Decrease radiated emission from high frequency links

(1)

Decrease radiated emission from high

frequency links

Anna Vestin

Engineering Physics and Electrical Engineering, master's level (120 credits) 2020

Luleå University of Technology

(2)

(3)

This page was left intentionally blank

(4)

Abstract

There is a need for increasing speed on external links and harmonics are often seen and are a challenge to master. Especially for products within the space industry where the requirements are narrower due to the harsh environment it needs to withstand.

This thesis is investigating a potential leakage path around the interface from the board to harness through Spacewire connectors. This leakage is seen as radiated emission and the focus of this thesis is to see if there is a possibility to decrease these emissions by minimising the ground loop that is potentially causing these issues in the UHF- and S-band.

In order to verify that the proposed solution a reverberation chamber was used to analyse the performance of the solution when a twisted cord attached to a ground plane and wrapped around the harness was used to minimise the difference in potential between the interface of the board to harness.

The results showed that the proposed solution was inconclusive, but another discovery was made, it was found that the radiated emissions seen in the S-band can be decreased significantly by improving the contact between the connector and the frame. This could be done with a beryllium-copper gasket. A decrease could also be seen in the UHF-band but not as substantial as the S-band.

(5)

(6)

Acknowledgements

This master thesis was carried out at RUAG Space in Gothenburg. I would like to thank Jan Hellenberg and Hans Corin for giving me this opportunity to write my thesis at the company.

A Special thanks to Hans Corin who was an excellent supervisor and made me feel very welcome when being in a new environment. Many thanks for the support and advice throughout this thesis.

I also would like to thank my examiner Jonas Ekman for his advice and comments on this thesis.

Gothenburg, June 2020 Anna Vestin

(9)

(10)

Nomenclature & Abbreviations

βo Phase constant in free space

ηo Intrinsic impedance of medium in free space

o Permittivity in free space

µo Permeability in free space

λo Wavelength in free space

EMC Electromagnetic compatibility EMI Electromagnetic interference ESD Electro Static Discharge EUT Equipment Under Test GND Ground

LNA Low-Noise Amplifier LUF Lowest Usable Frecuency

LVDS Low Voltage Differential Signalling PCB Printed Circuit Board

RE Radiated Emission UHF Ultra High Frequency

(11)

(12)

Chapter 1 Introduction

The demand of building more powerful circuits with a small interface is increasing more and more, to be able to fulfil the requirements of future development such as being lightweight or taking up a significant smaller space on a device than before. These narrow specifications might lead to increasing electromagnetic interference, this may cause the desired behaviour of the developed product to change or make another device not function properly [1].

With an expanding market of electrical or electronic devices there are also an expan-sion of electrical and electronic devices that emit electromagnetic energy that interfere with other devices, whether its intentional or unintentional, four conditions or subgroups should be considered in this electromagnetic dilemma, radiated emissions, radiated sus-ceptibility, conducted emissions and conducted susceptibility. Susceptibility is the ability of the device to withstand electrical noise to the point that it can function as intended, and emission is the electromagnetic energy that is radiating out from your device, this energy can be measured in strength and can cause other devices to not perform as ex-pected [2]. The three aspects needed to cause an EMI problem is illustrated in figure 1.1.

Figure 1.1: Breakdown of the EMC coupling problem.

(13)

2 _Introduction

This undesired behaviour of interference cannot be stopped, but you can minimise this electromagnetic interference by different ways of methods, better shielding could be one of many ways to improve the problem. By implementing these kinds methods into your design to minimise electromagnetic emissions from a device or make it more susceptible to interference is in other words making it more compatible with each other [3].

The International Electrotechnical Commission, published in the International Elec-trotechnical Vocabulary [4], defines electromagnetic compatibility like this:

” The ability of equipment or a system to function satisfactorily in its electromagnetic environment without introducing intolerable electromagnetic disturbances to anything in

that environment. ”

RUAG space in Gothenburg is a Swiss owned company that specialises towards the defence and aerospace industry and has a long experience as a supplier of products for electronics on satellites and launchers, by developing electronic products for space travel the demands and requirements increase significantly because of the extreme and harsh environment the device needs to withstand. Verification of requirements during EMC tests is essential and important with increasing speed on external links. Harmonics are often seen and are a challenge to master, especially in programs where severe requirements are tailored for the mission in typically the UHF-, L- and S-band. Reducing very low emission involves understanding of the requirements and the potential sources of the emission.

1.1 Aim

(14)

1.2. Research questions 3

Figure 1.2: Drawing of RJ45 connector.

