Development of a FastPulsed Gas Valve for the PlanarPulsed Inductive Plasma Thruster

GuoDawei, ChengMousen and Li Xiaokang

College of Aerospace Science and Engineering, National University of Defense Technology, Changsha, 410073, China

Afast pulsed valve for the gas propellant injection in planar pulsed inductive plasma thruster has been developed. The valve is basedon the eddy currentrepulsion mechanism. The distinctive designof the valve is employ a truncated conicaldiaphragm made by elastic metal with high electrical conductivity and without ferromagnetism as the active part, which greatly contributes to the virtue of simplicity by avoiding an additional structure to provide the closingforce. An optical transmission technique is adopted to measure the opening and closing characters of the valve. The experimental results show that the delay time before the valve action is less than 60s,and the fully opening and closing movement time are less than 100s.Moreover, the throughput characteristicsarepredicted based on the valve opening times and the orifice size,which reveal that the throughput regulation could be accomplishedby adjusting the drive voltage.

Keywords:Fast pulsed gas valve, Pulsed inductive plasma thruster,Dynamic characteristic

*Corresponding author: Tel:+86-731-8700-1170, Fax: +86-731-8451-2301,

e-mail:

1.Introduction

The planar pulsed inductive plasma thrusters(PPIPT)are spacecraft propulsion devices in which electrical energy is capacitively stored and then quickly discharged through a planarcoil[1, 2], forming fast rising azimuthal current in the coil and inducing a plasma current sheet within the gas propellant near the face of the coil. The current in the plasma sheet and the current in the coil flow in opposite directions, providing a mutual repulsion that rapidly blows the plasma sheet away from the coil, which in turn entrains surrounding gases to high exhaust velocity ( (10km/s)) , thus causing useful impulsive thrust.

The formation and acceleration of the plasma current sheet are severely affected by the distribution of the gas propellant on the coil surface. In order to achieve high energy-efficiency, the entire gaspropellant puff injected by the feed valve must be uniform across and compressed to within the decoupling distance from the face of the coil at the same time that the thruster is fired[3, 4]. Efficient mass utilizationis also necessary for maximum thrust efficiency and Isp (specific impulse), which can only be achieved if the trailing edge of the gas pulse reaches the coil face before the leading edge has time toescape[5]. All the features mentioned above require that the valve as the key part of the propellant injection system can be fully opened in a time of a few hundred microseconds at most, and be completely closed in the same time[6]. Further, the valve is required to operate in a leak-free manner for many (>106) shots according the preconceived missions.

The fast pulsed gas valve have been widely used in controlled thermonuclear fusion areas where require a puff of gas to inject into a vacuum [7-14]. The actuator based on the eddy current repulsion mechanism is typically used. The closing force exerted by the closure assembly is usually a spring force[7-11]ora gas pressure differential[12-14].Because of the asymmetrical structure of the springand the closing force on the valve is not even, poor repetitiveness will be resulted.The valve using a gas pressure differential as the closing mechanism require anadditional plenum, what’s more, a sliding friction surface exist between the piston and the O-rings, which is directly related to the reliability and the performance.

In this paper we present the design and the performance of a fast pulsed gas valve for our planar pulsed inductive plasma thruster. The operation principle of the valve is also based on the eddy current repulsion mechanism.Adistinctive designimprovement in this valve is the use of a truncated conicalelasticdiaphragm instead of the conductive disks and the piston that have been used in previous valves. The valve design has certainpreferred characteristics, such as avoiding an additional closure assembly forthe closing force is the resultant force of the diaphragm deformation,elimination of slid friction, and reduction in size and weight. The test results show that the action delay time is not greater than 60 s, and the fully opening and closing movement time are less than 100 s. The throughput analysis results reveal that the valve has the capacity of throughput regulationby adjusting the valve excitation voltage.

2. Design and construction

A cross section view of the valve is shown in Fig. 1. The valve consists of the spiral coil, theannular truncated conical diaphragm, the back stop and the valve body. The spiral coil is made of 6 turns copper enameled wire with an inductance of 3.75 H, and embedded in the valve body by epoxy resin together with the seal plate which has an O-ring type seal surface. The metal diaphragm covers a gas plenum and seals against two O-rings underthe preload exerted by the diaphragm.

