintering temperature and the addition of 5 wt% talc sintered the porous matrix together sufficiently for use as a heat pipe wick.

High frequency inductor↗
DOE OSTI2024

Improving Cathode Testing with a High-Gradient Cryogenic Normal Conducting RF Photogun

Future electron accelerator applications such as X-ray free electron lasers and colliders are dependent on significantly increasing beam brightness. With the observation that linac beam manipulation’s best preservation of max brightness is at the cathode, we are incentivized to create an environment where we can study how to achieve the highest possible photogun brightness. In order to do so, we intend to extract beams from high-brightness photocathodes with the highest achievable accelerating gradients we can manage in a klystron-powered radiofrequency (RF) photogun. We utilize here cryogenic normal conducting cavities to achieve ultra-high gradients via limitation of breakdown rates (BDR). The low temperatures should also reduce cathode emittance by reducing the mean transverse energy (MTE) of electrons near the photoemission threshold. To this end, we have designed and produced a new CrYogenic Brightness-Optimized Radiofrequency Gun (CYBORG) for use in a new beamline at UCLA. We will introduce the enabling RF and photoemission physics as a primer for the new regime of high field low temperature cathodes we intend to enter. We further report the current status of the beamline commissioning, including the cooling of the photogun to 100 K, and producing 0.5 MW of RF feed power, which corresponds to cathode accelerating fields in the range of 80–90 MV/m. We further plan iterative improvements to both to 77 K and 1 MW corresponding to our ultimate goal >120 MV/m. Our discussion will include future beamline tests and the consideration of the initial realization of an ultra-high-gradient photoinjector concept.

Lawler, Gerard Emile (ORCID:0000000160113435)↗
NASA NTRS2015

Compact Vapor Chamber Cools Critical Components

Advancements in the production of proton exchange membrane fuel cells have NASA considering their use as a power source for spacecraft and robots in future space missions. With SBIR funding from Glenn Research Center, Lancaster, Pennsylvania-based Thermacore Inc. developed strong, lightweight titanium vapor chambers to keep the fuel cells operating at optimum temperatures. The company is now selling the technology for cooling electronic components.

Source record↗
NASA NTRS1982

A lightweight electronically commutated dc motor for electric passenger vehicles

A functional model breadboard converter and a rare-earth-cobalt, permanent magnet motor; as well as an engineering model converter and PM motor suitable for vehicle installations were developed and tested. The converter and motor achieved an 88% peak efficiency, a maximum output of 26 kW at 26,000 rpm, and a continuous rating of 15 kW. The system also generated power to the source during braking, with a demonstrated peak power available at the converter terminals of approximately 26 kW at 88% efficiency. Major conclusions include: (1) the SAE J227a(D) driving cycle efficiency for the converter/motor is 86% to 88% when energy available for recovery at the converter terminals is included; (2) the converter initial cost is approximately five times that of the permanent magnet motor, but can be reduced by means of LSI logic and integrated liquid cooled semiconductor packages; and (3) an electronically commutated motor with a liquid cooled converter will operate reliably without service or maintenance for the life of a passenger vehicle.

Echolds, E. F.↗
NASA NTRS1985

Absolute response and noise equivalent power of cyclotron resonance-assisted InSb detectors at submillimeter wavelengths

Spectra are presented of the responsivity and noise equivalent power (NEP) of liquid-helium-cooled InSb detectors as a function of magnetic field in the range 20-110 per cm. The measurements are all made using a Fourier transform spectrometer with thermal sources. The results show a discernable peak in the detector response at the conduction electron cyclotron resonance (CCR) frequency for magnetic fields as low as 3 kG. The magnitude of responsivity at the resonance peaks is roughly constant with magnetic field and is comparable to the low-frequency hot-electron bolometer response. The NEP at the peaks is found to be comparable to the best long wavelength results previously reported. For example, NEP = 4.5 x 10 to the 13th W/(square root of Hz) at 4.2 K, 6 kG, and 40 per cm was measured. The InSb CCR will provide a much improved detector for laboratory spectroscopy, as compared with hot electron bolometers, in the 20-100 per cm range.