1.2 Research questions

In order to accomplish the aim set in 1.1, research questions has been appointed, and are the following:

• Which parameters are influencing emission?

• How can a change in design mitigate emissions from the board to harness? • Is the new design cost-effective?

1.3 Delimitation

• This thesis will only perform EMC tests using equipment provided by RUAG Space • This thesis will not consider to change the layout of the PCB to decrease emissions

1.4 Introduction to Spacewire

It started around 60 years ago with the lack of performance with bus interfaces. There was a need to increase speed for high-performance data-handling systems, and with adding nodes to the bus only made the system to limit its performance [5]. Another problem with the bus interface was that if any node had a failure it could immobilise the bus, or if a break would transpire a separation between the two buses would take place and communication would be terminated.

(15)

4 _Introduction

solve this dilemma and created what would be Spacewire, this was then released in 2003. According to Spacewire’s user guide by Dundee [6], it aims to:

• Facilitate the construction of high-performance on-board data-handling systems

• Help reduce system integration costs

• Promote compatibility between data-handling equipment and subsystems

• Encourage re-use of data-handling equipment across different missions

Spacewire is capable of delivering at high-speed rates as 2 Mbit/s to 200 Mbit/s through a network of point-to-point data-links that are connected via a Spacewire node to a new node or router. Spacewire has bi-directional, full-duplex, serial data links that send information as a serial bit stream, it uses Data-Strobe encoding to transmit data [7]. By XORing the two signals Data and Strobe the clock can be retrieved, an example of this is shown in figure 1.3.

Figure 1.3: Data-Strobe encoding and clock.

(16)

1.5. Outline 5

Figure 1.4: Specification of LVDS.

By having two differential signals that runs in two different paths, one carries the Data and Strobe in one way and the other one the opposite way. This concludes in a cable with four twisted pairs, which are shielded on different levels. There are several types of choices for connector, but the recommended one is specified as the nine pin micro-miniature D-type [9]. This connector is certified for use in space. The micro-D connector is illustrated in figure 1.5 with its pin-out.

Figure 1.5: Spacewire connector pin-out.

Spacewire is also compatible with the RJ45 connector as seen in figure 1.2.

1.5 Outline

(17)

6 _Introduction

• Chapter 2: Mathematical and physical descriptions in order to describe and verify the research questions formulated in previous section 1.2.

• Chapter 3: Methodology of this thesis, which includes a description of test setup, test equipment and EUT.

• Chapter 4: The results of the performed tests and other findings of this thesis. • Chapter 5: A analysis of the presented results from previous chapter.

(18)

Chapter 2 Theory

This chapter will present the mathematical and physical descriptions and theory to verify the research questions formulated in section 1.2 and also lay a foundation to the methods later presented in chapter 3.

2.1 Equipment

Equipment used during EMC testing for this thesis will be presented in this section and given a brief description.

2.1.1 Signal generator

A signal generator is a piece of electronic equipment and is generally used in testing, for designing or trouble shooting. The generator produces a repetitive signal in the form of a wave, analogue or digital. The most frequent types of waves produced by the generator are sine, saw tooth, square, triangular and pulse waves. These waves can be altered in frequency and amplitude to suit the specifications of the user.

2.1.2 FFT Spectrum analyser

A Fast Fourier Transform spectrum analyser is a device that can measure the magnitude or the strength of a spectrum by sampling the input signal. Though it is not feasible to use the standard Fourier Transform to convert data in to frequency domain, an alternative of the Fourier Transform is used instead, the Discrete Fourier Transform [2].

Fourier’s Transform states that

”Any waveform in the time domain can be represented by the weighted sum of sines and cosines.”

(19)

8 _Theory

The spectrum analyser computes the magnitude of its sine and cosine components and displays the spectrum of these measured frequency components. The different stages of how the analyser samples the input signal is described in figure 2.1.

Figure 2.1: Different stages of the spectrum analyser from sampling the input.

2.1.3 Reverberation chamber

A reverberation chamber is a shielded environment for electromagnetic compatibility or electromagnetic interference measurements where the absorption of the field is very low due to the high conductivity [10]. The shielded cavity is equipped with a stirrer to give a statistically uniform field inside the chamber. The stirrer can be seen in figure 2.2 and is a large metallic reflector and is used to stir the electromagnetic field inside the chamber. The purpose is to change the stochastic field inside the chamber by altering the standing wave pattern [11]. By changing the rotation of the stirrer, the boundary for the electromagnetic field inside the chamber will also change. In other words, the distribution of the electromagnetic field is statistically homogeneous for different positions inside the chamber.

