Pedestal Pressure Research Articles

Exploration of the Super H-mode regime on DIII-D and potential advantages for burning plasma devices

A new high pedestal regime (“Super H-mode”) has been predicted and accessed on DIII-D. Super H-mode was first achieved on DIII-D using a quiescent H-mode edge, enabling a smooth trajectory through pedestal parameter space. By exploiting Super H-mode, it has been possible to access high pedestal pressures at high normalized densities. While elimination of Edge localized modes (ELMs) is beneficial for Super H-mode, it may not be a requirement, as recent experiments have maintained high pedestals with ELMs triggered by lithium granule injection. Simulations using TGLF for core transport and the EPED model for the pedestal find that ITER can benefit from the improved performance associated with Super H-mode, with increased values of fusion power and gain possible. Similar studies demonstrate that the Super H-mode pedestal can be advantageous for a steady-state power plant, by providing a path to increasing the bootstrap current while simultaneously reducing the demands on the core physics performance.

Discovery of stationary operation of quiescent H-mode plasmas with net-zero neutral beam injection torque and high energy confinement on DIII-D

Recent experiments in DIII-D [J. L. Luxon et al., in Plasma Physics and Controlled Nuclear Fusion Research 1996 (International Atomic Energy Agency, Vienna, 1987), Vol. I, p. 159] have led to the discovery of a means of modifying edge turbulence to achieve stationary, high confinement operation without Edge Localized Mode (ELM) instabilities and with no net external torque input. Eliminating the ELM-induced heat bursts and controlling plasma stability at low rotation represent two of the great challenges for fusion energy. By exploiting edge turbulence in a novel manner, we achieved excellent tokamak performance, well above the H98y2 international tokamak energy confinement scaling (H98y2 = 1.25), thus meeting an additional confinement challenge that is usually difficult at low torque. The new regime is triggered in double null plasmas by ramping the injected torque to zero and then maintaining it there. This lowers E × B rotation shear in the plasma edge, allowing low-k, broadband, electromagnetic turbulence to increase. In the H-mode edge, a narrow transport barrier usually grows until MHD instability (a peeling ballooning mode) leads to the ELM heat burst. However, the increased turbulence reduces the pressure gradient, allowing the development of a broader and thus higher transport barrier. A 60% increase in pedestal pressure and 40% increase in energy confinement result. An increase in the E × B shearing rate inside of the edge pedestal is a key factor in the confinement increase. Strong double-null plasma shaping raises the threshold for the ELM instability, allowing the plasma to reach a transport-limited state near but below the explosive ELM stability boundary. The resulting plasmas have burning-plasma-relevant βN = 1.6–1.8 and run without the need for extra torque from 3D magnetic fields. To date, stationary conditions have been produced for 2 s or 12 energy confinement times, limited only by external hardware constraints. Stationary operation with improved pedestal conditions is highly significant for future burning plasma devices, since operation without ELMs at low rotation and good confinement is key for fusion energy production.

Scaling studies of H-mode pedestal characteristics on Experimental Advanced Superconducting Tokamak

The characteristics of high-confinement mode (H-mode) pedestal are examined on the Experimental Advanced Superconducting Tokamak. It is found that they are closely dependent on each other between electron pedestal characteristics and global parameters for all types of edge localized mode (ELM). The scaling of pedestal temperature based on thermal conduction and pedestal pressure width is carried out. Based on pedestal pressure gradient and pedestal density, six pedestal pressure width models are applied to predict the pedestal temperature height of type I ELMy H-mode. Compared to experimental results, the normalized poloidal beta model is more consistent than other models.