Brown, E. R.↗
DOE OSTI2025

The STAR Forward Silicon Tracker

The Forward Silicon Tracker (FST) is a pivotal component of the forward upgrade of the Solenoidal Tracker at RHIC (STAR), designed to discern hadron charge signs with a momentum resolution better than 30% for 0.2 < p T < 2GeV/c in the 2.5 < η < 4 pseudorapidity range. Its compact design features three disks along the beam direction, minimized material budget, and scattering effects. The FST uses Hamamatsu’s p-in-n silicon strip sensors with a double metal layer that enables efficient signal routing to the readout electronics, enhancing overall detector performance. The flexible hybrid boards, essential for the readout system, are constructed with Kapton and copper layers to optimize signal handling and power distribution. These boards connect silicon strips to analogue pipeline ASIC APV25-S1 chips, which read up to 128 channels each. A cooling system with nonconducting, volatile NOVEC 7200 coolant at 22.2 °C mitigates ASIC-generated heat. Furthermore, the FST enhances forward tracking performance at STAR as an integral part of the forward upgrade.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

First conduction cooled photoinjector status

SRF photoguns become a promising candidate to produce highly stable electrons for UEM/UED applications because of the ultrahigh shot-to-shot stability compared to room temperature RF photoguns. SRF technology was prohibitively expensive for industrial use until two recent advancements: Nb3Sn and conduction cooling. SRF gun can provide a CW operation capability while consuming only 2W of RF power which eliminates the need of an expensive high power RF system and saves a facility footprint. Euclid is developing a continuous wave (CW), 1.5-cell, MeV-scale SRF conduction cooled photogun operating at 1.3 GHz. We aim for generation of the first beam in 2024. In this paper, we present the most up-to-date progress including results of the first cool down of the gun-cavity in the newly developed conduction cooled cryomodule and LLRF system development.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Pulse-by-pulse transient thermal deformation in crystal optics under high-repetition-rate FEL

Time-domain modeling of the thermal deformation of crystal optics can help define acceptable operational ranges across the pulse-energy repetition-rate phase space. In this paper, we have studied the transient thermal deformation of a water-cooled diamond crystal for a cavity-based X-ray free-electron laser (CBXFEL), either an X-ray free-electron laser oscillator (XFELO) or a regenerative amplifier X-ray free-electron laser (RAFEL), by numerical simulations including finite-element analysis and advanced data processing. Pulse-by-pulse transient thermal deformation of a 50 µm-thick diamond crystal has been performed with X-ray pulse repetition rates between 50 kHz and 1 MHz. Results for temperature and thermal deformation have been compared with the results of transient analysis using a continuous wave (CW) power loading. Temperature and thermal deformation results from pulse-by-pulse transient analysis vary with time about the results for the CW case for the same average power. The variation amplitude increases with pulse energy and decreases with repetition rate. When the repetition rate increases to infinity, both temperature and thermal deformation converge to the results for the CW case. Two critical time scales for the operation of crystal optics in a CBXFEL are (1) first-turn time, i.e. the time for the XFEL pulse to complete the first turn around the cavity so that the crystal sees the recirculated XFEL pulse, and (2) period-end time, i.e. the time that the next electron bunch arrives for the amplification, so that the crystal outcouples the amplified FEL power. For the same average power, simulation results show that the crystal thermal deformation seen by the XFEL beam decreases with repetition rate at the first-turn time of a 300 m-long cavity and increases with repetition rate at the period-end time. For the wavefront preservation requirement of the crystal optics, a pulse-energy versus repetition-rate phase space has been established. The upper bounds of the pulse energy at both first-turn and period-end times decreases with repetition rate, especially at the period-end time. The upper bound of the thermal deformation of the crystal at the period-end time for any repetition frequency can be estimated from the CW case. For a water-cooled diamond crystal of dimension 5 mm × 5 mm × 0.05 mm, the time to reach a quasi steady-state is about 50 ms for temperature and 50 µs for thermal deformation.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