The resonance frequency for a rectangular cavity is given by derivation from Maxwell’s equations [2] fmnp = c 2π s mπ w 2 + nπ h 2 + pπ l 2 , (2.1)

(20)

2.1. Equipment 9

Figure 2.2: Model of the reverberating chamber with stirrer used during testing.

2.1.4 Antenna & LNA

An antenna can either be a transmitter or receiver of electromagnetic waves, they come in many shapes and sizes to comply with frequency ranges and radiation characteristics. The purpose of the antenna during radiated emissions tests is for the receiving antenna to measure the field strength emitted from the EUT [1].

The type of antenna used for these measurements was an open boundary quad-ridge horn, model Satimo QH400. The benefit of this antenna is the wide frequency range, covering the general interesting frequencies for the companies products. Electrical char-acteristics for the antenna are presented in table 2.1 and a figure of the model is presented in figure 2.3. A Low Noise Amplifier was used together with the antenna, as hinted in the name the LNA amplifies low-power signals without compromising its SNR. The LNA model used for these measurements is 50M6G, parameters and correction factors are listed in the appendix.

(21)

10 _Theory

Note that the gain of an antenna is not of interest in a reverberating chamber as the power measured is captured from all angles.

Table 2.1: Electrical characteristics of the Satimo QH400 antenna.

Frequency range 0.4 - 6 GHz Gain 4 - 15 dBi

VSWR <1.9

Return loss <-10 dB Polarisation Dual linear Cross-polar discrimination >30 dB Port-to-Port isolation >30 dB Impedance 50 Ohms

2.2 Radiated emission

Radiated emissions occur when unintentional or intentional electromagnetic energy is released from an electronic device, the electromagnetic waves propagates through air to be picked up by a possible receiver [1]. This may cause an interference as mentioned in chapter 1, but radiated emission doesn’t always have to cause a problem, if the emissions are within the regulated requirement then the receiving device should function properly according to its intended functions. But if it doesn’t meet those requirements there are two alternatives, either you improve the device susceptibility, or you control the emissions emitted from the device [12].

By focusing on controlling radiated emissions, there are some different approaches you can take to reduce these emissions depending on the type of problem you have, for instance better shielding, grounding or cable design. A significant contributor to radiated emissions are common-mode currents on cables within a device or interconnect cables [2]. How common-mode currents are connected to radiated emissions will be explained further in the next section.

2.3 Differential and common-mode currents

Imagine two currents I1 and I2 flowing in two wires in proximity to each other, each

of those wires are carrying common-mode IC and differential-mode ID currents. These

currents can be written as following

I1 = IC + ID (2.2a)

(22)

2.3. Differential and common-mode currents 11

By adding and subtracting equations in 2.2. This will yield

IC = I1+ I2 2 (2.3a) ID = I1− I2 2 . (2.3b)

Seen from equation 2.3, small common-mode currents can contribute to a considerable higher amount of radiated emissions than differential-mode currents [13]. A figure illus-trating this argument can be seen in figure 2.4.

Figure 2.4: Common-mode current emission model.

An estimation model can be derived from the Hertzian dipole and the given currents. The elements of the magnetic field intensity vector turns into

ˆ Hr = 0 (2.4a) ˆ Hθ = 0 (2.4b) ˆ Hφ= ˆ Idl 4πβ 2 osin θ j 1 βor + 1 β2 or2 , (2.4c)

where βo = 2π/λo and λo = uo/f is a wavelength at frequency f. uo = √_µ1_o_o is the

(23)

12 _Theory

And the elements of the electric field are

ˆ Er = 2 ˆ Idl 4πηoβ 2 ocos θ 1 β2 or2 − j 1 β3 or3 e−jβor _(2.5a) ˆ Eθ = ˆ Idl 4πηoβ 2 osin θ j 1 βor + 1 β2 or2 − j 1 β3 or3 e−jβor _(2.5b) ˆ Eφ= 0. (2.5c)

Where ηo = pµo/o = 120π = 377 Ω is the intrinsic impedance of free space and

µo = 4π × 10−7 H/m, o = 1/36π × 10−9 F/m. ~ˆ Ef arf ield = jηoβo ˆ Idl 4π sin θ e−jβor r ~aθ (2.6a) ~ˆ Hf arf ield = jβo ˆ Idl 4π sin θ e−jβor r ~aφ (2.6b) I1 = IC (2.7a) I2 = IC (2.7b) ˆ EC,max= j2π × 10−7 f ˆICL d e −jβod_{e−jβos₂ _{+ e}jβos₂ _} = j4π × 10−7f ˆICL d e −jβod_cos1 2βos. (2.8)

By substituting 1₂βos = πs/λo and assuming that the distance s is small, so that

cos1₂βos ∼= 1, the magnitude reduces to

| ˆEC,max| = 1.257 × 10−6

| ˆIC|f L

d . (2.9)

2.4 Ground loops

(24)

2.4. Ground loops 13

in noise and interference in the device. To get a better concept of this phenomena an illustration is given in figure 2.5.

