Vega radar book

��

��

��

��

��

��

��

��

��

��

Radar level measurement- The users guide

Peter Devine

© VEGA Controls / P Devine / 2000All rights reseved. No part of this book may reproduced in any way, or by any means, without priorpermissio in writing from the publisher:VEGA Controls Ltd, Kendal House, Victoria Way, Burgess Hill, West Sussex, RH 15 9NF England.

British Library Cataloguing in Publication Data

Devine, PeterRadar level measurement - The user´s guide1. Radar2. Title621.3´848

ISBN 0-9538920-0-X

Cover by LinkDesign, Schramberg.Printed in Great Britain at VIP print, Heathfield, Sussex.

written byPeter Devine

additional informationKarl Grießbaum

type setting and layoutLiz Moakes

final drawings and diagramsEvi Brucker

Foreword ixAcknowledgement xiIntroduction xiii

Part I1. History of radar 12. Physics of radar 133. Types of radar 33

1. CW-radar 332. FM - CW 363. Pulse radar 39

Part II4. Radar level measurement 47

1. FM - CW 482. PULSE radar 543. Choice of frequency 624. Accuracy 685. Power 74

5. Radar antennas 771. Horn antennas 812. Dielectric rod antennas 923. Measuring tube antennas 1014. Parabolic dish antennas 1065. Planar array antennas 108Antenna energy patterns 110

6. Installation 115A. Mechanical installation 115

1. Horn antenna (liquids) 1152. Rod antenna (liquids) 1173. General consideration (liquids) 1204. Stand pipes & measuring tubes 1275. Platic tank tops and windows 1346. Horn antenna (solids) 139

B. Radar level installation cont. 1411. safe area applications 1412. Hazardous area applications 144

Contents

ix

To suggest that any one type of leveltransmitter technology could be regar-ded as 'universal' would be unrealisticand potentially irresponsible due to thevariation and complexity of availableapplications when liquids, powders andsolids are all considered. However, therate at which radar based level trans-mitters have established themselvesover the last couple of years wouldtend to suggest that this technology iscloser to that definition that any princi-ple has ever been.

I have personally been involved inthe development, applications, salesand marketing of level transmitters,controllers and indicators of most typesover the last twenty years. In that timenothing has, in my opinion, come closeto matching the significance of radar interms of its overall suitability, for notjust conventional but extreme processconditions applications for the vastmajority of substances in vessels of vir-tually any size or complexity.

This unique principle combined withcurrent reflections processing software,materials of construction, simplicity ofinstallation and transmitter digital com-munications allows this to be conside-red as a day to day 'first consideration'for level, whereas only a very shorttime ago it was regarded as expensiveand specialised - this is no longer thecase.

The purpose of this publication isquite specific, and that is to explainsome of the principles involved, and toshow that by applying some simpleguidelines, what is obviously a sophi-sti-cated technology can be simple andreliably used in an enormously widerange of industrial and process applica-tions.

We make no apology for including achapter on Vega specific products, andhope this guide stimulates a radar user,or some greater depth of knowledge ifyou have some experience, we lookforward to hearing from you.

Mel HenryManaging DirectorVega Controls Ltd.

Foreword

xi

In writing and compiling this book Ihad the invaluable assistance of severalcolleagues from VEGA in Schiltachboth in the developing department andwithin the product management.

Particular thanks must go to KarlGriessbaum for his lucid explanationsof the 'secrets' of pulse radar; his insi-ght into the workings of FM - CWradar and the drawings to accompanythe explanations. Thanks also toJuergen Skowaisa and Juergen Motzerfor their technical contributions to thebook.

The publication of 'radar level mea-surement - the user´s guide' is a reflec-tion of the wealth of product knowled-ge of radar level application experiencein the VEGA group of companies andour agents and distributors world wide.

This experience has acceleratedsince the advent of the VEGAPULS 50series two wire, loop powered radar.

I would like to thank all those whocontributed to the section on radarapplications. This in-cludes DougAnderson, Dave Blenkiron, ChrisBrennan, Graeme Cross and JohnHulme in the UK, Paal Kvam ofHyptech in Norway, Dough Groh andhis colleagues at Ohmart VEGA in theUSA, and Juergen Skowaisa and RogerRamsden from VEGA Germany. Thankalso to the VEGA marketing depart-ment in Germany and the UK for their

assistance in producing and collatingpictures and photographs.

Thank to all the other unnamed con-tributors.

Finally, the most important contribu-tors to this book are all VEGA radarusers world wide without whom ourhigh level of expertise in process radarmeasurement applications would not bepossible.

Peter DevineTechnical managerVega Controls Ltd.

Acknowledgements

xiii

The technical benefits of radar as alevel measurement technique are clear.

Radar provides a non-contact sensorthat is virtually unaffected by changesin process temperature, pressure or thegas and vapour composition within avessel.

In addition, the measurement accu-racy is unaffected by changes in densi-ty, conductivity and dielectric constantof the product being measured or by airmovement above the product.

These benefits have become moresignificant to the process industry sincethe advent of low costs, high perfor-mance, two wire loop powered radarlevel transmitters.

This breakthrough, in the summer of1997, produced an unprecedentedboom in the use of non-contact micro-wave radar transmitters for liquid andsolids process level application.

'Radar level measurement - theuser´s guide' is offered as a referencebook for all those interested in the tech-nology, the application, and the prac-tical installation of radar level sen-sors.We cover many practical process levelapplications rather than the closedniche market of custody transfer mea-surement.

Radar history, physics and techni-ques are presented as well as descripti-ons of types of ra-dar antenna andmechanical and electrical installations.

Now radar is an affordable optionfor process level measurement. Wecompare it closely with all of the otherprocess level techniques and give manyexamples of the myriad applications ofradar across all industries.

Radar level measurement has comeof age. We hope that this book will beinvaluable in helping you to see thepotential of this latest and almost uni-versal level measurement technology.

More than anything, we hope thatyou enjoy delving into the pages of thisbook.

Peter DevineTechnical managerVega Controls Ltd

Introduction

1

James Clerk Maxwell predicted theexistence of radio waves in his theoryof electromagnetism as long ago as1864. He showed mathematically thatall electromagnetic waves travel at thesame velocity in free space,independent of their wavelength. Thisvelocity is of the order of 300,000 kilo-metres per second, the speed of light.

Heinrich Rudolf Hertz, verifiedMaxwell’s theory by experiments car-ried out in 1886-87 at KarlsruhePolytechnic. He used a spark gap trans-mitter producing bursts of high fre-quency electromagnetic waves at about455 MHz, or a wavelength of 0.66metres.

