Study of a Battery Driven Electrohydrodynamic Thruster

Study of a Battery Driven Electrohydrodynamic Thruster

Kajsa Hjort and Elias Hachichou

Abstract—Electrohydrodynamic (EHD) thrusters hold promise to provide more efficient thrust than propeller driven systems for small drones. However, the fact is that there is incomplete analysis in the area and that no work has yet studied the capacities of battery driven EHD thrusters. This study shows how a battery system from a commercial Arc-Lighter can produce a suitable voltage for an EHD thruster, with 7.1 kV. Using the battery system to drive a single EHD thruster, the EHD thruster could only lift 13.3 ppm of the total weight of a battery system and EHD thruster. However, the low value is due to a low-quality thruster. If the battery system would be used with suitable high- quality EHD thrusters from literature, approximately 5% of the total weight could be reached when at least 20% is needed for airplane drones to lift. In conclusion, the battery system is not sufficient for lifting the total weight. Still, in future, with more work on both EHD thruster and a battery system providing a higher voltage, there might be possible to have small airplane drones with battery driven EHD thrusters.

I. INTRODUCTION

D

The different fields have made drones a good research topic for improvements. One of the problems with drones is their power source, nowadays most often propellers. The blades make them both louder and less efficiency than they could be. One alter- native power source for drones could be electrohydrodynamic (EHD) thrusters. Compared to propellers, the EHD thrusters is quieter and have theoretically better efficiency. [1]

The EHD thruster ionizes the air to create a wind, instead of using moving parts. [1] The lack of moving parts could make the EHD thruster more reliable, easier to produce and possibly give a longer life span due to less tear on the parts. However, EHD thruster has still problems, which are the reason there are no ion-drones on the market. One of the big problems is the current electrical system delivering the high voltage needed to power the thruster, which is not portable or easy to implement into the drones in a commercial way [1]

The concept of EHD was developed in the early 60’s when the EHD thruster got introduced by Alexander de Seversky.

[2]EHD is used in air filtration, cooling, jet printing, EHD- speakers, propulsion and other situations where an air flow is needed and ordinary methods, such as fans, are not suitable.

The EHD thruster has also been implemented in deep space vessels by NASA for several missions in which they bring the propellant with them. [3] The advantage with ion thrusters in space, is the high efficiency they have compared to traditional rocket propulsion. Even though they run at high voltages they can keep the power consumption low because of the very low currents.

This bachelor thesis focuses on the feasibility to implement a battery driven system to an EHD thruster, to explore if the battery driven EHD thruster could produce enough force to be able to lift itself.

II. THEORY

The theory for understanding the steps and components to make the battery system for an EHD thruster is presented in this section. In addition to what is presented, it is good to read on transformers and AC-to-DC-converters. (Since this thesis is a Bachelor’s Thesis, it is advised that AC-to-DC-converters and transformers should not be included.) The theory is needed to better understand the challenges in powering an EHD thruster.

A. EHD Thruster

The EHD thruster is first introduced to give an understand- ing of the thrusters’ components and important parts, why the high voltage is needed and why an EHD thruster could be a suitable alternative to propellers.

EHD thrusters consist of two electrodes, with a distance L between them (figure 1a). [1] One of the electrodes acts as a cathode and the other as an anode. The anode is where the ions, which come from the cathode is collected. The ions are attracted to the anode because of the Coulomb’s force.

The attraction happens in the so-called ’drift layer’. [4] The attraction will create a wind because the ions clashes with neutral molecules (figure 1b), which then create a new electron and positive charged air molecule and the collision starts again.

The phenomenon is called electron avalanche effect and will create the thrust, which is needed to lift the EHD thruster.

Figure 1. a) The EHD thruster with a needle as cathode and a mesh as anode.

b) An enlargement of the drift region, where free electrons collide with the air and create new free electrons

The cathode side is made of a defined number of sharp emitters with a distance to the anode, L, and an EHD thrusters

diameter, D, (figure 1). The sharp emitters are used because of the electrical field, E is larger at a sharp emitter than at a curvature. They should be spaced equally over the whole cathode to make the best airflow. Therefore, the air can be ionized at lower voltages. [1], [5] For an Example of how a cylinder-shaped thruster may look, see the prototypes (figure 7).