(25)

(26)

Chapter 3 Method

Methodology of this thesis, which includes a detailed description of test setup, test equipment and EUT.

3.1 Methodological quality

The chosen method of measuring the radiated emissions in this thesis will be carried out in an reverberation chamber. The method was chosen over the option of testing in a anechoic chamber due to the many advantages of the reverberation chamber such as shorter testing time which in turn leads to economic benefits, good repeatability during tests and that the chamber provides good electromagnetic isolation from the external surroundings [10]. Of course, there are always some disadvantages to this type of measurements which are high LUF, loss of information with respect to polarisation and directional properties of the EUT, but short testing time and good repeatability overcomes these disadvantages. The reverberation chamber was therefore more suitable for this thesis than the anechoic chamber in this case.

3.2 Description of test site

The test setup used during the test is based on the MIL-STD-461F [14] clean room setup for the functional tests as follows:

• Clean room area - inside reverberating chamber with stirrers • Solid ground plane

• Harness used during the test should be the standard test harness for functional tests, i.e. functional flight representative configuration

The EUT shall be placed at least λ₄ from any conductive ground surface, where λ is the lowest test frequency used. The exact position (x,y,z) of the EUT and harness shall be

(27)

16 _Method

noted, see figure 2.2 for origin of the chamber. The preferred position of the EUT is at the height of the fixed copper tables, 25cm from the edge of the table, 50cm from the wall. The measurement system shall be checked by configuring the test equipment as shown in figure 3.1. To setup the EUT antenna positioning, the physical reference points of the antenna shall be documented for measuring heights of the antennas and distances of the antennas from the test setup boundary shall be used as follows:

• Position antenna 1 m from the front edge of the test setup boundary for all setups. • Position the antenna above the floor ground plane.

• Ensure that no part of any antenna is closer than 1 m from the walls and 0,5 m from the ceiling of the shielded enclosure.

• Orient the antenna away from direct view of both the stirrer and the EUT. • Record the antenna position.

(28)

3.3. Design solution 17

3.3 Design solution

The proposed solution to be tested consists of a wire that is twisted around the Spacewire harness from the board to the connector. The idea of this solution was to minimise the ground loop that occurs between the interface of the board to the harness by connecting the wire to the ground plane of the board and twisting this wire around the harness to finally be soldered at the end of the wire against the frame so that it is grounded at both ends. To make this solution clearer, a drawing was made, figure 3.2, where the orange pattern in the picture represents the wire being twisted around the harness seen from the side. The design solution presented in this thesis were carefully developed through discussions and interviews with mechanical and system engineers as well as production technicians who all contributed with their expertise in the area.

Figure 3.2: Description of twisted cord solution.

3.4 Description of EUT & Test

(29)

18 _Method

Figure 3.3: Circuit scheme of test object.

After verification of the board was confirmed, the board was mounted on to a frame and then fitted in to a chassi, the front view of the chassi is pictured in figure 3.4, there after it was placed in the reverberation chamber. The antenna was placed on the copper table in the chamber and the LNA was connected in between the antenna and spectrum analyser, the power supply and signal generator were connected and the Spacewire harness attached. For the full layout of the setup in the reverberation chamber, see figure 3.1

Before the measurements could take place, a configuration of the settings was done on the analyser and the stirrer to suit the UHF- and S-band measurements. The range of the UHF-band lies between 300-3000 MHz and the S-band lies on the span of 2.0-4.0 GHz, these configurations can be viewed in the appendix. Because of the wide range of the UHF- and S-band a decision was made to only test in the areas where the demands of the requirements are tougher to meet. The range that was tested for the UHF-band was 420-460 MHz and for the S-band 2020-2080 MHz.

(30)

3.4. Description of EUT & Test 19

Figure 3.4: Front view of equipment under test.

(31)

20 _Method

Figure 3.6: Description of test setup e, foil tape against the front frame.

(32)

Chapter 4 Results & Discussion

This chapter presents the results from the conducted measurements done in the rever-beration chamber with the described methods from chapter 3. The results from the S-and UHF-bS-and are presented in separate sections were a discussion regarding the results also take place.