Plasma confinement at JET

Operation with a Be/W wall at JET (JET-ILW) has an impact on scenario development and energy confinement with respect to the carbon wall (JET-C). The main differences observed were (1) strong accumulation of W in the plasma core and (2) the need to mitigate the divertor target temperature to avoid W sputtering by Be and other low Z impurities and (3) a decrease of plasma energy confinement. A major difference is observed on the pedestal pressure, namely a reduction of the pedestal temperature which, due to profile stiffness the plasma core temperature is also reduced leading to a degradation of the global confinement. This effect is more pronounced in low βN scenarios. At high βN, the impact of the wall on the plasma energy confinement is mitigated by the weaker plasma energy degradation with power relative to the IPB98(y, 2) scaling calculated empirically for a CFC first wall. The smaller tolerable impurity concentration for tungsten (<10−5) compared to that of carbon requires the use of electron heating methods to prevent W accumulation in the plasma core region as well as gas puffing to avoid W entering the plasma core by ELM flushing and reduction of the W source by decreasing the target temperature. W source and the target temperature can also be controlled by impurity seeding. Nitrogen and Neon have been used and with both gases the reduction of the W source and the target temperature is observed. Whilst more experiments with Neon are necessary to assess its impact on energy confinement, a partial increase of plasma energy confinement is observed with Nitrogen, through the increase of edge temperature. The challenge for scenario development at JET is to extend the pulse length curtailed by its transient behavior (W accumulation or MHD), but more importantly by the divertor target temperature limits. Re-optimisation of the scenarios to mitigate the effect of the change of wall materials maintaining high global energy confinement similar to JET-C is underway and JET has successfully achieved H98(y,2) = 1 for plasma currents up to 2.5 MA at moderate βN.

Dependence of pedestal structure on collisionality at fixed beta in JT-60U

The dependence of pedestal structure on collisionality at fixed beta has been investigated in JT-60U. In the ITER-relevant low collisionality regime, the pedestal width does not change with edge collisionality. In the high collisionality regime, the pedestal width broadens with increased edge collisionality. The pedestal pressure gradient and width are not significantly changed when the pedestal is close to an intermediate n peeling-ballooning mode boundary at low collisionality. The experimental result indicates that conventional pedestal models where the pedestal width is independent of collisionality and is determined by at the pedestal is not a bad assumption in the ITER-relevant low collisionality regime. On the other hand, the pressure gradient decreases and the pedestal width increases at high collisionality. The pedestal broadening becomes significant when the pedestal is marginal to be unstable at the high n ballooning mode in the high collisionality regime.

Taming the Heat Flux Problem: Advanced Divertors Towards Fusion Power

The next generation fusion machines are likely to face enormous heat exhaust problems. In addition to summarizing major issues and physical processes connected with these problems, we discuss how advanced divertors, obtained by modifying the local geometry, may yield workable solutions. We also point out that: (1) the initial interpretation of recent experiments show that the advantages, predicted, for instance, for the X-divertor (in particular, being able to run a detached operation at high pedestal pressure) correlate very well with observations, and (2) the X-D geometry could be implemented on ITER (and DEMOS) respecting all the relevant constraints. A roadmap for future research efforts is proposed.

Experimental study of pedestal turbulence on EAST tokamak

Turbulence in the pedestal region of the EAST tokamak has been observed and studied using reflectometry. In lower hybrid wave (LHW) or neutral beam injection (NBI) dominated heating plasma, a coherent mode (CM) was usually observed in the ELM-free phase just after the L-H transition. The CM rotated in the electron diamagnetic drift (EDD) direction in the laboratory frame with a poloidal wave number (kθ) of 0.5 cm−1–0.7 cm−1 and its frequency usually chirped from 80 – 100 kHz down to 40 – 50 kHz as the pedestal evolved. The appearance of this mode reduced the increasing rate of pedestal pressure, implying that the CM may have an effect on outward pedestal transport. This mode can exist every ten milliseconds and is finally replaced by broadband (BB) fluctuation in the later ELM-free phase. It was found that the appearance and disappearance of the CM was correlated to the pedestal pressure. In the inter-ELM phase, the pedestal turbulence is generally dominated by BB fluctuation with poloidal wave numbers from 0 to 3 cm−1 rotating in the EDD direction in the laboratory frame. Analysis shows that the pedestal pressure increasing rate dpe, ped/dt decreases with the amplitude of the BB fluctuation, implying that the BB fluctuation may play an important role in pedestal evolution. The preliminary observation on the fluctuation just inside the pedestal top is also presented.