The wires from the EHD thruster are connected to a power supply, which delivers a defined voltage. The lowest voltage used for EHD thrusters is around 3 kV because at 3 kV the air is ionized, creating a Corona discharge. [6]

The longer L is, the higher voltage is needed to give the optimal force. The optimal force will be larger if the voltage is increased and therefore the length is important. If the length is too short and the voltage is too high, an arc discharge bridge will appear which will short circuit the EHD thruster.

1) Corona Discharge: To understand the optimal structure of an EHD thruster, the corona discharge needs to be known.

If the voltage reaches a value higher than the Peek’s Corona Inception Voltage, CIV, a corona is generated. The air starts to be ionized at the CIV voltage. The corona occurs because the E-field in the system is shifting. [7] To decide the CIV, Equation 1 can be used. This Equation is specified for air. The value depends on the distance between anode and cathode, L, and the needle radius, rw. [6]

CIV = 3 · 10⁶· (1 +0.0301

√rw

) · r_w· lnL rw

) (1)

For Example, if the radius is 0.1 mm and the air gap is 30 mm, the CIV is 6.86 kV. With a voltage of little over 6.86 kV, there will be a corona discharge.

The created corona discharge is especially notable by the conductors’ surface. The corona discharge is strongest at this point because the conductor’s area has the highest dielectric flux density. The flux density will increase when the voltage is increased beyond the critical value. If the voltage becomes too high an arc discharge will happen. The arc discharge appears because of the ionization layer. The arc discharge will short- circuit the system and no thrust is produced. However, if the value is exactly below the critical value before the arc discharge, the maximum air flow is produced. [6], [8]

2) EHD Thruster Design: The thruster consists of two major parts, the cathode and the anode. The cathode radius, r_w, should optimally be small to be able to create a corona, see Equation 1. The design of the anode is built on two things: the collection area, and the airflow capability. A larger surface leads to a higher collection of ions, resulting in a more efficient airflow. But a larger surface area, depending on the construction, can block the airflow or make the construction inefficient and heavy. The vertical sheet has the advantage that the sheet does not block the airflow. The other type of cathode is a horizontal sheet of metal. The horizontal collector gives an even spread of ions giving a better” pull” of the surrounding air. To allow airflow, the sheet has several holes.

A more convenient solution is to use a mesh instead of a sheet.

The mesh type of collector has been proven in earlier work to be the best option. [9]

Another important thing to consider when constructing EHD thrusters is to have a constant E-field in the thruster. If there are sharp corners, there will be a higher electrical field at these corners. The corners can cause an arc discharge and a short circuit if the voltage is too high. The corners can also cancel parts of the thrust. Because of these problems, an optimal figure is a circle for both the anode and the cathode.

However even though the circle structure is important, the most important part is to use a direct current power supply.

[1]

B. Electric system

The battery system used in the report is a system which includes an AC-to-DC-transformer, micro-controller and a bipolar transistor. The battery system is a circuit which is specifically made to create an arc. The specialized circuit means that the system needs to be modified to work with EHD applications. One problem is that the output is something closer to AC and not DC. To solve the problem, a DC-to- AC-converter can be used. The converter makes the output to AC, which is not favorable for the EHD thruster. One way to convert AC-to-DC is the use of a rectifier.

1) Rectifier: The most common use of a diode is to be part of a rectifier. A rectifier converts AC to DC and the rectifier works because the diode only allows electrons to flow in one direction. The simplest rectifier consists of only one diode is called a half bridge and a rectifier with four diodes (figure 2) is called a full bridge. A full wave bridge must be used if both of the half cycles should contribute. Because if not, the filter will either take the negative or positive part only, erasing the other part. [10]

+

-

Figure 2. Full-bridge rectifier that works as an AC-to-DC-converter.