Hertz confirmed that these electro-magnetic radio waves had the samevelocity as light and could be reflectedby metallic and dielectric bodies. Inaddition to their reflective properties,Hertz demonstrated that radio wavesexhibit refraction, diffraction, polariza-tion and interference in the same wayas light. These early experiments inreflecting radio waves off metal plateswere the first manifestations of radar aswe know it today.

The first practical form of radar wasproduced by a German engineer,Christian Hülsmeyer. Patented in vari-ous countries in 1904 as the‘Telemobiloscope’, Hülsmeyer’s appa-ratus was described as ‘A Hertzianwave projecting and receiving appara-tus adapted to indicate or givewarning of the presence of a metallicbody, such as a ship or a train, in theline of projection of such waves’.

An addition to the patent in the sameyear described ‘Improvements inHertzian wave projecting and receiving

1. History of radar

James Clerk Maxwell - predicted the existence of radio waves in

his theory of electromagnetism (Pic. 1.1 - J.C.M.F)

Heinrich Hertz - Hertz confirmed by experiment that elec-tromagnetic radio waves have the samevelocity as light and can be reflected by

metallic and dielectric bodies(Pic. 1.2 - I.N.T)

2

apparatus for locating the position ofdistant metal objects’.

A successful demonstration of thetelemobiloscope was made at theInternational Shipping Congress inRotterdam in 1904, and also to theGerman navy. However, the telemo-biloscope was considered to be limitedand was not a commercial success.

Guglielmo Marconi, is famous forpioneering trans-Atlantic radio commu-nications. In 1922 Marconi had alsorecognised the potential of using shortwave radio for the detection of metallicobjects. Marconi envisaged the use ofradio for ship to ship detection at nightor in fog. However, he did not appearto receive the support or have theresources to carry these ideas further atthe time.

Prior to World War II, radar wasbeing developed independently in anumber of different countries, includ-ing Britain, Germany, the UnitedStates, Italy, France and the SovietUnion.

In 1934, following a series of exper-iments at the Naval ResearchLaboratory in the United States, apatent was granted to Taylor, Youngand Hyland for a ‘System for detectingobjects by radio’.

In February 1935, British scientist,Robert Watson-Watt presented a paperon ‘The detection and location of air-craft by radio methods’ to the TizardCommittee for the Scientific Survey ofAir Defence.

Christian Hülsmeyer produced the first practical radar

patented in 1904 (Pic. 1.3 - D.M.M)

Guglielmo Marconi recognised the potential of using shortwave radio for the detection of metallic

objects in 1922(Pic. 1.4 - GEC Marconi)

1. History of radar

3

Subsequently, a practical demonstra-tion was carried out using a BBC radiotransmitter at Daventry. About five anda half miles (9 km) away, a separateradio receiver connected to an oscillo-scope was used to detect the presenceof a Handley Page Heyford aircraft asit flew between the transmitter andreceiver.

Both the American system andWatson-Watt’s Daventry experimentwere types of continuous wave (CW)radar. Called CW wave-interferenceradar or bistatic CW radar, a continu-ous single frequency was transmittedfrom one point and detected by areceiver at a separate location. Thereceiver also detects doppler shiftedechoes from the target object. Theinterference between the frequency ofthe direct signal and reflected signals at

a slightly different frequency indicatedthe presence of the target object.

If you are unfortunate enough to liveon an airport flight path, you may havewitnessed this effect on your televisionscreen. As an aircraft approaches, thepicture on the screen may flicker withregular horizontal bands scrolling verti-cally on the screen. These diminishwhen the aircraft is directly overheadand then continue as the aircraft movesaway.

Although it proved a point atDaventry, CW wave-interference radarwas not a practical device. It coulddetect the presence but not the positionof the target.

After Daventry, the British effortcontinued at Orford Ness and thennearby Bawdsey Manor on the Suffolkcoast. It was clear that pulse radarwould be needed to provide therequired distance and direction infor-mation essential for a defensive radiodetection system.

The British, under the direction ofWatson-Watt developed a defensivesystem of CH (Chain Home) radar sta-tions which eventually covered all ofthe coastal approaches to Britain. Thestandard chain home radars had a rela-tively low frequency of between 22 &30 MHz (wavelength 10 to 13.5metres). They had a power of 200 kilo-watts and a range of up to 190 kilo-metres.

However, the long range CH radartransmitters were blind to low flyingaircraft and therefore they were supple-mented by CHL (Chain Home Low)radar transmitters which had a shorter range and covered the lower altitudesthat were overlooked by the main CH

Sir Robert Watson - Watt was a senior figure in the development

of British radar in the 1930’s & 40’s(Pic. 1.5 - I.W.M)

4

transmitters. They operated on a fre-quency of 200 MHz (wavelength 1.5metres).

It is well documented that the CHand CHL network of radar stationswere a crucial factor during the Battleof Britain in the summer of 1940. Itenabled the fighters of the Royal AirForce to be deployed when and wherethey were needed and rested when thethreat receded. The limited resources inmen and machines were not wasted inlong standing patrols.

German radar research was also con-ducted in secret in the late 1930’s.Whereas the development effort inBritain was focused on air defence, inGermany separate radar developmentswere carried out for the Navy, Armyand Luftwaffe.

Companies involved in Germannaval research produced a range of ship

mounted sea search radar transmitterscalled Seetakt. These were delivered asearly as 1938 with a frequency of 366MHz (wavelength 82 cm) and wereinstalled on many vessels including thefamous battleships, Bismarck and GrafSpee.

German Naval developments alsoproduced the Freya range of searchradars operating on 125 MHz (wave-length 2.4 metres). These were found tobe effective for tracking aircraft at longrange, and were subsequently suppliedto the Luftwaffe for early warning.However, they could not provide alti-tude information.

Other German radars in wide usewere the parabolic antenna Würzburgand Würzburg Riese (Giant Würzburg)transmitters. The standard Würzburgswere generally used for directingsearchlights and flak batteries and theWürzburg Riese for tracking individualintruders and directing night fighters tointercept them.

In a similar fashion to the BritishChain Home system, the Germans builta defensive network of ‘Himmelbett’radar stations. The literal translation ofHimmelbett is four poster bed. Thefour ‘posts’ of the bed consisted of aFreya early warning radar, a Würzburgradar for tracking the intruding aircraft,a Würzburg radar to guide the nightfighter to the intruder and a Seeburgplotting table (Seeburgtisch) to monitorthe interception.

This defensive radar system becameknown by the British as the‘Kammhuber Line’ after the Germangeneral in charge of night fighters.