Assessment of operational space for long-pulse scenarios in ITER

The operational space (Ip − n) for long-pulse scenarios (Δtburn ∼ 1000 s, Q ⩾ 5) of ITER has been assessed by 1.5D core transport modelling with pedestal parameters predicted by the EPED1 code by a set of transport codes under a joint activity carried out by the Integrated Operational Scenario ITPA group. The analyses include the majority of transport models (CDBM, GLF23, Bohm/gyroBohm (BgB), MMM7.1, MMM95, Weiland, scaling-based) presently used for interpretation of experiments and ITER predictions. The EPED1 code was modified to take into account boundary conditions predicted by SOLPS4 for ITER. In contrast to standard EPED1 assumptions, EPED1 with the SOLPS boundary conditions predicts no degradation of the pedestal pressure as density is reduced. Lowering the plasma density to ne ∼ (5–6) × 1019 m−3 leads to an increased plasma temperature (similar pedestal pressure), which reduces the loop voltage and increases the duration of the burn phase to Δtburn ∼ 1000 s with Q ⩾ 5 for Ip ⩾ 13 MA at moderate normalized pressure (βN ∼ 2). These ITER plasmas require the same level of additional heating power as the reference Q = 10 inductive scenario at 15 MA (33 MW NBI and 17–20 MW EC heating and current drive power). However, unlike the ‘hybrid’ scenarios considered previously, these H-mode plasmas do not require specially shaped q profiles nor improved confinement in the core for the transport models considered in this study. Thus, these medium density H-mode plasma scenarios with Ip ⩾ 13 MA present an attractive alternative to hybrid scenarios to achieve ITER's long-pulse Q ⩾ 5 scenario and deserve further analysis and experimental demonstration in present tokamaks.

Enhanced H-mode pedestals with lithium injection in DIII-D

Periods of edge localized mode (ELM)-free H-mode with increased pedestal pressure and width were observed in the DIII-D tokamak when density fluctuations localized to the region near the separatrix were present. Injection of a powder of 45 µm diameter lithium particles increased the duration of the enhanced pedestal phases to up to 350 ms, and also increased the likelihood of a transition to the enhanced phase. Lithium injection at a level sufficient for triggering the extended enhanced phases resulted in significant lithium in the plasma core, but carbon and other higher Z impurities as well as radiated power levels were reduced. Recycling of the working deuterium gas appeared unaffected by this level of lithium injection. The ion scale, kθρs ∼ 0.1–0.2, density fluctuations propagated in the electron drift direction with f ∼ 80 kHz and occurred in bursts every ∼1 ms. The fluctuation bursts correlated with plasma loss resulting in a flattening of the pressure profile in a region near the separatrix. This localized flattening allowed higher overall pedestal pressure at the peeling–ballooning stability limit and higher pressure than expected under the EPED model due to reduction of the pressure gradient below the ‘ballooning critical profile’. Reduction of the ion pressure by lithium dilution may contribute to the long ELM-free periods.

Improved confinement in high-density H-modes via modification of the plasma boundary with lower hybrid wavesa)

Injecting Lower Hybrid Range of Frequency (LHRF) waves into Alcator C-Mod's high-density H-mode plasmas has led to enhanced global energy confinement by increasing pedestal temperature and pressure gradients, decreasing the separatrix density, modifying the pedestal radial electric field and rotation, and decreasing edge turbulence. These experiments indicate that edge LHRF can be used as an actuator to increase energy confinement via modification of boundary quantities. H98-factor increases of up to ∼35% (e.g., H98 from 0.75 to 1.0) are seen when moderate amounts of LH power (PLH/Ptot ∼ 0.15) are applied to H-modes of densities n¯e ∼ 3 × 1020 m−3, corresponding to values ∼0.5 of the Greenwald density. However, the magnitude of the improvement is reduced if the confinement quality of the target H-mode plasma is already good (i.e., H98target ∼ 1). Ray-tracing modeling and accessibility calculations for the LH waves indicate that they do not penetrate to the core. The LHRF power appears to be deposited in p...

Improved confinement in JET high β plasmas with an ITER-like wall

The replacement of the JET carbon wall (C-wall) by a Be/W ITER-like wall (ILW) has affected the plasma energy confinement. To investigate this, experiments have been performed with both the C-wall and ILW to vary the heating power over a wide range for plasmas with different shapes. It was found that the power degradation of thermal energy confinement was weak with the ILW; much weaker than the IPB98(y,2) scaling and resulting in an increase in normalized confinement from H98 ∼ 0.9 at βN ∼ 1.5 to H98 ∼ 1.2−1.3 at βN ∼ 2.5 − 3.0 as the power was increased (where H98 = τE/τIPB98(y,2) and βN = βTBT/aIP in % T/mMA). This reproduces the general trend in JET of higher normalized confinement in the so-called ‘hybrid’ domain, where normalized β is typically above 2.5, compared with ‘baseline’ ELMy H-mode plasmas with βN ∼ 1.5 − 2.0. This weak power degradation of confinement, which was also seen with the C-wall experiments at low triangularity, is due to both increased edge pedestal pressure and core pressure peaking at high power. By contrast, the high triangularity C-wall plasmas exhibited elevated H98 over a wide power range with strong, IPB98(y,2)-like, power degradation. This strong power degradation of confinement appears to be linked to an increase in the source of neutral particles from the wall as the power increased, an effect that was not reproduced with the ILW. The reason for the loss of improved confinement domain at low power with the ILW is yet to be clarified, but contributing factors may include changes in the rate of gas injection, wall recycling, plasma composition and radiation. The results presented in this paper show that the choice of wall materials can strongly affect plasma performance, even changing confinement scalings that are relied upon for extrapolation to future devices.