With the help of the circuit, the signal will go from full wave to just positive or negative part.

A smoothing capacitor can be used to get a signal closer to DC. By putting a smoothing capacitor after the rectifier, the smoothing capacitor minimizes the ripple and creates a signal with more DC characterizations. The smoothing capacitor gives a higher voltage because the signal is more stable because of less ripple. The capacitor gives a better signal overall if DC is wanted. To calculate the size of the capacitor, the following Equation can be used [11]:

τ = 1

f ' R_load· C (2)

In the Equation, C is the capacitance, I is the current, f is the frequency, R_load is the resistance of the load, and tau, τ , is the time constant.

C. Thrust

To determine if the thruster can lift itself the thrust is needed. To lift the EHD thruster straight up, as in a quadcopter or a helicopter, the thrust (T) needs to be greater than the gravity (T ≥ mg). To calculate the thrust for the thruster, the following Equation can be used, if the pressure before and after is simplified to the same value:

∂m

∂t = Density · V elocity · Area (3) F = (∂m

∂t )1· V1− (∂m

∂t )0· V0 (4) V1is the outlet velocity and V0is the velocity in. ^∂m_∂t is the mass flow rate. [12] If the experiment is conducted in a space without any wind the air velocity is 0 m/s in the beginning. For Example, if the mass is 2 g and the collector area is 7.07 cm², the thruster need to create 4.7 m/s to lift.

D. Efficiency

Knowing the efficiency of the system is important to be able to determine if there is more power to get out and how much energy is lost. To calculate the efficiency the input power and output power should be known.

1) Electric Power: The input power for the EHD thruster can be calculated with the current and voltage from the battery system using this Equation:

Pelectric= U · I (5)

In the Equation, U is the voltage and I is the current.

2) Mechanical Power: The output power for the EHD thruster can be calculated with the wind speed from the thruster using the following Equation:

Pmechanic=1

2Aρv³ (6)

where A is the outlet area of the thruster, ρ is the air density and v is the wind speed.

3) Efficiency: Efficiency is then calculated by getting the ratio of the input and output power.

η =Pmechanic

Pelectric

(7)

III. EXPERIMENTAL

The purpose of the study was to test if a battery system could produce enough power to drive an EHD thruster i.e.

that the EHD thruster could lift itself and the battery system.

The design for the electrical system consisted of an electrical lighter, bought from eBay (figure 3). The Arc-Lighter uses the arc created from the high voltage as a heat source, to be able to light various things such as cigarettes.

Figure 3. Arc-Lighter from eBay, from where the battery system is extracted.

The reason why an Arc-Lighter was chosen was because the battery system had to be a small electrical system that can deliver high enough voltage to be able to create a Corona Discharge. The implemented components made it easy to go to the other problem, instead of making the circuit from scratch (figure 4). The Arc-Lighter consists of a DC-to-AC- converter with a transformer to amplify the voltage. In this Arc-Lighter a double set of converters and transformers was present (figure 4), since the manufacturer wanted to create a cross pattern with two arcs (figure 3).

Figure 4. There are two sets of DC-to-AC-converters as well as transformers and one charge controller and a timer/trigger.

A rectifier was designed to modify the battery system, converting the high-voltage AC signal to resemble a DC signal. The Rload, the EHD thruster resistance, was calculated and found by measuring the highest frequency of the battery system’s signal. Since the voltage was constantly higher than 1 kV, to protect the equipment either a home-built voltage divider or a commercially available high voltage probe [13]

were used before the signal was introduced to the measurement tools. The high voltage probe is basically a voltage divider with a division factor of 1000. The high voltage probe was of a much higher quality than the voltage divider made in the lab and the probe gave a better value for the voltage from the lighter. However, the high voltage probe could not handle

a signal with AC characterizes, which is the reason for also using the voltage divider in the tests without the rectifier.

To find the load, node analysis was used on one of the prototypes. By knowing the current through and the voltage over the resistance of the EHD thruster, Rload, could be calculated (figure 5). Measurement of the current and voltage before and after the thruster was made with a Fluke 87 multimeter [14], The voltage source is a high voltage power supply. The Rload included the resistance in the air between the electrode surfaces and the resistances in the wire.