British Chain Home Radar aerials - Radar was instrumental in the defenceof Britain during the second world war

(Pic. 1.6 - I.W.M)

1. History of radar

Above - The famous aerial reconnaissancephotograph of a German Würzburg radarantenna at Bruneval in northern France.

This image alerted the British to thepresence and advanced state of German

defensive radar which led to a commandoaction in which components from theradar were taken back to Britain for

analysis(Pic. 1.7 - I.W.M)

Right - The German Würzberg radar wasused for directing searchlights and flakbatteries and for tracking individual tar-gets and directing interceptors to them

(Pic. 1.8 - P.D)

5

6

Both Britain and Germany devel-oped airborne radar for fighter inter-ception by night. British airborne radartrials started in 1937 with the produc-tion AI Mark 1 taking to the air in May1939. The first practical BritishAirborne Interception radar was the AIMark IV which was first tested inAugust 1940.

In Germany the Lichtenstein air-borne radar was available in mid 1941.The characteristic external radar aerialarray of the Lichtenstein caused signifi-cant aerodynamic drag. This couldreduce the aircraft speed by as much as40 kilometres per hour. By 1943 therange had been extended to 6000metres.

It became clear to radar researchersthat a shorter ‘centimetric’ wavelengthwould be more useful for a number ofapplications. This would enable a morefocused airborne radar that would notsuffer from the ground returns thatrestricted capabilities of the first air-borne radars. The higher frequencycould be used for a ground mappingradar unit to locate towns and othergeographic features.

The problem was how to find amethod of generating sufficient powerat the desired wavelength of 10centimetres.

German Airborne Radar ‘Lichtenstein’available in mid 1941 - the external aerialradar caused significant aerodynamic drag

(Pic. 1.10 - I.W.M)

British Airborne Radar - AI Mark IVdeveloped for fighter interception by night

in 1940(Pic. 1.9 - I.W.M)

1. History of radar

7

In late February 1940, an historicbreakthrough was made by JohnRandall and Harry Boot, researchers atthe University of Birmingham, whenthey tested their world changing inven-tion the Cavity Magnetron.

The heart of this cavity magnetronwas a simple solid copper block withsix cavities machined into it. In thecentre was the cathode. When a strongmagnetic field and high voltage wasapplied between the copper block andthe cathode, the stream of electrons res-onated in unison within the cavitiesinstead of passing directly to the copperblock anode. The frequency of oscilla-tion was calculated to be about 3 GHz(10 centimetre wavelength).

The theoretical calculations of theprototype cavity magnetron were cor-rect. The actual wavelength was foundto be 9.87 centimetres and the allimportant power of the prototype was400 Watts.

Production of cavity magnetrons fol-lowed very quickly and the power out-put was significantly increased. Britaindeveloped microwave airborne inter-ception AI radar sets for night fighterswhich had a vastly improved long andnear range. The British microwave air-borne interception radar was the AIMark VII which was introduced in mid1942. The improved AI Mark VIII wasmass produced and in wide use byearly 1943.

Cavity Magnetron -the world changing invention by John

Randall and Harry Boot invented in 1940(Pic. 1.11 - GEC)

The Cavity Magnetron was used in centrimetric ‘microwave’ airborne radar and pro-duced a quantum leap in performance. The radar dish was protected inside a

plastic nose assembly(Pic. 1.12 & 1.13 - H.R.A)

8

Britain also used the cavity mag-netron in the development of a groundmapping radar called H2S. This deviceenabled aircraft to be accurately navi-gated to their destinations without theaid of ground based beacons or beams.

Britain shared this secret microwavetechnology with the United Stateswhere additional development tookplace at the Radiation Laboratory with-in the Massachusetts Institute ofTechnology. From the work carried outat MIT, further airborne interceptionradars and gun laying radars were massproduced and delivered to the alliedforces. The American SCR-720 (knownas AI Mark X in Britain) was firstdelivered to the USAAF by late 1942.

This radar unit became a standarddevice long after the war had finished.

War time secrecy meant that radiodetection devices were given codednames. In Britain, the early chain homeradar was called RDF after the existingRadio Direction Finding systems in thehope that it would mislead their real

function.In the same way in Germany,

radar was disguised as ‘DezimeterTelegraphie’ or ‘De-Te’, translated asdecimetric telegraphy

It was the Americans who intro-duced the now universally used palin-drome, RADAR or RAdio DetectionAnd Ranging.

The history of the development ofradar during the course of the SecondWorld War is a huge subject initself. Many devices were developed.Measures and counter measures weretaken in the radar war.

Since 1945, radar has been used foran increasing number of peaceful appli-cations. The giant Würzburg parabolicradar transmitters of the Second WorldWar became post war radio telescopes.The basic designs were developed andenlarged and can be seen at the wellknown Jodrell Bank Observatory nearManchester which has a dish diameterof 75 metres.

Viewed from Earth, the planet Venus

Modern radar systems are exemplified by this ‘AWAC’ airborne early warning aircraft.Multiple targets can be detected at extreme range

(Pic. 1.14 - P.D)

1. History of radar

is one of the brightest celestial bodies.However, the mysteries of our closeneighbour in the Solar System wereonly uncovered with the assistance ofradar. The surface of Venus is shroudedin dense clouds of vapour includingcarbon dioxide gas at pressures of 90bar and an average temperature of750 K.

Earth bound pulse radar measure-ments over an extended period of timewere used to calculate the radius of theorbit of Venus. Doppler shift measure-ments from the surface were used tocalculate the rate of rotation of theshrouded planet. The Venus ‘day’ wasfound to be 243 Earth days.

During the 1970’s, radar mapping ofthe planet’s surface by space probeuncovered surface features such ascraters.

Jodrell Bank - the observatory nearManchester which has a 75 metre

dish diameter(Pic. 1.15 - P.D)

Detection by radar is not always desirable. Huge sums of money have been spentreducing the radar signature of the F117 stealth fighter

(Pic. 1.16 - P.D)

9

10

Radar technology is part of oureveryday lives. The cavity magnetronis used in microwave ovens.Continuous wave (CW) radars are usedin automatic door detection and vehiclespeed measurement. Other well knowncivilian radar applications include airtraffic control, shipping and weatherradar.

Radar altimeters developed in the1930’s use a form of radar calledFM - CW or Frequency ModulatedContinuous Wave radar.

In the 1970’s, the same FM - CWmeasurement technique was used inthe production of the first radar leveltank gauge. Initially these radar leveltransmitters were used to measurepetroleum products in supertankers.Further developments of FM - CWlevel transmitters led to their use onshore based storage tanks in the mid1980’s. Originally these were expen-sive, high accuracy systems for fiscalmeasurement of petroleum products.