4.1 S-band

The results presented in this section were measured at 2020-2080 MHz which is in the region of the S-band. The results are presented in the order so that a comparison of all gathered measurements is shown first and there after different comparisons between the reference and twisted cord measurements are shown.

The results of the reference measurement with the different setup’s described in section 3.4 are shown in figure 4.1. There it can be seen that option a, foil tape on the inside of the connector from figure 3.5 and option e, foil tape against the front frame from figure 3.6 made the radiated emissions completely silent compared against option d, the reference without any addition. This concludes that there is not only a leakage between the interface of the board between the harness and the connector, there is also a leakage from the perforation made to fit the connector on the front frame.

It can also be seen that the other two options, b and c, foil tape on the inside of the back shell and respectively foil tape at the beginning of the harness also had a minimising effect on the emissions but not as great as option a and e.

(33)

22 _{Results & Discussion} 2 .020 ₂.025 ₂.030 ₂.035 ₂.040 ₂.045 ₂.050 ₂.055 ₂.060 ₂.065 ₂.070 ₂.075 ₂.080 ·10 9 −137.5 −135.5 −133.5 −131.5 −129.5 −127.5 Frequency[Hz] dB

Reference measurements, S-Band

Foil tape against the front frame

Foil tape at the beginning of the harness Foil tape on the inside of the back shell Measurement without any addition

Foil tape on the inside of the connector socket

Figure 4.1: This figure compares the different setup’s tested on the reference.

2 .020 ₂.025 ₂.030 ₂.035 ₂.040 ₂.045 ₂.050 ₂.055 ₂.060 ₂.065 ₂.070 ₂.075 ₂.080 ·10 9 −137.5 −132.5 −127.5 −122.5 Frequency[Hz] dB

Twisted cord, S-Band

Foil tape at the beginning of the harness Foil tape on the inside of the back shell Measurement without any addition

Foil tape on the inside of the connector socket

(34)

4.1. S-band 23

The results of when the twisted cord solution was added on to the board and tested together with the different setups from section 3.4, they are presented in figure 4.2. There it can be seen that there is an overall increase of the emissions and especially between frequencies 2055-2065 MHz compared to the reference in figure 4.1 where it was silent. For the case where option a, foil tape on the inside of the connector socket and e, foil tape against the front frame are tested together with the twisted cord, it can be seen that it is completely silent there as well.

2 .020 ₂.025 ₂.030 ₂.035 ₂.040 ₂.045 ₂.050 ₂.055 ₂.060 ₂.065 ₂.070 ₂.075 ₂.080 ·10 9 −137.5 −132.5 −127.5 −122.5 Frequency[Hz] dB

Comparison of reference and twisted cord, S-Band

Reference - Foil tape at the beginning of the harness Twisted cord - Foil tape at the beginning of the harness

Figure 4.3: Comparison of reference and the twisted cord option with option c.

(35)

24 _{Results & Discussion} 2 .020 ₂.025 ₂.030 ₂.035 ₂.040 ₂.045 ₂.050 ₂.055 ₂.060 ₂.065 ₂.070 ₂.075 ₂.080 ·10 9 −137.5 −132.5 −127.5 −122.5 Frequency[Hz] dB

Reference - Foil tape on the inside of the back shell Twisted cord - Foil tape on the inside of the back shell

Figure 4.4: Comparison of reference and the twisted cord option with option b.

2 .020 ₂.025 ₂.030 ₂.035 ₂.040 ₂.045 ₂.050 ₂.055 ₂.060 ₂.065 ₂.070 ₂.075 ₂.080 ·10 9 −137.5 −132.5 −127.5 −122.5 Frequency[Hz] dB

Reference - Measurement without any addition Twisted cord - Measurement without any addition

(36)

4.1. S-band 25

Moving on from the setup in section 3.4, in figure 3.5 and 3.6. It is now more clear where the leakage exists therefore the copper tape can be exchanged against the more realistic solution the beryllium-copper gasket.

2 .020 2 .025 2 .030 2 .035 2 .040 2 .045 2 .050 2 .055 2 .060 2 .065 2 .070 2 .075 2 .080 ·10 9 −137.5 −135.5 −133.5 −131.5 −129.5 −127.5 Frequency[Hz] dB

Reference measurements with gasket added, S-Band

Reference - Gasket added

Reference - Without any addition

Figure 4.6: Comparison of the reference with and without gasket.