The operation cycle of the valve can be outlined as following. The valve is normally shut. When the valve is activated by discharging the capacitor via a thyristor, the pulsed current in the coil would generate a time varying magnetic field which induces eddy currents in the diaphragm. The Lorentz force resulting from the interaction of the magnetic field and the eddy currents, which much greater than the seal preload, will repel the diaphragm away from the seal surface of the seal plate, and the annular orifice isopen. The high pressure working gas in the plenum expands through the annular orifice. When the Lorentz forcedecays due to the attenuation of the current in the coil to the extent that is not enough to overcome the elasticrestoringforce, the diaphragm returns to its original position, wherein the valve completes an operation cycle.

Fig. 1. HERE

All the metal materials used in the valve are non-ferromagnetic, so the magnetic fields stemming from the valve coil and the thruster coil could not interfere with each other. Therefore, the valve can be used near the surface of the thruster coil. The diaphragm is made from beryllium copper of a high conductivity (3.0e7 S/m) and is placed as closeas possible to the coil to obtain greater magnetic force. The other metal parts of the valve are made from low conductivity non-ferromagnetic steels to reduce the eddy current loss and to improve the efficiency of the valve.The seal plate is made from an impact-resistant rubber,which assures the absorption of kinetic energybyinelastic collision when the diaphragm returns to its original position and canceling the reopening on the secondary bounce.

The electrical power system used to active the valve consists of a charging power supply and a 350F capacitor switched by a thyristor into the driver coil. This gives a half cycle current pulse of roughly 115s. The eddy currents induced in the diaphragm are the distribution currents which decay to 1/e from the surface to a depth. The depth is circuit frequency dependent and can be expressed as:

(1)

Where,is the magnetic permeability of the diaphragm which is approximate to , f is the frequency of the drive current, is the electrical conductivity of the diaphragm. For the given current pulse this yields a penetration of 1.38 mm for the diaphragm. For acquiringmore eddy currents and minimizing the diaphragm mass, the thickness of the diaphragm adopted in the valveis about 1.5 mm.

3. Test apparatus

An optical transmission technique is used to measure the opening and closing characteristics, which involves using a laser light beam and a photodetector. The laser light beam aligns with the gap between the outer edge of the diaphragm and the back stop and is irradiated onto the photodetector.The varieties of the photodetector output signalis in direct proportion to the decrease of the gap that is the separation of the diaphragm from the seal surface during the valve action. Because the gas pressure in the plenum is less than 1 atm, the force loaded on the diaphragm by gas is much smaller than the electromagnetic force and the elastic force of the diaphragm. Thus, the test is performed at atmospheric pressure and the gas inlet of the valve is open.The schematic of the test setup is shown in Fig. 2.

Fig. 2. HERE

The diameter of the laser light beam is a little larger than the maximum gap, which can make sure that there is no light on the photodetector when the valve is fully opened.The wave length of the laser light is about 650nm. The response time of the photodetectoris 1ns, which satisfies the measurement requirementsbecause the response time of the valve is in the order of microsecond. The trigger signal for thyristor is a square pulse generated by the function generator with a cycle much less than the coil current half cycle,which front edge is served as the oscilloscope trigger. The current in the excitation circuit is measured by a current transformer and the voltage of the capacitor is monitored by a high voltage probe. All the dataare collected and recorded using an oscilloscope.

4. Test results and discussion

The influence of the excitation voltage on the valve opening and closing characteristics is of much concern. The time evolution of the current in the coil energized by different capacitor charge voltages is shown in Fig. 3.Figure 4 shows the photodetector output at different charge voltage. It is clear that the photodetector signalsare smooth.It is believed that the magnitude of the photodetector signal is positively related to the lift off displacement of the diaphragm outer edge. It is seen from the photodetector signal that the valve can be opened when the excitation voltage is greater than 1100V, and there is a threshold voltage between 1300V and 1400V for the full opening of the valve.

Fig. 3. HERE

Fig. 4. HERE

The dynamic response characteristic (actuation delay time, the opening movement time andthe closing movement time) and the valve pulse characteristic(the maximum final lift off displacement and the duration of the opening state) are shown in Fig. 5. The maximum final lift off displacement is expressed in the form of dimensionless displacement(,x(t) is the instantdisplacement;xf is the full opening displacement, it is about 0.65mmand the duration of the valve opening state is expressed in the form of the full width at half maximum (FWHM) of the photodetector output signal. The FWHMdecreases with the increase of the charge voltage when the voltage is larger than the threshold value.