Lighter+Rectifier

A RLoad

V

Figure 5. Measuring circuit for current and voltage on the battery system and the thruster.

The frequency of the battery system was measured by a Tektronix TDS 340 oscilloscope [15] Equation 2 provided then the value of the capacitor.

To find the optimal thrust for an EHD thruster powered by a battery system, five unique EHD thrusters were made. All designs of the thrusters were drawn in SolidWorks [16] and 3D printed in a Stratasys uPrint SE 3D-printer [17]. All EHD thrusters consisted of a 2.5 cm long cylinder with a 15 mm radius (figure 7a, b, c). Each of the thrusters had a unique design of the collector. However, their tops looked the same.

The top consisted of two beams holding a hollowed circle in the center for the wires and had three holes for each of the emitters, the needles (figure 7a, b, c). Needles were used instead of conventional metal wire because they are easier to implement in the construction and the current was easier to control. With wires, there are difficulties with the length of the wire, not knowing if the wire is the same between models.

These needles were eventually spread throughout the collector.

The needles were from American Probe and Technologies, Inc and made of nickel-plated tungsten. The radius, rw, of the needles was 0.6 µm, except for a thicker needle with a radius of 1 µm. A glue gun was used to glue the needles in place and to isolate for safety reasons. The glue was also used to secure the collector to the bottom and isolate the collector’s parts, which extruded from the EHD thruster. To connect the collector and the emitter, high voltage wires were soldered to the collector and emitter for each of the prototypes. To make the transition, from EHD thruster to high voltage wires safe, heat shrink tubes were used.

To find the thrust for the five EHD thrusters (Table II), an experiment was conducted inside a sealed chamber. The thrusters were connected to the rectifier, the battery system and to the flow meter. The flow meter was connected to a 3D-printed funnel to connect better to the EHD thruster (figure 7d). the funnel was needed since the flow meter’s tube and EHD thrusters had different diameters.

The flow meter gave an output voltage that could be converted to Standard Litre Per Minute, SLPM, from a curve

given in the data sheet, given by a researcher in the laboratory [13]. From the SLPM the trust was calculated and the thrust value was used to calculate the thruster’s thrust to weight ratio (If the ratio would reach 1, the EHD thruster could lift itself).

The schematic experimental setup and a picture of the actual setup are shown in figure 6a and 6b, respectively. The current was also measured and the power output of the EHD thruster was calculated.

(a)

(b)

Figure 6. (a) Schematic drawing of the setup that connected to a flow meter measures wind speed for an EHD thruster.(b) Setup of the experiment in real life, measuring the thrust for an EHD thruster.

In the setup, five prototypes with different collectors or nee- dles were measured (Table II). Three different collectors were used to see how the battery system could handle the different thrusters. The following thrusters were tested; one prototype with rough mesh (figure 7c), one fine mesh (figure 7c) and one vertical collector (figure 7b) to see the difference of thrust for the battery system with different collectors. Both the thruster with fine and the one with rough mesh used the same Design (figure 7c).

The mesh was made of stainless steel and had different density. The density of the fine mesh was 0.48 kg/m² and that of the rough mesh was 1.57 kg/m². For the prototypes using a vertical collector, aluminum was used as the material because aluminum is light weighted and easily forms different structures.

To have a lighter structure, an updated version of the fine mesh was made with two pillars as support instead of a full cylinder, referred to as “open system” (figure7a). If the open and closed systems had similar thrust, the open system should be used since its lower weight would give a higher thrust to weight ratio. Two prototypes were tested with the same collector and structure but with different sized needles, to see if the radius of the needle impacted on the thrust. The size of the needles was 1 µm for the thicker one, and 0.6 µm for the thinner one, as used in the previous prototypes ( figure 7). All the prototypes were weighted to be used for the calculation of the trust and weight ratio. Enlarged versions of all prototype’s figures are available in the appendix.