Later, lower accuracy FM - CW radartransmitters became available for theprocess industry.

In the late 1980’s, pulse radar leveltransmitters were developed for processmeasurement applications. The avail-ability of suitable crystals and solidstate components such as GaAs FEToscillators enabled cost effective radarlevel transmitters to enter the market.

In 1997 a significant improvementin the specification of radar level trans-mitters was achieved. VEGA producedthe world’s first two wire, loop pow-ered, intrinsically safe radar level trans-mitter. For the first time low cost, highspecification radar level transmittersbecame available.

It is likely that these advances willcontinue into the new millennium andthat radar level transmitters willbecome a commodity item in the sameway as differential pressure transmit-ters.

In the field of radar levelmeasurement, technologicaladvances have resulted intwo wire, intrinsically safetransmitters(Pic. 1.17 - Vega)

1. History of radar

11

A raw oscilloscope echo trace had to be interpreted by skilled operators using the Britishwar time Chain Home Low radar(Pic. 1.18 & 1.19 - I.W.M)

Comprehensive information is available on the PC echo trace of the latest two wire looppowered radar level transmitters(Pic. 1.20 - Vega Pic. 1.21 - Vega)

Comparing the old with the new










Inhalt

13

The velocity of light in free space is299,792,458 metres per second, butwho is timing? For the purposes of thecalculations in this book, we will call it300,000 kilometres per second or3 x 108 metres per second.

Maxwell’s theories of electro-magnetism were confirmed by theexperiments of Heinrich Hertz. Theseshow that all forms of electromagneticradiation travel at the speed of light infree space. This applies equally to longwave radio transmissions, microwaves,infrared, visible and ultraviolet lightplus X-rays and Gamma rays.

Maxwell showed that the velocity oflight in a vacuum in free space is givenby the expression :Examples :-

The original cavity magnetron hada wavelength of 9.87 centimetres.This corresponds to a frequency of3037.4 MHz (3.0374 GHz).

The frequency of a pulse radarlevel transmitter may be 26 GHzor 26 x 108 metres per second. The wavelength is 1.15 centimetres.

The electromagnetic waves have anelectrical vector E and a magnetic vec-tor B that are perpendicular to eachother and perpendicular to the directionof the wave. This will be discussed andillustrated further in the section onpolarization. The electrical vector hasthe major influence on radar applica-tions.

2. Physics of radar

Fig 2.1

ampl

itude

λ direction of wave

Electromagnetic waves

c velocity of electromagneticwaves in metres / second

f frequency of wave in second -1

λ wavelength in metres

[Eq. 2.1]

[Eq. 2.2]

co

c

)1

=

=

(µo x εo

f x λ

The velocity of an electromagneticwave is the product of the frequencyand the wavelength.

co velocity of electromagntic wavein a vacuum in metres / second

µµo the permeability of free space (4 π x 10 -7 henry / metre)

εεo the permittivity of free space(8.854 x 10 -12 farad / metre)

14

108 107 106 105 104 103 102 101 100 10-1 10-2 10-3 10-4

101 102 103 10

100 MHz 1 GHz 10 GHz 100 GHz

3 m 0.3 m 3 cm 3 mm

4 105 106 107 108 109 1010 1011 1012

infraradio waveselectric waves

The microwave frequencies of the electromagnetic spectrum.Radar level transmitters range between 5.8 GHz (5.2cm) and 26 GHz (11.5mm)

The Electromagnetic spectrum

15

2. Physics of radar

10-5 10-6 10-7 10-8 10-9 10-10 10-11 10-12 10-13 10-14 10-15 10-16 m

1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 Hz

gamma raysX raysultra violetred

Fig 2.2 Electromagnetic spectrum.All electromagnetic waves travel at the speed of light in free space. This spectrumshows the range of frequencies and wavelengths from electric waves togamma rays

PermittivityIn electrostatics, the force between

two charges depends upon the magni-tude and separation of the charges andthe composition of the mediumbetween the charges. Permittivity ε isthe property of the medium that effectsthe magnitude of the force. The higherthe value of the permittivity, the lowerthe force between the charges. Thevalue of the permittivity of freespace (in a vacuum) εo, is calculatedindirectly and empirically to be:8.854 x 10-12 farad / metre.

Relative permittivity ordielectric constant εεr

The ratio of the permittivity of amedium to the permittivity of freespace is a dimensionless propertycalled ‘relative permittivity’ or ‘dielec-tric constant’. For example, at 20° Cthe relative permittivity of air is closeto that of a vaccum and is only about1.0005 whereas the relative permittivi-ty of water at 20° C is about 80.(Dielectric constant is also widelyknown as DK.)

The value of the dielectric constantof the product being measured is veryimportant in the application of radar tolevel measurement. In non-conductiveproducts, some of the microwave ener-gy will pass through the product andthe rest will be reflected off the surface.

This feature of microwaves can beused to advantage or, in some circum-stances, it can create a measurementproblem.

Permeability µ and relativepermeability µr

The magnetic vector, B, of an elec-tromagnetic wave also has an influenceon the velocity of electromagneticwaves. However, this influence is neg-ligible when considering the velocity ingases and vapours which are non-mag-netic. The relative permeability of theproduct being measured has no signifi-cant effect on the reflected signal whencompared with the effects of the rela-tive permittivity or dielectric constant.For the non-magnetic gases above theproduct being measured, the value ofthe relative permeability, µr = 1.

Frequency, velocity and wave-length

As we have already stated, the fre-quency (f), velocity (c) and wavelength(λ) of the electromagnetic waves arerelated by the equation c = f x λ.

The frequency remains uninfluencedby changes in the propagation medium.However, the velocity and wavelengthcan change depending on the electricalproperties of the medium in which theyare travelling. The speed of propaga-tion can be calculated using equation2.3.

16

c velocity of electromagnetic wavein the medium in metres/second

co velocity of electromagneticwaves in free space

µ r the relative permeability(µ medium / µo)

εεr the relative permittivity

[Eq. 2.3]

c)

co

=(µr x εr

2. Physics of radar

Changes in the wavelength andvelocity of microwaves are apparent incertain radar level applications.Changes in temperature, pressure andgas composition have a small effect onthe running time of microwavesbecause the dielectric constant of thepropagation medium is altered to agreater or lesser extent. This is dis-cussed in detail later.

Radar level transmitters can be usedto measure conductive liquids throughlow dielectric ‘windows’ such as glass,polypropylene and PTFE. The opti-mum thickness of the low dielectricwindow is a half wavelength or multi-ple of half wavelength.

