1. Introduction
Pulsed plasma thrusters (PPTs) are considered an attractive propulsion option for stationkeeping and drag makeup purposes for mass- and power-limited satellites that require micro-Newton second to milli-Newton second impulse bits [
1,
2,
3]. The low power requirements, simple design, robustness, and high specific impulse (>1000 s) are some of the key factors contributing to the application of PPT [
4]. With plenty of illustrations of the maturity of this technology, it has also been shown that PPT can be easily scaled down in power and size. A micro-PPT (μPPT), which is a miniature version of the traditional PPT, has been designed for delivery of very small impulse bits. The μPPT can deliver an impulse bit in the 10 μN·s range to provide attitude control and stationkeeping for microsatellites. In this thruster, the discharge across the propellant surface ablates a portion of the propellant, ionizes it, and then accelerates it, generating the thrust in a predominantly electromagnetic way [
1,
4,
5].
There is an increasing need for precision from the control of micro/nano satellites, which can be achieved by optimizing μPPT processes. Since experiments are cost-intensive, numerical methods are often used to simulate the working processes of the μPPT, so as to explore the effects of relevant parameters on improving performance [
6,
7,
8].
The modeling of μPPT has many aspects in common with the simulation of PPT, which has recently been extensively reported. Among these models, the electromechanical model is one of the most frequently used in the simulation of PPT. This model approximates the PPT system as an electromechanical device with an electric circuit interacting with a dynamic system. The electric circuit is theoretically idealized in both models as a Lenz-Capacitor-Resistance (LCR) circuit with discrete, albeit moveable, elements, as shown in
Figure 1. With these assumptions, the electromechanical model is relatively mature and has the advantages of small computation, short computing period, and simple modeling steps compared with the slug model [
9] and magnetohydrodynamic (MHD) model [
10]. Moreover, the simulation results were in good agreement with the experimental results, and the influences of parameters on the performance of μPPT can be investigated through this model. The electromechanical model, which has gained wide application in modeling, was first described by Hart in his fundamental 1962 study [
11]. Jahn [
9] developed a one-dimensional model that can predict the macroscopic properties of PPT with simple experimental results. However, it was inconsistent with the actual situation of increased ablation mass and could not reflect the ablation and plasma flow processes, as the ablation mass was assumed to be a constant value and the ablated working fluid was assumed to be directly attached rather than flowing to the current sheet during the discharge process. Subsequently, scholars have made many improvements to the electromechanical model. Vondra et al. [
12] developed a modified model that took the effect of aerodynamic force on the acceleration process of the working fluid into account and applied it to the optimized design of the Lincoln Experimental Satellite 6 (LES-6) PPT. Wei [
13] proposed a diffusion model that can reflect variations in the thickness of the current sheet with the discharge current. In these papers, the models were mainly modeled with empirical parameters or experimental results and could predict macroscopic performance rapidly, but they failed to accurately reflect the ablation process. Moreover, they assumed that the ablation mass as a constant value was no longer negligible in the simulation of μPPT due to the low magnitude in volume, power, and weight. It is therefore necessary to establish a theoretical model that contains the ablation process and relates the accumulation of the ablation mass to the mechanism process of impulse formation [
14,
15,
16].
With relatively low density, steady physical and chemical properties, low conductivity, and high specific heat, Teflon has become an ablation-resistant propellant, and is applied in PPT [
17,
18,
19] and other aerospace applications [
20]. There have been a number of investigations into the ablation process of Teflon, and some models have been presented to help study the ablation problem [
21]. To survey the transient ablation of Teflon, heat conduction and phase transition in intense radiation and convective environments were investigated by Arai [
22]. The mechanisms of surface recession and the evolution of the internal temperature and thickness of the gel layer were investigated with the transient one-dimensional two-layer ablation model of Teflon. Arai’s work took into account the optical transmittance of the amorphous and crystalline zones of Teflon, and studied the effect of the internal absorption of radiation in a numerical simulation. On the basis of Arai’s work, Gatsonis et al. [
23] modified the ablation model to simulate PPT operations and developed a volume fraction method to capture the moving interface between the semi-crystalline and gel phases; the simulation results were in good agreement with the experimental results. However, in situations that include extremely high temperature gradients, extremely large heat fluxes, and extremely short transient durations, the propagation speeds are finite and the mode of heat conduction is propagative and nondiffusive. As a result, the effect of non-Fourier heat conduction on the ablation process of Teflon cannot be neglected [
24]. In particular, the non-Fourier effect would be more obvious under the operation of μPPT due to the lower magnitude. But Gatsonis et al. did not take the non-Fourier effect into account. To analyze the physical process in the Teflon cavity of a coaxial PPT, Keidar [
25] also built an ablation model to calculate the electron temperature, electron and neutral densities, and ablation surface temperature with electron temperature and electron and neutral densities, considering the plasma energy balance, Joule heating, thermal conductivity, and mass balance. However, Keidar also paid little attention to the non-Fourier effect and some performance parameters could not be calculated in his paper. Additionally, there is little focus on the non-Fourier effect in similar attempts like those of Cho [
26], Huang [
17], and Tahara [
27]. It is therefore significant to take the non-Fourier effect into account in the ablation process.
This paper presents a numerical modeling method that combines a modified electromechanical model with a Teflon ablation model. In our method, a modified ablation model is established to calculate the ablation mass accumulation of work fluid, taking into account the reflectivity of material, heat conduction, phase transition, and non-Fourier effect. After this, the cumulative ablation mass is assumed to be an input parameter of the modified electromechanical model to simulate the operation of μPPT. Implementing the combined modeling method, the precision is improved while the calculation increases only very little. Meanwhile, by utilizing the model, not only are the macroscopic performance parameters predicted, but the specific ablation process and the temperature distribution of the working fluid can be given [
14,
28].
The remainder of the paper is organized as follows.
Section 2 presents the theoretical modeling process of the ablation model and modified electromechanical model.
Section 3 compares the numerical results with the experimental results in order to verify the effectiveness of the modeling method.
Section 4 explores the influences of different parameters on the performance of μPPT. Finally, some conclusive remarks are given in
Section 5.