(a) (b)

(c) (d)

Figure 7. Different prototypes of thrusters (a) Two pillars for support structure, i.e. open system, EHD Thruster. (b) Full cylindrical structure, i.e.

closed system, with a Vertical collector. (c) Closed system with horizontal collector, (d) Funnel to connect the flow meter to the EHD thruster to decrease losses

In the last test conducted, several thrusters were connected to the battery system and the rectifier, to put a greater load on the system. The thrusters were all the same type, fine mesh with a thin needle (0.6 µm). However, one was an open system and the other were closed systems. The thrusters were connected in parallel to the battery system and the wind speed was measured on one of the thrusters. The total current was also measured, the setup of the experiment is in figure 8.

Figure 8. Test set-up for evaluating how much load the battery system could handle. The battery system was connected to a rectifier which was then connected to 1, 2 or 3 parallel thrusters.

IV. MEASUREMENTS

A. Battery system

When the lighter is connected to the oscilloscope through a voltage divider, a pulsing signal is observed. The highest pulses had a frequency of 15.2 kHz and a positive pulse of 10 V and a negative pulse of 25 V (figure 9). Since the exact voltage division is not known, the figure only shows the relative variations of the voltage.

A 10 pF capacitor capable of handling 10 kV is used [18], as the closest available smoothing capacitor to the ideal 12 pF.

Figure 9. Measurement of pulse signals from the lighter connected to an oscilloscope through a voltage divider. The highest pulses had a positive pulse of 10 V and a negative pulse of 25 V with a frequency of 15.2 kHz.

The output from the lighter connected to a rectifier with a smoothing capacitor and a voltage divider and is -50 V, (figure 10), once again the figure only provides relative values.

The ripple and pulses are smaller and the signal showed more DC characteristics than without the rectifier (figure 9).

Figure 10. Measurement of pulse signals with an oscilloscope from the lighter connected to a rectifier with a smoothing capacitor and a voltage divider. The pulse is -50 V.

The output is 7.1 kV from the lighter when the lighter is connected to a rectifier with a smoothing capacitor. The value is measured with a new and fully charged battery. The voltage decreased when a used battery with reduced charge is connected.

B. Wind Speed and Thrust

From the thrust experiment, the results are that the thrust from the open system with a fine mesh and thinner emitters gave the best thrust of 2.20 µN, and the vertical collector with the closed system showed the lowest value of 1.86 µN (Table I, figure 7a and b).

From the thrust value, Table I, the thrust/weight ratio could be obtained. The open structure with the fine mesh and needle,

TABLE I

THRUSTERSTHRUST VERSUS WEIGHT

Thruster Wind speed m/s Thrust (µN)

Closed¹rough mesh 0.0489 2.08

Closed¹vertical collector 0.0464 1.86

Closed¹fine mesh 0.0504 2.20

Open²fine mesh³ 0.0496 2.13

Open²fine mesh⁴ 0.0504 2.20

1Closed system refers to the structure of the cylinder. Closed means the cylinder is a full cylinder

2Closed system refers to the structure of the cylinder. Closed means the cylinder is a full cylinder

3Pillar structure EHD thruster with a thicker needle, radius of 1 µm.

4Pillar structure EHD thruster with the normal needle, radius of 0.6 µm.

0.6 µm, gave a ratio of 13.3 ppm when the weight for the entire system is used (weight of battery system and thruster with glue combined). In comparison, a ratio of 82.4 ppm is obtained when only relating to the weight of the thruster (Table VI). The prototype with the least thrust/weight ratio is the vertical collector that showed 36.1 ppm for the thruster and 9.78 ppm for the entire system including the EHD thruster.

TABLE II

THRUSTERSTHRUST VERSUS WEIGHT

Thruster Thrust/Weight

for Thruster ¹

Thrust/Weight for whole system²

Rough Mesh 0.0000428 0.000011061

Vertical Collector 0.0000361 0.00000978 Closed fine mesh 0.0000468 0.0000118 Open fine mesh³ 0,0000800 0.0000129 Open fine mesh⁴ 0.0000824 0.0000133

1Weight of EHD thruster, including glue, collector and emitters (14.19 g plus EHD thruster)

2Weight of the whole system, including the EHD thruster and the battery system (14.19 g).