(37)

26 _{Results & Discussion}

4.2 UHF-band

The results from the UHF-band was swept over 400-460 MHz and was carried out in the same way as the measurements done in the S-band. The results will also be displayed in the same order, where the gathered results from the reference and the twisted cord solution are shown first with the different setups from section 3.4 in separate plots and there after compared to each other. The field in this band focuses on emissions from the harness and there for shows a different result than in the S-band

The reference measurements are shown in figure 4.7, in this measurement it is quite difficult to see the differences between the different setups due to the small deviations. One could say that option e, foil tape to the front frame had the best overall outcome in this test by comparing the bars.

4 .02 ₄.05 ₄.08 ₄.11 ₄.14 ₄.17 4.2 ₄.23 ₄.26 ₄.29 ₄.32 ₄.35 ₄.38 ₄.41 ₄.44 ₄.47 4.5 ₄.53 ₄.56 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 −112.5 Frequency[Hz] dB

Reference measurement, UHF-Band

Foil tape at the beginning of the harness Foil tape on the inside of the back shell

Foil tape on the inside of the connector socket Measurement without any addition

Figure 4.7: This figure compares the different setup’s tested on the reference.

(38)

4.2. UHF-band 27

more inconclusive. There are very small differences that gives the reference a lower gain than the twisted cord option, but a conclusion cannot be drawn considering these mixed results more than that the twisted cord solution is an unsatisfactory solution.

4 .02 4 .05 4 .08 4 .11 4 .14 4 .17 4.2 4 .23 4 .26 4 .29 4 .32 4 .35 4 .38 4 .41 4 .44 4 .47 4.5 4 .53 4 .56 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 Frequency[Hz] dB

Twisted cord, UHF-Band

Foil tape at the beginning of the harness Foil tape on the inside of the back shell

Foil tape on the inside of the connector socket Measurement without any addition

(39)

28 _{Results & Discussion} 4 .02 4 .05 4 .08 4 .11 4 .14 4 .17 4.2 4 .23 4 .26 4 .29 4 .32 4 .35 4 .38 4 .41 4 .44 4 .47 4.5 4 .53 4 .56 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 −112.5 Frequency[Hz] dB

Comparison of reference and twisted cord, UHF-Band

Reference - Foil tape against the front frame Twisted cord - Foil tape against the front frame

Figure 4.9: Comparison of reference and the twisted cord with option e.

4 .02 ₄.05 ₄.08 ₄.11 ₄.14 ₄.17 4.2 ₄.23 ₄.26 ₄.29 ₄.32 ₄.35 ₄.38 ₄.41 ₄.44 ₄.47 4.5 ₄.53 ₄.56 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 Frequency[Hz] dB

Reference - Foil tape at the beginning of the harness Twisted cord - Foil tape at the beginning of the harness

(40)

4.2. UHF-band 29 4 .02 4 .05 4 .08 4 .11 4 .14 4 .17 4.2 4 .23 4 .26 4 .29 4 .32 4 .35 4 .38 4 .41 4 .44 4 .47 4.5 4 .53 4 .56 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 Frequency[Hz] dB

Reference - Foil tape on the inside of the back shell Twisted cord - Foil tape on the inside of the back shell

Figure 4.11: Comparison of reference and the twisted cord with option b.

4 .02 ₄.05 ₄.08 ₄.11 ₄.14 ₄.17 4.2 ₄.23 ₄.26 ₄.29 ₄.32 ₄.35 ₄.38 ₄.41 ₄.44 ₄.47 4.5 ₄.53 ₄.56 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 −112.5 Frequency[Hz] dB

Reference - Foil tape on the inside of the connector socket Twisted cord - Foil tape on the inside of the connector socket

(41)

30 _{Results & Discussion} 4 .02 4 .05 4 .08 4 .11 4 .14 4 .17 4.2 4 .23 4 .26 4 .29 4 .32 4 .35 4 .38 4 .41 4 .44 4 .47 4.5 4 .53 4 .56 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 Frequency[Hz] dB

Reference - Measurement without any addition Twisted cord - Measurement without any addition

Figure 4.13: Comparison of reference and the twisted cord with option d, no addition.

4 .02 ₄.05 ₄.08 ₄.11 ₄.14 ₄.17 4.2 ₄.23 ₄.26 ₄.29 ₄.32 ₄.35 ₄.38 ₄.41 ₄.44 ₄.47 4.5 ₄.53 ₄.56 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 −112.5 Frequency[Hz] dB

Reference measurement with gasket added, UHF-Band

Reference - Gasket added

Reference - Without any addition

(42)

4.3. Choice of voltage source 31

The beryllium-copper gasket was also tested in the UHF-band as seen in figure 5.6, there it is evident that the gasket has a minimising effect on this band too but not as significant as in the S-band.