Key impact of finite-beta and fast ions in core and edge tokamak regions for the transition to advanced scenarios

Extensive linear and non-linear gyrokinetic simulations and linear magnetohydrodynamic (MHD) analyses performed for JET hybrid discharges with improved confinement have shown that the large population of fast ions found in the plasma core under particular heating conditions has a strong impact on core microturbulence and edge MHD by reducing core ion heat fluxes and increasing pedestal pressure in a feedback mechanism. In the case of the ITER like wall, it is shown how this mechanism plays a decisive role for the transition to plasma regimes with improved confinement and it can explain the weak power degradation obtained in dedicated power scans. The mechanism is found to be highly dependent on plasma triangularity as it changes the balance between the improvement in the plasma core and the edge. The feedback mechanism can play a similar role in the ITER hybrid scenario as in the JET discharges analysed due to its high triangularity plasmas and the large amount of fast ions generated in the core by the heating systems and the alpha power.

Roles of argon seeding in energy confinement and pedestal structure in JT-60U

The mechanism of improving energy confinement with argon seeding at high density has been investigated in JT-60U. Better confinement is sustained at high density by argon seeding accompanied by higher core and pedestal temperatures. The electron density profiles become flatter with increasing density in conventional H-mode plasmas, whereas peaked density profiles are maintained with argon seeding. Density peaking and dilution effects lower the pedestal density at a given averaged density. The pedestal density in the argon seeded plasmas, which is lower than that in plasmas with deuterium puff, enables the pedestal temperature to be higher, whereas the increase in the pedestal pressure with argon seeding is small. High pedestal temperature is a boundary condition for high core temperature through profile stiffness, which leads to better confinement with argon seeding. The density peaking is a key factor of sustaining better confinement in argon seeded H-mode plasmas. The radiative loss power density is predominantly enhanced in the edge region by argon puff. The role of argon seeding in the pedestal characteristics has also been examined. The pedestal width becomes larger continuously with edge collisionality, but is nearly independent of the presence of argon seeding.

Progress at JET in integrating ITER-relevant core and edge plasmas within the constraints of an ITER-like wall

This paper reports the progress made at JET-ILW on integrating the requirements of the reference ITER baseline scenario with normalized confinement factor of 1, at a normalized pressure of 1.8 together with partially detached divertor whilst maintaining these conditions over many energy confinement times. The 2.5 MA high triangularity ELMy H-modes are studied with two different divertor configurations with D-gas injection and nitrogen seeding. The power load reduction with N seeding is reported. The relationship between an increase in energy confinement and pedestal pressure with triangularity is investigated. The operational space of both plasma configurations is studied together with the ELM energy losses and stability of the pedestal of unseeded and seeded plasmas. The achievement of stationary plasma conditions over many energy confinement times is also reported.

Low EffortLiNuclear Fusion Plasma Control Using Model Predictive Control Laws

One of the main problems of fusion energy is to achieve longer pulse duration by avoiding the premature reaction decay due to plasma instabilities. The control of the plasma inductance arises as an essential tool for the successful operation of tokamak fusion reactors in order to overcome stability issues as well as the new challenges specific to advanced scenarios operation. In this sense, given that advanced tokamaks will suffer from limited power available from noninductive current drive actuators, the transformer primary coil could assist in reducing the power requirements of the noninductive current drive sources needed for current profile control. Therefore, tokamak operation may benefit from advanced control laws beyond the traditionally used PID schemes by reducing instabilities while guaranteeing the tokamak integrity. In this paper, a novel model predictive control (MPC) scheme has been developed and successfully employed to optimize both current and internal inductance of the plasma, which influences the L-H transition timing, the density peaking, and pedestal pressure. Results show that the internal inductance and current profiles can be adequately controlled while maintaining the minimal control action required in tokamak operation.