3Pillar structure EHD thruster with a thicker needle, radius of 1 µm

4Pillar structure EHD thruster with the normal needle, radius of 0.6 µm.

When several thrusters are connected the total current increased from 116 µA for a single thruster, to 180 µA for two thrusters and 205 µA for three thrusters, Table III. When three thrusters are connected the battery system started to overheat.

The wind speed did not change as much as the current did.

A single thruster showed a wind speed of 0.0528 m/s, and for two and three thrusters in parallel, the value is 0.0538 m/s and 0.0481 m/s, respectively.

TABLE III

PARALLEL THRUSTERS PERFORMANCE Thrusters IOut(µA) Wind speed(m/s)

1 116 0.0528

2 180 0.0538

3 205 0.0481

The efficiency of the thrusters is very low for all of them, around 10⁻⁸, and the highest power draw from the battery system is 0.909 W (Table IV)

TABLE IV

THE FINE MESH THURSTER’SPOWER

Thruster Current (µA) Power⁴(W) Power⁵(nW) Efficiency

Closed¹ 124 0.880 55.3 6.28E-08

Open² 111 0.781 52.8 6.77E-08

Open³ 130 0.909 55.3 6.09E-08

1Closed system refers to the structure of the cylinder. Closed means the cylinder is a full cylinder

2Pillar structure EHD thruster with a thicker needle, radius of 1 µm.

3Pillar structure EHD thruster with the normal needle, radius of 0.6 µm.

4Electric Input Power.

5Mechanical output Power.

V. DISCUSSION

The highest thrust to weight ratio measured is 13 ppm (when using the weight for the whole system). A lower value than first anticipated and 13 ppm is also low to be able to lift only the thruster. A ratio of 1 is needed to be able to lift the thruster without additional support. Airplanes have with their wings as lifting support at high enough wind speed. Therefore, they need only a thrust to weight ratio of around 0.2.

The main reason the thrust is so low seems to be the thruster design since other reports have shown higher thrust from similar types of EHD thrusters.

The data from Table I shows that there is no significant difference in thrust between a closed system and an open system, which is why the open design is used in the latter prototypes.

According to the result, it is not possible for the EHD thrusters including the battery system to lift external help.

as indicated in Table II there is no thrust/weight ratios that are even close to 1, which need for being able to lift itself.

However, if one of the EHD thruster models from Moreau and Touchard [9] is used, the battery system in this report should be able to lift the EHD thruster when using an airplane drone (Table VI), Example 1.

In Example 1 when the EHD thruster is provided with 14.2 kV, which yields the highest wind speed and therefore also the highest thrust, of 10 m/s and 86.86 mN respectively.

The lowest values of wind speed and thrust of the Examples from other articles, [9] is the one with the lowest voltage giving 1.95 m/s in wind speed and a thrust of 3.29 mN (Example 4, Table V)

TABLE V

THRUSTERS WIND SPEED,THRUST AND OUTPUT POWER FROM ANOTHER RESEARCH PUBLICATION

Thruster Wind speed (m/s) Thrust (mN) Power²(W)

Example 1 10 86.6 0.039

Example 2 5 21.6 0.0096

Example 3 4.55 17.9 0.0080

Example 4 1.95 3.29 0.00015

1Values are taken from figure 4b and 2b in [9] and therefore not exact numbers

2P˙electric, Current taken from figure 4b and voltage from 2b in [9]

In the article [9], which focused on designing the thruster, a wind speed of 4.55 m/s is reached at around 7 kV showing that the best thruster in this report is relatively inefficient (Table V,

Example 3). The thrusters in the article [9] do also have a lower power draw even though they run at higher voltages.

The lower power draw shows that the main reason for the relatively low air flow achieved in this report is due to bad thruster design rather than the battery system itself.