4.3 Choice of voltage source

In the beginning of this thesis a choice was made to have a 3V battery to supply the circuit. The decision was made based on the need for good EMC with as little interference as possible. Keeping the battery on the inside of the chassi would reduce the potential leakage of having a voltage source being connected from the outside. Measurements were done and analysed with the 3V battery setup and a conclusion was made that depending on the potential of the battery during the measurements, this had an immediate effect on the results. The results could differ as much as up to 20 db depending on if the battery was recently changed or not, which made the results non-consistent.

The voltage source was therefor changed into a power supply unit to eliminate such behaviour and to give better repetitiveness in measurements.

4.4 Choice of Spacewire harness

The Spacewire harness used in the beginning was replaced due to the fact that it was not showing consistent behaviour, let’s call this harness the ”blue harness”, to a harness that was flight representative. After making the replacement, the results showed that the ”blue harness” was emitting radiated emissions on a higher level than the flight representative harness and was therefore exchanged in this test setup. The difference of the harnesses can be seen in figure 4.15.

(43)

32 _{Results & Discussion} 4 .00 4 .05 4 .10 4 .15 4 .20 4 .25 4 .30 4 .35 4 .40 4 .45 4 .50 4 .55 4 .60 ·10 8 −137.5 −132.5 −127.5 −122.5 −117.5 −112.5 Frequency[Hz] dB

Comparison of Spacewire harness - UHF-band

Flight representative Blue harness

(44)

Chapter 5 Conclusion and future work

From the results obtained in chapter 4, a conclusion can be made that there is not only leakage in the form of radiated emissions between the interface of the PCB to the harness, it exists as well on the harness of the plug connector which was thought not to be a culprit in the beginning of this thesis.

When discovering this issue, the test setup described in section 3.4, in figure 3.5 and 3.6 was added in to the testing to see where the leakage was coming from so that it could be eliminated. The test showed that in the S-band there was limited contact between the connector and the front frame which made it possible for the emissions to slip out of the perforation made to fit the connector on to the front frame. When testing setup a and e from mentioned figures, the signal is completely silent, and the issue is resolved for that particular band. Copper tape made it clear where the leakage was coming from and was then substituted for a beryllium-copper gasket, which can be seen in figure 4.6. The results showed a decrease in emissions by 5-8 dB which is a satisfactory result. In the future it is recommended to have a gasket on all Spacewire connections to minimise emissions.

For the twisted cord solution in the S-band it could be seen that an increase of the emissions was discovered when adding the solution on to the board. The purpose of the solution was to minimise the ground loop with this wire, but instead an increase of the emissions occurred.

Trying to reduce the radiated emissions from the UHF-band was not as clear as the results obtained from the S-band. The results were inconclusive and showed only a small improvement when looking at the reference with the different test setup’s and also when comparing the reference with the twisted cord option it could only be seen some small changes and it was not possible to say that either of these solutions had an effect on the emissions in the UHF-band. Only when adding the beryllium-copper gasket on to the reference there was a consistent, but small decrease in emissions, an average of 3 dB.

For further work another solution needs to be implemented to see whether it has an effect on the UHF-band to be able to reduce emissions as wished, one solution could be to add a spring gasket mounted on the plug connector to improve EMI further. Also another

(45)

34 _{Conclusion and future work}

(46)

References

[1] H. W. Ott, Noise reduction techniques in electronic systems. Wiley Interscience, 1936.

[2] P. R. Clayton, Introduction to electromagnetic compatibility. Wiley Interscience, 1992.

[3] M. Mardiguian, How to control electrical noise. Don White consultants Inc, 1983.

[4] I. electrotechnical commission, International electrotechnical vocabulary.

[5] P. Walker, “The origins of spacewire,” International spacewire symposium, 2003.

[6] S. Parkes, SpaceWire user’s guide. STAR-Dundee, 2012.

[7] A. T. Baklezos, C. D. Nikolopoulos, C. N. Capsalis, and S. Tsatalas, “Effect of lvds link speed and pattern length on spectrum measurements of a spacewire harness,” in 2017 International Workshop on Antenna Technology: Small Antennas, Innovative Structures, and Applications (iWAT), pp. 38–41, 2017.

[8] T. Instruments, Interface circuits for TIA/EIA-644 (LVDS). Texas Instruments, 2002.

[9] European cooperation for space standardization, Space engineering - Electromag-netic compatibility. ECSS Secritariat, ESA-ESTEC, 2012.

[10] M. Bäckström, O. Lundén, and P.-S. Kildal, “Reverberation chambers for emc sus-ceptibility and emission analyses,” pp. 429–452, Review of Radio Science, 2002.