For example, polypropylene has adielectric constant εr of 2.3 and thehalf wavelength at a frequency of 5.8GHz is 17 mm compared with a halfwavelength of about 26 mm in a vacu-um. It follows that the speed of

microwaves in polypropylene is abouttwo thirds of the speed in air.

As with low dielectric windows,non-conductive, low dielectric constantliquids may absorb more power thanthey reflect from the surface. Thevelocity of the microwaves within theliquid is slower than in the vapourspace above.

For example, if there is about 0.5metres of solvent in the bottom of ametallic vessel, a radar level transmittermay see a larger echo from the vesselbottom than from the product. Thislarge echo will appear to be furtheraway than it really is because the run-ning time within the solvent is slower.For this reason, special considerationsmust be made within the echo process-ing software to ensure that the radarfollows the solvent level and does notfollow the vessel bottom as it apparent-ly moves away!

Empty vessel: large echofrom metalbottom

As the vessel fills withsolvent two echoesare received. Theecho from the vesselbottom appearsfurther away becausethe running time ofthe microwaves insolvent is slower

Fig 2.3 - Effect of dielectric constant on the running time of a microwave radar

solvent echo

17

18

Effects on the propagationspeed of microwaves

Microwave radar level transmitterscan be applied almost universallybecause, as a measurement technique,they are virtually unaffected by processtemperature, temperature gradient, vac-uum and normal pressure variations,gas or vapour composition and move-ment of the propagation medium.

However, changes in these processconditions do cause slight variations inthe propagation speed because thedielectric constant of the propagationmedium is altered.

Calculating the propagationspeed of microwaves

The temperature, pressure and thegas composition of the vapour space allhave an effect on the dielectric constantof the propagation medium throughwhich the microwaves must travel.This in turn affects the propagationspeed or running time of the instru-ment.

The dielectric constant or relativepermittivity can be calculated asfollows :

The same effect can be experienced when looking at interface detection usingguided microwave level transmitters to detect oil and water or solvent and aqueousbased liquids.

reference echo(water without oil)

water echooil echo

εεr calculated dielectric constant(relative permittivity)

εεrN dielectric constant of gas/vapourunder normal conditions (temperature 273 K, pressure 1 barabsolute)

θN temperature under normalconditions, 273 Kelvin

PN pressure under normalconditions, 1 bar absolute

θ process temperature in Kelvin

P process pressure in bar absolute

Fig 2.4 Oil/water interfacedetection using aguided microwavelevel transmitter. Notethat the water echohas a reduced ampli-tude and appears to befurther away. Therunning time ofmicrowaves in oil isslower than in air

εr = + x1 θN x P

θ x PN

(εrN - 1)[Eq. 2.4]

2. Physics of radar

From equation 2.4 and equation 2.3,we can calculate the percentage errorcaused by variations in the dielectricconstant of different gases and vapoursand the relative effects of changes inprocess temperature and pressure.

Gases and vapoursBy definition, the dielectric constant

in a vacuum is equal to 1.0. The dielec-tric constants of the gases and vapoursthat may be present above the product

differ but they have only a very smalleffect on the accuracy of radar.

Radar level transmitters are usuallycalibrated in air. For this reason, thefollowing tables show

1. Dielectric constant of different gasesat normal temperature and pressure(273K, 1 Bar A)

2. Percent error in the running time inthe gases compared with air

Gas / VapourεεrN (dielectric

constant at normalconditions)

% Error from air (atnormal temperature

and pressure)

Vacuum 1.0000 + 0.0316Air 1.000633 0.0

Argon 1.000551 + 0.0041

Ammonia / NH 3 1.006976 + 0.3154

Hydrogen Bromide HBr 1.002994 - 0.1178Hydrogen Chloride HCl 1.004078 - 0.1717

Carbon Monoxide / CO 1.000692 - 0.00295Carbon Dioxide / C0 2 1.000985 - 0.0176

Ethane / C 2H6 1.001503 - 0.0434

Ethylene / C 2H4 1.001449 - 0.0407Helium 1.000072 + 0.0280

Hydrogen / H 2 1.000275 + 0.0179

Methane / CH 4 1.000878 - 0.0122Nitrogen / N 2 1.000576 + 0.00285

Oxygen / O 2 1.000530 + 0.0052

Table 2.1 The dielectric constants under normal conditions, εεrN and the error caused bythe dielectric constant of typical process gases under normal conditions

19

20

Temperature

Fig 2.5 Temperature effect on radar measurement of air at a constant pressure of 1 BarA

High temperature or large temperature gradients have very little effect on thetransit time of microwaves within an air or vapour space. At a temperature of2000° C the variation is only 0.026% from the measurement value at 0° C. Radarlevel transmitters with air or nitrogen gas cooling are used on molten iron and steelapplications.

Temperature in ° C

0

0.005

0.0

0.01

0.015

0.02

0.025

0.03

250 500 750 1000 1250 1500 1750 2000

% e

rror

21

2. Physics of radar

Fig 2.6 The influence of pressure on radar measurement in air at a constant temperatureof 273 K

Pressure does have a small but more significant influence on the velocity ofelectromagnetic waves. At a pressure of 30 Bar, the error is only 0.84%. Howeverthis becomes more significant and at a pressure of 100 Bar there is a velocitychange of 2.8%. If the pressure is varying constantly between atmospheric pressureand 100 Bar, the velocity variations can be compensated using a pressure transmit-ter.

Pressure

0

0

2

4

6

8

10

50 100 150 200 250 300 350 400

Pressure in Bar (absolute)

% e

rror

22

In the preceding equations, we haveassumed that the microwaves aretravelling in ‘free space’ in a vacuum.However, in practice the proximityof metallic vessel walls and otherstructures will have an influence onthe propagation velocity of themicrowaves. This is particularly truewhen microwave radar level transmit-ters are fitted inside bypass tubes orstilling tubes or when a horn antenna isfitted with a waveguide extension.

When microwaves are propagating

within a metallic tube the running timeappears to slow down because themicrowaves travel further bouncingoff the inside wall of the tube andcurrents are set up on the inside surfaceof the tube.

This effect is discussed in moredetail in the chapters on antennas andmechanical installations. The wave-guide effect can be compensated duringcalibration and the use of stilling tubesand bypass tubes can be beneficial insome level applications.

Conductive productsUsing a spark gap transmitter,

Heinrich Hertz demonstrated that elec-tromagnetic waves could be reflectedoff metallic objects and objects with arelatively high dielectric constant.

In the same way, radar can easilymeasure conductive aqueous liquidssuch as acids and caustic and otherconductive products ranging frommolten metal to saturated spent grain inthe brewing process.