TABLE VI

THE THRUST-TO-WEIGHT RATIO OF THE THRUSTER FROM ANOTHER RESEARCH PUBLICATION.

Thruster Thrust/Weight for Thruster

1 Thrust/Weight for whole system

2

Example 1 2.035 0.244

Example 2 0.509 0.0609

Example 3 0.421 0.0504

Example 4 0.077 0.00926

1Weight of EHD thruster, including glue, collector and emitters

2Weight of the whole system, including the EHD thruster and the battery system (14.19 g).

In the article by, Moreau and Touchard there is no weight given, so the weight is approximated using the measurements given and the weights from the thrusters in this report. [9] The mesh would weight around 0.15 g and the cylinder around 1.78 g. That would make a total weight of the thruster of 1.93 g. As mentioned before, the Example 1 thruster (Table VI) can lift the EHD thruster and the whole system as an airplane drone. For Example 3, the Example with voltage as the battery systems output, the trust weight ratio for the EHD thrust is 0.421 (Table [9]). The EHD thruster’s ratio of thrust/weight could with the battery system, produced in this report, lift itself as an airplane drone. However, the EHD thruster could not lift itself and the entire system because the ratio of thrust/weight is too low (0.0504).

The highest power from the article by Moreau and Touchard is 0.244 W and compared to this report’s (0.909 W), i.e. their output power is much smaller. The low output power, for the optimal EHD thruster, means that if modifications to the system would be done, such as increasing the voltage more than 7.1 kV. The battery system might be able to provide enough power for the thruster in Example 1. [9] If so, there would be possibilities that the EHD thruster could lift itself and the battery system as an airplane drone

The experiment with the thrusters connected in parallel showed that the battery system can handle a relatively large power draw. The thrusters used in this study have already a large power draw, compared to other reports, which shows that if used the article’s thruster many more could have been able to be connect together in parallel. Important to note is that the battery system started to overheat when three thrusters are connected, probably due to the higher current.

One of the problems with the experiment is that they run consecutively, without charging the battery in between.

The uncharged battery could have lead to inaccuracy when measuring, because of the lower voltage from a slightly less charged battery can give.

The signal of the battery system after the rectifier had different characteristics than a typical high voltage power supply’s output signal. Therefore, this report made its own research on an optimal EHD. The results are the same results

as in Moreau and Touchard [9] publication, as expected.

However, the thrust at the same voltage is lower for this report’s EHD thruster than the article’s EHD thruster. An inaccuracy could be the needles, which is an old needle from another study. The fact that these old needles are bent should affect results negatively since the effective radius will be larger and may be much larger.

VI. FUTUREWORK

There are three big concerns which need to be addressed in future work: sustainability, higher voltage and lighter weight.

First, the battery system needs to be sustainable, work in all type of weathers and all type of conditions. The battery needs to have a longer lifespan, so the battery system can be used more broadly and effectively.

As mention before, there are Examples of the EHD thruster that can deliver 10 m/s flow with 14.2 kV. However, the battery system delivered just slightly above 7.1 kV. To make the EHD lift both the EHD and the EHD thruster’s battery system, the battery system needs to deliver a higher voltage.

The voltage could be increased by using a transformer that provides a higher voltage than the one in the battery system.

The transformer would give a higher voltage which could sustain the optimal EHD thruster in the article with an EHD thruster with a flow of 10 m/s. Higher voltage transformer is possible because the high current contributed to a higher power output than needed for the 14.2 kV EHD thruster. [9]

The final task which needs to be done to make the project work is to make both the EHD thruster and the battery system lighter. A lighter system will enable the system to lift itself easier without the same need for a higher thrust.

VII. CONCLUDING REMARKS

The conducted study of a battery system for an EHD thruster concludes that there might be a future for small airplane drones if the battery system would sustain 14 kV. Now with an optimal EHD thruster, the battery system should be able to lift only the EHD thruster, if the drone is an airplane model.

However, this observation is not set in stone because there has not been any tests with the battery system and the optimal EHD thruster yet.