Fig. 5. HERE

The valve throughput measurement is not performed due to the limitations of the current test conditions, but the throughput can be roughly predicted based on the valve opening times and the orifice size. Assuming the temperature and the pressure of the gas in the plenum maintain constant,the throughput injected into the vacuum by the valve could be expressed as[15]:

(2)

Where, is the ratio of the specific heats; p and are the pressure and density of the gas in the plenum respectively; A(t) is the instant area of the valve orifice and is given by,r(36 mm) is the radius of the valve annular orifice. The gasmass per pulse versus the capacitor charge voltage for a unite atmospheric pressure are show in Fig. 6. There is a maximumvalue wherein the excitation voltage is 1400 V,which indicates that the valve could achieve throughput regulation by adjusting the excitation voltage.

Fig. 6. HERE

5. Conclusion

A fast pulsed gas valve based on inductive eddy current repulsion mechanismhas been developed and tested for the PPIPT.The valve employs a conical diaphragm as the active part, which greatly contributes to the virtue of simplicity. The test results show that the action delay time is not greater than 60 s, and the fully opening and closing movement time are less than 100 s. Moreover,the valve can achieve throughput regulation and valve pulse width adjustment by adjusting the valve excitation voltage.

Reference

[1]K. A. Polzin,J.Propul.Power.27, 513 (2011).

[2]K. Martin, A. Dominguez, R. H. Eskridge, K. A. Polzin and D. P. Riley,Joint Conference of the 30th International Symposium on Space Technology and Science / 34th International Electric Propulsion Conference and 6th Nano-satellite Symposium, Hyogo-Kobe, Japan, July 4-10, 2015.

[3]R. H.Lovberg and C. L. Dailey, AIAA Paper No. 89-2266,1989.

[4]D.Russell, J. H.Poylio, W.Goldstein, B. Jackson, R. H.Lovbergand C. L. Dailey, AIAA Paper No. 2004-6054, 2004.

[5]J. H. Poylio, D. Russell, W. Goldsteinand B. Jackson,AIAA Paper No. 2004-3640, 2004.

[6]R. H.LovbergandC. L. Dailey,AIAA Paper No. 91-3571, 1991.

[7]Y. Song, P. Coleman, B. H. Failor, A. Fisher, R. Ingermanson, J. S. Levine, H. Sze, E. Waisman, R. J.Commisso, T. Cochran, J. Davis, B. Moosman, A. L. Velikovich, B. V. Weber, D. Bell, and R. Schneider,Rev. Sci. Instrum. 71, 3080 (2000).

[8]Mahadevan Krishnan, C. G. R. Geddes, R. A. van Mourik, and W. P. Leemans, H. Murphy and M. Clover. Kristi Wilson Elliott ,Phys. Rev. STAccel. Beams 14, 033502 (2011).

[9]Mahadevan Krishnan, Kristi Wilson Elliott, Robert E. Madden, P. L. Coleman, John R. Thompson, Alex Bixler and et al, Rev. Sci. Instrum. 84, 063504 (2013).

[10]P. W.L. de Grouchy, E. Rosenberg, N. Qi, B. R. Kusse, E. Kroupp, A. Fisher, Y. Maron, and D. A. Hammer, AIP Conference Proceedings 1639, 43 (2014).

[11]H. D. Zhuang and X. D. Zhang, Rev. Sci. Instrum.86, 053502 (2015).

[12]A. Savtchkov, K. H. Finken, and G. Mank, Rev. Sci. Instrum. 73, 3490 (2002).

[13]S. A. Bozhenkov, K.-H. Finken, M. Lehnen, and R. C. Wolf, Rev. Sci. Instrum. 78, 033503 (2007).

[14]R. Raman, T. R. Jarboe, B. A. Nelson, S. P. Gerhardt, W.-S. Lay, and G. J. Plunkett, Rev. Sci. Instrum.85, 11E801 (2014).

[15]B. Novak and S. Pekarek, Rev. Sci.Instrum.41, 369 (1970).

Fig. 1.Cross sectional view of the valve

Fig. 2. Schematic of test setup for measuring valve response time

Fig.3. Coil current traces at various charge voltages

Fig. 4. Photodetector signals vs time at different capacitor charge voltages

Fig. 5. (a) Dynamic response characteristic and (b) valve pulse characteristics

Fig. 6. The gas mass per pulse versus the capacitor charge voltage for a unite atmospheric pressure.

Fig.1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.