When microwaves from a radar hit aconductive surface the electrical field Eis short circuited. The resultant currentin the conductive product causes themicrowaves to be re-transmitted orreflected from the surface.

Radar level transmitters have noproblem in measuring conductive liq-uids and solids because the microwaveswith frequencies between 5.8 GHz and26GHz are readily reflected off a con-ductive surface producing relativelylarge echoes.

Non-conductive productsIf a liquid or solid is non-conductive,

the value of the dielectric constant (rela-tive permittivity εr) becomes moreimportant. The theoretical amount ofreflection at a dielectric layer can be cal-culated using equation 2.5

Waveguides, stilling tubes & bypass tubes

Reflection of electromagnetic waves

Electromagnetic waves exhibit the same properties as light.

· Reflection · Refraction· Polarization · Interference· Diffraction

23

2. Physics of radar

TolueneSolvent with a low dielectric constant,

εr = 2.4

AcetoneSolvent with a dielectric constant,εr = 20

Fig 2.7 Reflected radar power depends upon the dielectric constant of the productbeing measured

Transmitted power: W1

Reflected power: W2

Dielectric constant: εrThen the percentage of reflectedpower at the dielectric layer,

Π x

100

% p

ower

ref

lect

ed

0

0

20

40

60

80

100

10 20 30 40 50 60 70 80

Π

Π [Eq. 2.5]

=

=

1-

W2

W1

4 x εr

1 + εr( )2

Typical examples are as follows:

4.46% power is reflected 40 % power is reflected

Π = 1- 4 x 4 x2.4

1 + (2.4)(

(

)

)2

Π = 1- 20

1 + (20)(

(

)

)2

Dielectric constant, εr

24

In radar level measurement the reflected energy from a product surface becomesmore critical at a dielectric constant (εr) of less than 5. The following graph showsthis important region.

Most electrically conductive products or products with a dielectric constant ofmore than 1.5 can be measured using microwave radar level transmitters. Stillingtubes can be used to concentrate the microwaves for lower dielectric constantproducts.

Fig 2.8 Reflected radar power depends upon the dielectric constant of the product beingmeasured. This graph shows the critical region where care must be taken overchoice of radar antenna

Π x

100

% p

ower

ref

lect

edLo

ssL,

dB

1.0

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

0

5

10

15

20

1.5 2.0 3.0 3.5 4.0 4.5 5.02.5

Fig 2.9 Reflection loss in dB: loss L = 10 log ΠΠ

- 60

- 40

- 20

- 10

0



2. Physics of radar

Electromagnetic waves have anelectrical vector E and magnetic vectorB that are in phase but perpendicular toeach other. The direction of propaga-tion of the waves is perpendicular tothe electrical and magnetic vectors asshown in the diagram below.

Polarization defines the orientationof the electromagnetic waves and refersto the direction of the electrical vectorE. Most process radar level transmittersexhibit linear polarization as in the dia-

gram. The direction of the linear polar-ization is set by the orientation of thesignal coupler from the microwavemodule. The properties of the polariza-tion of microwaves can be important inthe application of radar to level mea-surement.

In television and microwave com-munications, linear polarization is alsoreferred to as horizontal or verticalpolarization depending on the relativeorientation of the aerials or antennas.

Fig 2.10 Diagram showing linear polarization and the relative orientation of the electricvector E, the magnetic vector B and the direction of propagation of themicrowaves

direction of wave

E

B

Polarization

25

26

Fig 2.11 Circular polarization involves rotation of the electrical and magnetic vectorsthrough 360° within a wavelength

Another form of polarization iselliptical polarization. A specific formof elliptical polarization is circularpolarization where the electrical vectorE and magnetic vector B rotate through360° within the space of a single wave-length, when a linear or circular polar-ized signal is reflected the direction ofpolarization is reversed. With circularpolarization it is possible to use thereversal of polarization to distinguishbetween a direct echo and an echo thathas made two reflections.

Circular polarization can also beused in search radars to separate thereflections from aircraft or ships frominterference echoes from rain. Thealmost spherical shape of the rain dropscauses a definite reversal of polariza-tion which can be easily rejected by thereceiving antenna. However, the scat-tered reflections from the ship or air-craft provide roughly equal amounts ofreversed and un-reversed energy thatenables detection.

λλ

27

2. Physics of radar

The linear polarization that is com-mon with process radar level transmit-ters can be used to minimise the effectsof false echo returns from the internalstructure of a process vessel. Thesefalse echoes could be reflected fromprobes, welds, agitators and baffles.

In some applications, the effect offalse echoes within a vessel can be sig-nificantly reduced by rotating the radarin the connection flange or boss. Theprinciple is illustrated below anddetailed in the section on mechanicalinstallations in Chapter 6.

Fig 2.12 If a metallic or high dielectric object is orientated in the same plane as theelectrical vector of the polarized microwaves, the radar level transmitter willreceive a large amplitude echo

Fig 2.13 If the same object is orientated at right angles to the plane of the electrical vector,the received echo will have a smaller amplitude

Large echo

Small echo

Direction of wave

E

B

Polarization can be used to reduce the amplitude of false echoes

Direction of wave

B

E

Beam angle is often discussed inrelation to radar transmitters. This cangive the impression that the radarantenna can direct a finely focusedbeam towards the target. Unfortunatelythis is not the case.

In practice, although they aredesigned to produce a directed beam, aradar antenna radiates some energy inall directions. As well as the main lobe

which accounts for most of the radiatedpower, there are also weaker side lobesof energy. This phenomenon is caused,in part, by diffraction. In addition tothis, destructive interference causes thenull points or notches that form thecharacteristic side lobes.

Chapter 5 provides a detailed expla-nation of beam angles, side lobes andtypes of antennas.

Fig 2.14 The lobe structure of antenna beams is caused by diffraction and destructiveinterference

Fig 2.15 Refraction & reflection

RefractionIn the same way as light is refracted

at an air/glass or air/water interface,microwaves are refracted when theyencounter a change in dielectric. This could be a low dielectric window(PTFE/glass/polypropylene) or a non-conductive low dielectric liquid such asa solvent.

The angle of refraction depends onthe angle of the incident wave and alsoon the ratio of the dielectric constantsat the interface.

It is possible to utilise the refractiveproperties of electromagnetic waves toconstruct a dielectric lens that willfocus microwaves.

Diffraction

main lobeside lobes

antenna

a a

B

microwave

interface

refracted energy

dielectric window / product

reflected energy

28

29

2. Physics of radar

Problematic interference effects are caused primarily by the inadvertent mixingof signals that are out of phase. The microwave signals have a sinusoidal wave-form.