VIII. ACKNOWLEDGMENTS

The authors could never have finished this thesis if it is not for some people; Sven Hedin, Laila Ladhani, Hans Olof Sohlstr¨om, Mikael Bergqvist and the supervisor Wouter Metsola Van Der Wijngaart. We want to thank Sven for his help with inventory and extra knowledge of the battery system.

Laila helped with everything from measurement to testing, and for all the help. Thanks, Mikael for helping with purchases and for the extra knowledge at the Micro and Nanosystems department. Thanks to Hans for helping with the math behind the measurements and without him, the battery system would not have been possible to measure correctly. To our supervisor, Wouter, you gave us hope and supported us during the project so the project could be completed.

REFERENCES

[1] C. Wan, “Electro-hydrodynamic (ehd) thruster analysis and optimiza- tion,” Ph.D. dissertation, The Cooper Union for the Advancement of Science and Art, New York, 2009.

[2] H. Fantel, “Major de seversky’s ion-propelled aircraft.” Popular Me- chanic, vol. 58, no. 61, pp. 194–196, 1964.

[3] E. Y. Choueiri, “A critical history of electric propulsion: The first 50 years (1906-1956),” Journal of Propulsion and Power, vol. 20, no. 2, pp. 193–203, 2004.

[4] E. Moreau, N. Benard, F. Alicalapa, and A. Douy`ere,

“Electrohydrodynamic force produced by a corona discharge between a wire active electrode and several cylinder electrodes – application to electric propulsion,” Journal of Electrostatics, vol. 76, pp. 194 – 200, 2015. [Online]. Available:

http://www.sciencedirect.com/science/article/pii/S0304388615000637 [5] E. A. Christenson and P. S. Moller, “Ion-neutral propulsion in atmo-

spheric media.” AIAA Journal, vol. 5, no. 10, pp. 1768–1773, 1967.

[6] (2017, May) Blaze labs ehd thrusters research. [Online]. Available:

http://www.blazelabs.com/l-intro.asppeek

[7] Y. Zhang, L. Liu, Y. Chen, and J. Ouyang, “Characteristics of ionic wind in needle-to-ring corona discharge,” Journal of Electrostatics, vol. 74, pp. 15–20, 2015.

[8] E. Kuffel, W. Zaengl, and J. Kuffel, “High voltage engineering funda- mentals 2nd ed,” ISBN: 0, vol. 7506, no. 3634, p. 3, 2000.

[9] E. Moreau and G. Touchard, “Enhancing the mechanical efficiency of electric wind in corona discharges,” Journal of Electrostatics, vol. 66, no. 1–2, pp. 39 – 44, 2008. [Online]. Available:

http://www.sciencedirect.com/science/article/pii/S0304388607000939 [10] (2017, May) Rectifier circuits. [Online]. Avail-

able: https://www.allaboutcircuits.com/textbook/semiconductors/chpt- 3/rectifier-circuits/

[11] AspenCore. (2017, May). [Online]. Available: http://www.electronics- tutorials.ws/diode/diode˙6.html

[12] Nasa. (2017, May) General thruster equations. [Online]. Available:

https://www.grc.nasa.gov/www/k-12/airplane/thrsteq.html

[13] (2017, May) High voltage probe. [Online]. Available:

http://www.farnell.com/datasheets/1725284.pdf

[14] (2017, May) Fluke 87v industrial multimeter. [Online].

Available: http://en-us.fluke.com/products/digital-multimeters/fluke-87v- digital-multimeter.html

[15] (2017, May) Test equipment solutions. [Online].

Available: http://www.testequipmenthq.com/datasheets/TEKTRONIX- TDS340-Datasheet.pdf

[16] (2017, May) Solidworks cad. [Online]. Available:

https://www.solidworks.com

[17] (2017, May) Stratasys uprint se. [Online]. Available:

http://www.stratasys.com/3d-printers/idea-series/uprint-se

[18] (2017, May) High flow mass airflow/amplified. [Online]. Available:

https://www.elfa.se/Web/Downloads/˙t/ds/AWM5000˙eng˙tds.pdf