[11] Svensk Elstandard, Electromagnetic compatibility (EMC) - Part 4-21: Testing and measurement techniques - Reverberation chamber test methods. Svensk Elstandard, 2004.

[12] M. Mardiguian, Controlling radiated emissions by design. Kluwer academic publish-ers, 2001.

[13] C. R. Paul and D. R. Bush, “Radiated emissions from common-mode currents,” pp. 1–7, 1987 IEEE International Symposium on Electromagnetic Compatibility, 1987.

(47)

36 _References

(48)

Appendices

A

Correction factors

Table 5.1: Correction factor for Satimo QH400.

(49)

38 _Appendices

Table 5.2: Parameters for LNA50M6G

Frecuency [MHz] Gain [dB] Cable loss [dB] Correction factors [dB]

(50)

Appendices 39

B

Spectrum analyser setup

Table 5.3: Table of the used settings of the analyser.

Basic settings

Number of Points 40001

Sweep Time 0,109333333 s Average Type Power(RMS)

RBW 2000

RBW Filter Gaussian RBW Filter BW 3 dB

VBW 1000000

(51)

40 _Appendices

C

Enlarged figures

2.020

2.025

2.030

2.035

2.040

2.045

2.050

2.055

2.060

2.065

2.070

2.075

2.080 ·10

9

−

137 .5

−

135 .5

−

133 .5

−

131 .5

−

129 .5

−

127 .5

F

requency[Hz]

dB

Reference

measuremen

ts,

S-Band

F

oil

tap

e

against

th

e

fron

t

frame

F

oil

tap

e

at

the

b

eginning

of

the

harness

F

oil

tap

e

on

the

inside

of

the

bac

k

shell

Measuremen

t

without

an

y

addition

F

oil

tap

e

on

the

inside

of

the

connector

so

ck

et

(52)

Appendices 41

2.020

2.025

2.030

2.035

2.040

2.045

2.050

2.055

2.060

2.065

2.070

2.075

2.080 ·10

9

−

137 .5

−

132 .5

−

127 .5

−

122 .5

F

requency[Hz]

dB

Twisted

cord,

S

-Band

F

oil

tap

e

against

the

fron

t

frame

F

oil

tap

e

at

the

b

eginning

of

the

harness

F

oil

tap

e

on

the

inside

of

th

e

bac

k

shell

Measuremen

t

without

an

y

addition

F

oil

tap

e

on

the

inside

of

th

e

connector

so

ck

et

(53)

42 _Appendices

2.020

2.025

2.030

2.035

2.040

2.045

2.050

2.055

2.060

2.065

2.070

2.075

2.080 ·10

9

−

137 .5

−

135 .5

−

133 .5

−

131 .5

−

129 .5

−

127 .5

F

requency[Hz]

dB

Reference

measuremen

ts

with

g

ask

et

added

,

S-Band

Reference

-Gask

et

added

Reference

-Without

an

y

addition

(54)

Appendices 43

4.02

4.05

4.08

4.11

4.14

4.17

4.2

4.23

4.26

4.29

4.32

4.35

4.38

4.41

4.44

4.47

4.5

4.53

4.56 ·10

8

−

137 .5

−

132 .5

−

127 .5

−

122 .5

−

117 .5

−

112 .5

F

requency[Hz]

dB

Reference

measuremen

t,

UHF-Band

F

oil

tap

e

against

th

e

fron

t

frame

F

oil

tap

e

at

the

b

eginning

of

the

harness

F

oil

tap

e

on

the

inside

of

the

bac

k

shell

F

oil

tap

e

on

the

inside

of

the

connector

so

ck

et

Measuremen

t

without

an

y

addition

(55)

44 _Appendices

4.02

4.05

4.08

4.11

4.14

4.17

4.2

4.23

4.26

4.29

4.32

4.35

4.38

4.41

4.44

4.47

4.5

4.53

4.56 ·10

8

−

137 .5

−

132 .5

−

127 .5

−

122 .5

−

117 .5

F

requency[Hz]

dB

Twisted

cord,

UHF-Ban

d

F

oil

tap

e

against

the

fron

t

frame

F

oil

tap

e

at

the

b

eginning

of

the

harness

F

oil

tap

e

on

the

inside

of

the

bac

k

shell

F

oil

tap

e

on

the

inside

of

the

connector

so

ck

et

Measuremen

t

without

an

y

addition

(56)

Appendices 45

4.02

4.05

4.08

4.11

4.14

4.17

4.2

4.23

4.26

4.29

4.32

4.35

4.38

4.41

4.44

4.47

4.5

4.53

4.56 ·10

8

Decrease radiated emission from high frequency links