An internal kink instability is observed to grow and saturate in a line-tied screw pinch plasma. Detailed measurements show that an ideal, line-tied kink mode begins growing when the safety factor q = (4pi2r2B(z))/(mu0I(p)(r)L) drops below 1 inside the plasma; the saturated state corresponds to a rotating helical equilibrium. In addition to the ideal mode, reconnection events are observed to periodically flatten the current profile and change the magnetic topology.

Initial results from the Madison Dynamo Experiment provide details of the inductive response of a turbulent flow of liquid sodium to an applied magnetic field. The magnetic field structure is reconstructed from both internal and external measurements. A mean toroidal magnetic field is induced by the flow when an axial field is applied, thereby demonstrating the omega effect. Poloidal magnetic flux is expelled from the fluid by the poloidal flow. Small-scale magnetic field structures are generated by turbulence in the flow. The resulting magnetic power spectrum exhibits a powerlaw scaling consistent with the equipartition of the magnetic field with a turbulent velocity field. The magnetic power spectrum has an apparent knee at the resistive dissipation scale. Large-scale eddies in the flow cause significant changes to the instantaneous flow profile resulting in intermittent bursts of non-axisymmetric magnetic fields, demonstrating that the transition to a dynamo is not smooth for a turbulent flow.

An axisymmetric magnetic field is applied to a spherical, turbulent flow of liquid sodium. An induced magnetic dipole moment is measured which cannot be generated by the interaction of the axisymmetric mean flow with the applied field, indicating the presence of a turbulent electromotive force. It is shown that the induced dipole moment should vanish for any axisymmetric laminar flow. Also observed is the production of toroidal magnetic field from applied poloidal magnetic field (the ω-effect). Its potential role in the production of the induced dipole is discussed. Many stars and planets generate their own nearlyaxisymmetric magnetic fields. Understanding the mechanism by which these fields are generated is a problem of fundamental importance to astrophysics. These dynamos are sometimes modeled using two components: a process which generates toroidal magnetic field from poloidal field and a feedback mechanism which reinforces the poloidal field [1]. The first process is easily modeled in an axisymmetric system: toroidal differential rotation of a highly-conducting fluid sweeps the pre-existing poloidal field in the toroidal direction creating toroidal field. This phenomenon, known as the ω-effect, is efficient at producing magnetic field and has been observed experimentally [2,3,4]. The second ingredient to the model is more subtle, as toroidal currents must be generated to reinforce the original axisymmetric poloidal field. Cowling's theorem [5] excludes the possibility of an axisymmetric flow generating such currents so some symmetry-breaking mechanism is required.The usual mechanism invoked [6] is a turbulent electromotive force (EMF), E = ṽ ×b , whereby small scale fluctuations in the velocity and magnetic fields break the symmetry and interact coherently to generate the large scale magnetic field. This EMF is sometimes expanded [7] in terms of transport coefficients about the mean magnetic field: E = αB + β∇ × B + γ × B; α is characterized by helicity in the turbulence, β by enhanced diffusion and γ by a gradient in the intensity of the turbulence. α is of particular interest as it results in current flowing parallel to a magnetic field, and when coupled with the ω-effect can generate the toroidal currents needed to reinforce the poloidal field. Experimental evidence for mean-field EMFs (such as the α-effect) in turbulent flows has been scarce. Three experiments, relying on a laminar α-effect, have generated an EMF [8] and dynamo action [9,10], but heavilyconstrained flow geometries were used to produce the needed helicity; the role of turbulence was ambiguous. Experiments with unconstrained flows have provided evidence for turbulent EMFs, though not the turbulent α- effect. Reighard and Brown [11] have attributed a measured reduction in the conductivity of a turbulent flow of sodium to the β-effect. Pétrélis et al. have observed [12] distortion of a magnetic field similar to an α-effect (currents generated in the direction of an applied magnetic field) and postulate that turbulence may be responsible for...

The magnetic field measured in the Madison Dynamo Experiment shows intermittent periods of growth when an axial magnetic field is applied. The geometry of the intermittent field is consistent with the fastest growing magnetic eigenmode predicted by kinematic dynamo theory using a laminar model of the mean flow. Though the eigenmodes of the mean flow are decaying, it is postulated that turbulent fluctuations of the velocity field change the flow geometry such that the eigenmode growth rate is temporarily positive. Therefore, it is expected that a characteristic of the onset of a turbulent dynamo is magnetic intermittency. Determining the onset conditions for magnetic field growth in magnetohydrodynamics is fundamental to understanding how astrophysical dynamos such as the Earth, the Sun, and the galaxy self-generate magnetic fields. These onset conditions are now being studied in laboratory experiments using flows of liquid sodium [1]. The conditions required for generating a dynamo can be determined by solving the magnetic induction equationwhere B is the magnetic field, V is the velocity field scaled by a characteristic speed V 0 , and the time is scaled to the resistive diffusion time τ σ = µ 0 σL 2 . The magnetic Reynolds number is Rm = µ 0 σLV 0 , where σ is the conductivity of the fluid and L is a characteristic scale length [2]. In the kinematic approximation, the velocity field is assumed to be a prescribed flow (either the flow is stationary or its time dependence is specified) and Lorentz forces due to the magnetic field are neglected. Equation 1 is then linear in B and can be solved as an eigenvalue problem. Several different types of stationary, helical flows have been shown theoretically to be kinematic dynamos [3,4,5,6], which in turn has lead to the design of current dynamo experiments. For particular flows, the kinematic model predicts a critical value of the magnetic Reynolds number, Rm crit , above which the magnetic field becomes linearly unstable, i.e. for Rm > Rm crit a small seed magnetic field will grow exponentially in time. The dynamo onset conditions have been tested in helical pipe-flow experiments at Riga [7,8] and Karlsruhe [9,10]. Both experiments generated magnetic fields at a value of Rm crit consistent with predictions from the kinematic theory.Fluids and plasmas such as the Earth's liquid core, the solar convection zone, and liquid metals are turbulent under the conditions required for a dynamo. The

The nature of Ohm's law is examined in a turbulent flow of liquid sodium. A magnetic field is applied to the flowing sodium, and the resulting magnetic field is measured. The mean velocity field of the sodium is also measured in an identical-scale water model of the experiment. These two fields are used to determine the terms in Ohm's law, indicating the presence of currents driven by a turbulent electromotive force. These currents result in a diamagnetic effect, generating magnetic field in opposition to the dominant fields of the experiment. The magnitude of the fluctuation-driven magnetic field is comparable to that of the field induced by the sodium's mean flow.