Fig 2.16 In this illustration both of the sine waves have an identical frequency andamplitude but the second wave has a 45° phase lag

Interference - Phase

Phase angle

45°

Interference can be ‘constructive’ where in-phase signals produce a signal with ahigher amplitude or it can be destructive where signals that are 180° out of phaseeffectively cancel each other out.

signals in-phase

180° out of phase

constructive interference

destructive interference

Fig 2.17 Illustration of constructive and destructive interference

30

Microwaves can manifest interfer-ence effects in exactly the same way aslight. Potentially this can cause mea-surement problems. The causes ofinterference should be understood andavoided by design and installation con-siderations.

The wrong choice of antenna, instal-lation of an antenna up a nozzle, posi-tioning transmitters too close to vesselwalls or other obstructions can all lead

to interference of the signal. The chap-ter on mechanical installation shouldhelp a radar level user to avoid thispotential problem.

However, we use destructive inter-ference to our advantage when weapply pulse radar level measurementthrough a low dielectric ‘window’ tomeasure conductive or high dielectricliquids.

Interference

Fig 2.18 Interference caused by positioning an antenna too close to the vessel wall. If aradar level transmitter is installed too close to the vessel wall it is possible thatinterference will occur. With indirect reflection A B’ B’’ C, the phase may bealtered by 180° when compared with the direct reflection A B C. For this reasonthe microwaves may partially cancel out due to destructive interference

+ =

C

A

B’

B B”

31

2. Physics of radar

The thickness of the dielectric win-dow must be a half wavelength of thewindow material. When the half wave-length is used, there is destructive inter-ference between the reflection off thetop surface of the window and thereflection off the internal second surfaceof the window.

There is a 180° phase shift betweenthese reflections and they cancel each

other out. This type of installationis explained more fully in Chapter 6on the mechanical installations ofradar level transmitters together witha table showing the optimum thicknessof most important plastics and glasseswhich are suitable for penetration withradar sensors.

Fig 2.19 Destructive interference is a benefit when using pulse radar to measure througha low dielectric window. The reflection from the top surface and the reflectionfrom the internal second surface cancel each other if the thickness is a halfwavelength

emitted wavereflection withphase shift from topsurface

plastic vessel ceiling

reflection withoutphase shift frominternal surface

D

emitted wave

reflection with phase shift offtop surface of window

reflection without phase shiftoff internal face of window










Contents

In continuous wave or CW Radar, acontinuous unmodulated frequency istransmitted and echoes are receivedfrom the target object. If the targetobject is stationary, the frequency ofthe return echoes will be the same asthe transmitted frequency. The range ofthe object cannot be measured.

However, the frequency of the returnsignal from a moving object is changeddepending on the speed and directionof the object. This is the well known‘doppler effect’. The doppler effect isapparent when the siren note of anemergency vehicle changes as it speedspast a pedestrian. The pitch of the siren

note is higher as it approaches the lis-tener and lower as it recedes. Thedoppler effect is also used byastronomers to monitor the expansionof the Universe. By measuring the ‘redshift’ of the spectrum of distant starsand galaxies the rate of expansion canbe measured and the age of distantobjects can be estimated.

In the same way, when an object thathas been illuminated by a CW Radarapproaches the transmitter, the frequen-cy of the return signal will be higherthan the transmitted frequency. Theecho frequency will be lower if theobject is moving away.

33

3. Types of radar

Fig 3.1 CW radar uses doppler shift to derive speed measurement

target velocity v

received frequency ft + fdp

transmitted frequency ft, wavelength λ

1a. CW, continuous wave radar

34

1b. CW wave-interference radar or bistatic CW radar

1c. Multiple frequency CW radar

In Fig 3.1, the aircraft is travellingtowards the CW radar. Therefore thereceived frequency is higher thanthe transmitted frequency and the signof fdp is positive. If the aircraft wastravelling away from the radar at the

same speed, the received frequencywould be ft - fdp.

The velocity of the target in thedirection of the radar is calculated byequation 3.1

c is the velocity of microwavesv is the target velocityft is the frequency of the

transmitted signal fdp is the doppler beat frequency

which is proportional to velocityft+fdp is received frequency. The sign

of fdp depends upon whether thetarget is closing or receding

Standard continuous wave radar isused for speed measurement and, asalready explained, the distance to a sta-tionary object can not be calculated.However, there will be a phase shiftbetween the transmitted signal and thereturn signal.

If the starting position of the objectis known, CW radar could be used todetect a change in position of up to halfwavelength (λ/2) of the transmittedwave by measuring the phase shift ofthe echo signal. Although furthermovement could be detected, the range

would be ambiguous. With microwavefrequencies this means that the usefulmeasuring range would be very limited.

If the phase shifts of two slightlydifferent CW frequencies are measuredthe unambiguous range is equal to thehalf wavelength (λ/2) of the differencefrequency. This provides a usable dis-tance measurement device.

However, this technique is limited tomeasurement of a single target.Applications include surveying andautomobile obstacle detection.

We have already mentioned that CWradar was used in early radar detectionexperiments such as the famousDaventry experiment carried out byRobert Watson - Watt and his col-leagues. In this case, the transmitterand receiver were separated by a con-siderable distance. A moving objectwas detected by the receiver becausethere was interference between the fre-

quency received directly from thetransmitter and the doppler shifted fre-quency reflected off the target object.Although the presence of the object isdetected, the position and speed cannotbe calculated.

In essence, this is what happenswhen a low flying aircraft interfereswith the picture on a television screen.See Fig 3.2.

v2

= =λ x fdp c x fdp

2 x ft

[Eq. 3.1]

3. Types of radar

targ

et

tran

smitt

erte

levi

sion

inte

rfer

ence

refle

cted

sig

nal

(dop

pler

shi

ft)tr

ansm

itted

sign

al in

dire

ct

tran

smitt

ed s

igna

l dire

ct

Fig

3.2

The

eff

ect o

f lo

w f

lyin

g ai

rcra

ft o

n te

levi

sion

rec

eptio

n is

sim

ilar

to th

e m

etho

d of

dete

ctio

n by

CW

wav

e-in

terf

eren

ce r

adar

35

Single frequency CW radar cannotbe used for distance measurementbecause there is no time reference markto gauge the delay in the return echofrom the target. A time reference markcan be achieved by modulating the fre-quency in a known manner.

If we consider the frequency of thetransmitted signal ramping up in alinear fashion, the difference betweenthe transmitting frequency and thefrequency of the returned signal will beproportional to the distance to thetarget.