The H-mode pedestal structure and its role on confinement in JET with a carbon and metal wall

We present the pedestal structure, as determined from the high-resolution Thomson scattering measurements, for a database of low and high triangularity (δ ≈ 0.22–0.39) 2.5 MA, type I ELMy H-mode JET plasmas after the installation of the new ITER-like wall (JET-ILW). The database explores the effect of increasing deuterium fuelling and nitrogen seeding with a view to explain the observed changes in performance (edge and global). The low triangularity JET-ILW plasmas show no significant change in performance and pedestal structure with increasing gas dosing. These results are in good agreement with EPED1 predictions. At high triangularity, for pure deuterium fuelled JET-ILW plasmas, there is a 20–30% reduction in global performance and pressure pedestal height in comparison to JET-C plasmas. This reduction in performance is primarily due to a degradation of the temperature pedestal height. The global performance and pressure pedestal height of JET-ILW plasmas can be partially recovered to that of JET-C plasmas with additional nitrogen seeding (Giroud et al 2013 Nucl. Fusion 53 113025). This observed improvement in performance is predominately due to a significant increase in density pedestal height as well as a small increase in the temperature pedestal height. A key result with increasing deuterium fuelling for JET-ILW plasmas is there is no improvement in pressure pedestal height however the pedestal still widens which is inconsistent with the scaling. Furthermore, a key result with increasing nitrogen seeding is the pressure pedestal widening is due to an increase in the temperature pedestal width whilst the density pedestal shows no clear trend. The comparison of EPED1 predictions with the measurements at high triangularity is complex as, for example, for pure deuterium fuelled plasmas there is very good agreement for the pedestal height but not the width. In addition, current EPED1 runs under-predict the pedestal height and width at high nitrogen seeding for JET-ILW plasmas however further work is required to determine the significance of these deviations. Understanding these deviations is essential as provides an insight to the physical mechanisms governing the pedestal structure and edge performance.

Progress in preparing scenarios for operation of the International Thermonuclear Experimental Reactor

The development of operating scenarios is one of the key issues in the research for ITER which aims to achieve a fusion gain (Q) of ∼10, while producing 500 MW of fusion power for ≥300 s. The ITER Research plan proposes a success oriented schedule starting in hydrogen and helium, to be followed by a nuclear operation phase with a rapid development towards Q ∼ 10 in deuterium/tritium. The Integrated Operation Scenarios Topical Group of the International Tokamak Physics Activity initiates joint activities among worldwide institutions and experiments to prepare ITER operation. Plasma formation studies report robust plasma breakdown in devices with metal walls over a wide range of conditions, while other experiments use an inclined EC launch angle at plasma formation to mimic the conditions in ITER. Simulations of the plasma burn-through predict that at least 4 MW of Electron Cyclotron heating (EC) assist would be required in ITER. For H-modes at q95 ∼ 3, many experiments have demonstrated operation with scaled parameters for the ITER baseline scenario at ne/nGW ∼ 0.85. Most experiments, however, obtain stable discharges at H98(y,2) ∼ 1.0 only for βN = 2.0–2.2. For the rampup in ITER, early X-point formation is recommended, allowing auxiliary heating to reduce the flux consumption. A range of plasma inductance (li(3)) can be obtained from 0.65 to 1.0, with the lowest values obtained in H-mode operation. For the rampdown, the plasma should stay diverted maintaining H-mode together with a reduction of the elongation from 1.85 to 1.4. Simulations show that the proposed rampup and rampdown schemes developed since 2007 are compatible with the present ITER design for the poloidal field coils. At 13–15 MA and densities down to ne/nGW ∼ 0.5, long pulse operation (&amp;gt;1000 s) in ITER is possible at Q ∼ 5, useful to provide neutron fluence for Test Blanket Module assessments. ITER scenario preparation in hydrogen and helium requires high input power (&amp;gt;50 MW). H-mode operation in helium may be possible at input powers above 35 MW at a toroidal field of 2.65 T, for studying H-modes and ELM mitigation. In hydrogen, H-mode operation is expected to be marginal, even at 2.65 T with 60 MW of input power. Simulation code benchmark studies using hybrid and steady state scenario parameters have proved to be a very challenging and lengthy task of testing suites of codes, consisting of tens of sophisticated modules. Nevertheless, the general basis of the modelling appears sound, with substantial consistency among codes developed by different groups. For a hybrid scenario at 12 MA, the code simulations give a range for Q = 6.5–8.3, using 30 MW neutral beam injection and 20 MW ICRH. For non-inductive operation at 7–9 MA, the simulation results show more variation. At high edge pedestal pressure (Tped ∼ 7 keV), the codes predict Q = 3.3–3.8 using 33 MW NB, 20 MW EC, and 20 MW ion cyclotron to demonstrate the feasibility of steady-state operation with the day-1 heating systems in ITER. Simulations using a lower edge pedestal temperature (∼3 keV) but improved core confinement obtain Q = 5–6.5, when ECCD is concentrated at mid-radius and ∼20 MW off-axis current drive (ECCD or LHCD) is added. Several issues remain to be studied, including plasmas with dominant electron heating, mitigation of transient heat loads integrated in scenario demonstrations and (burn) control simulations in ITER scenarios.

Impact of Te and ne on edge current density profiles in ELM mitigated regimes on ASDEX Upgrade

ELM resolved edge current density profiles are reconstructed using the CLISTE equilibrium code. As input, highly spatially and temporally resolved edge electron temperature and density profiles are used in addition to data from the extensive set of external poloidal field measurements available at ASDEX Upgrade, flux loop difference measurements, and current measurements in the scrape-off layer. Both the local and flux surface averaged current density profiles are analysed for several ELM mitigation regimes. The focus throughout is on the impact of altered temperature and density profiles on the current density. In particular, many ELM mitigation regimes rely on operation at high density. Two reference plasmas with type-I ELMs are analysed, one with a deuterium gas puff and one without, in order to provide a reference for the behaviour in type-II ELMy regimes and high density ELM mitigation with external magnetic perturbations at ASDEX Upgrade. For type-II ELMs it is found that while a similar pedestal top pressure is sustained at the higher density, the temperature gradient decreases in the pedestal. This results in lower local and flux surface averaged current densities in these phases, which reduces the drive for the peeling mode. No significant differences between the current density measured in the type-I phase and ELM mitigated phase is seen when external perturbations are applied, though the pedestal top density was increased. Finally, ELMs during the nitrogen seeded phase of a high performance discharge are analysed and compared to ELMs in the reference phase. An increased pedestal pressure gradient, which is the source of confinement improvement in impurity seeded discharges, causes a local current density increase. However, the increased Zeff in the pedestal acts to reduce the flux surface averaged current density. This dichotomy, which is not observed in other mitigation regimes, could act to stabilize both the ballooning mode and the peeling mode at the same time.

The stabilizing effect of core pressure on the edge pedestal in MAST plasmas

The pedestal pressure measured in Mega Ampere Spherical Tokamak plasmas has been shown to increase as the global plasma pressure increases. By deliberately suppressing the transition into the high-confinement regime, the core plasma pressure was systematically altered at the time of the first edge localized mode. Stability analysis shows that the enhanced Shafranov shift at higher core pressure stabilizes the ballooning modes driven by the pedestal pressure gradient, consequently allowing the pedestal to reach higher pressures.

Pedestal and edge localized mode characteristics with different first wall materials and nitrogen seeding in ASDEX Upgrade

A comparison of ASDEX Upgrade (AUG) discharges performed with carbon and the full tungsten wall shows that the pedestal performance at low triangularity is not altered without gas puffing. The pedestal electron pressure is the same for both wall materials as is the confinement. With the tungsten wall the natural density is higher even without an additional gas puff. In typical operation with gas puffing the density is again higher in tungsten. This results in a higher collisionality with the tungsten wall. Pedestal pressure and plasma confinement, however, are not degraded until very large amounts of deuterium are puffed.The edge localized mode (ELM) crash in typical AUG discharges is observed to be composed of two independent phases. This is observed for both the carbon and the tungsten wall. The 1st phase of the crash is unaffected by scans of the plasma parameters as long as the pedestal pressure remains constant. The duration of the 2nd phase is strongly anti-correlated with the separatrix density and can be suppressed by the application of nitrogen seeding for divertor cooling. A consistent explanation for the two phases of the ELM crash does not seem possible when considering only the pre-ELM pedestal profiles. The scrape off layer (SOL) plasma provides the necessary free parameter for a consistent explanation, indicating the importance of the SOL in understanding the ELM crash evolution.