If the distance to the target is R,and c is the speed of light, then thetime taken for the return journey is:-

We can see from Fig. 3.3 that ifwe know the linear rate of change ofthe transmitted signal and measure thedifference between the transmitted andreceived frequency fd, then we cancalculate the time ∆t and hence derivethe distance R.

36

time

fd

transm

itted fre

quency

receive

d frequency

freq

uenc

y

Fig 3.3 The principle of FM - CW radar

2. FM-CW, frequency modulated continuous wave radar

∆t = 2 x Rc

∆t =

∆t

2 x Rc

[Eq. 3.2]

37

3. Types of radar

FM - CW wave forms transmitted frequencyreceived frequency

Fig 3.6 Saw tooth waveMost commonly usedon most FM - CWprocess radar leveltransmitters

Fig 3.5 Triangular waveUsed on FM - CWradar transmitters

frequency

4.4GHz

4.2GHz

10 GHz

9 GHz

frequency

time

time

time

frequency

Fig 3.4 Sine waveCommonly used on air-craft radio altimetersbetween 4.2 and4.4 GHz

In practice, the FM - CW signal hasto be cyclic between two different fre-quencies. Radio altimeters modulatebetween 4.2 GHz and 4.4 GHz. Radarlevel transmitters typically modulatebetween about 9 GHz and 10 GHz or

24 GHz and 26 GHz.The cyclic modulation of FM - CW

radar transmitter takes different forms.These are sinusoidal, saw tooth ortriangular wave forms.

38

If we look at a triangular waveform we can see that there is an inter-ruption in the output of the differencefrequency , fd. In practice, the receivedsignal is heterodyned with part of thetransmitted frequency to produce thedifference frequency which has a posi-

tive value independent of whether themodulation is increasing or decreasing.

The diagram below makes theassumption that the target distance isnot changing. If the target is moving,there will be a doppler shift in the dif-ference frequency.

frequency

time

time

difference

frequency

fd

Fig 3.7 & 3.8 The change in direction between the ramping up and down of the frequencycreates a short break in the measured value of the difference frequency.This has to be filtered out. The transmitted frequency is represented by thered line and the received frequency is represented by the dark blue line.The difference frequency is shown in light blue on the bottom graph

Pulse radar is and has been usedwidely for distance measurement sincethe very beginnings of radar technolo-gy. The basic form of pulse radar is apure time of flight measurement. Shortpulses, typically of millisecond ornansecond duration, are transmittedand the transit time to and from the tar-get is measured.

The pulses of a pulse radar are notdiscrete monopulses with a single peak

of electromagnetic energy, but are infact a short wave packet. The numberof waves and length of the pulsedepends upon the pulse duration andthe carrier frequency that is used.

These regularly repeating pulseshave a relatively long time delaybetween them to allow the return echoto be received before the next pulse istransmitted.

39

3. Types of radar

The inter pulse period (the timebetween successive pulses) t is theinverse of the pulse repetitionfrequency fr or PRF. The pulse durationor pulse width, τ, is a fraction of theinter pulse period.

The inter pulse period t effectivelydefines the maximum range of theradar. ExampleThe pulse repetition frequency(PRF) is defined as

If the pulse period t is 500 microsec-onds, then the pulse repetition frequen-cy is two thousand pulses per second.In 500 microseconds, the radar pulseswill travel 150 kilometres. Consideringthe return journey of an echo reflectedoff a target, this gives a maximum the-oretical range of 75 kilometres.

If the time taken for the returnjourney is T, and c is the speed of light,then the distance to the target is

3. Pulse radar

Fig 3.9 Basic pulse radar

t

τ

3rd pulse

Transmitted pulses

2nd pulse 1st pulse

1t

T x c2

fr = R =[Eq. 3.3]

a. Basic pulse radar

The pulses transmitted by a standardpulse radar can be considered as a veryshort burst of continuous wave radar.There is a single frequency with nomodulation on the signal for the dura-tion of the pulse.

If the frequency of the waves of thetransmitted pulse is ft and the target ismoving towards the radar with velocityv, then, as with the CW radar alreadydescribed, the frequency of the returnpulse will be ft + fdp , where fdp is thedoppler beat frequency. Similarly, thereceived frequency will be ft - fdp if thetarget is moving away from the radar.

Therefore, a pulse doppler radar canbe used to measure speed, distance anddirection.

The ability of the pulse dopplerradar to measure speed allows the sys-tem to ignore stationary targets. This isalso commonly called ‘moving targetindication’ or MTI radar.

In general, an MTI radar has accu-rate range measurement but imprecisespeed measurement, whereas a pulsedoppler radar has accurate speed mea-surement and imprecise distance mea-surement.

The velocity of the target in thedirection of the radar is calculated inequation 3.4:

This is the same calculation as forCW radar. The distance to the target iscalculated by the transit time of thepulse, equation 3.3.

As well as being used to monitorcivil and military aircraft movements,pulse doppler radar is used in weatherforecasting. A doppler shift is measuredwithin storm clouds which can be dis-tinguished from general ground clut-ter. It is also used to measure theextreme wind velocities within a torna-do or ‘twister’.

40

b. Pulse doppler radar

c

R

2

2

=

=

=λ x fdp c x fdp

2 x ft

T x c

[Eq. 3.4]

[Eq. 3.3]

3. Types of radar

Fig

3.1

0 P

ulse

dop

pler

rad

ar p

rovi

des

targ

et s

peed

, dis

tanc

e an

d di

rect

ion

f t+

f dp

f t

R

Pul

se d

opp

ler

rada

r

41

42

With pulse radar, a shorter pulseduration enables better target resolutionand therefore higher accuracy.However, a shorter pulse needs a sig-nificantly higher peak power if therange performance has to be main-tained. If there is a limit to the maxi-mum power available, a short pulsewill inevitably result in a reducedrange.

With limited peak power, a longerpulse duration, τ , will provide more

radiated energy and therefore range but(with a standard pulse radar) at theexpense of resolution and accuracy.

Pulse compression within a ‘Chirp’radar is a method of achieving theaccuracy benefits of a short pulse radartogether with the power benefits ofusing a longer pulse. Essentially, Chirpradar is a cross between a pulse radarand an FM - CW radar.

Fig 3.11 Chirp radar wave form. Chirp is a cross between pulse and FM - CW radar

c. Pulse compression and ‘Chirp’ radar

f1

f2

t1 t2

time

time

freq

uenc

yam

plitu

de

ττ

43

3. Types of radar

Each pulse of a Chirp radar has lin-ear frequency modulation and a con-stant amplitude.

The echo pulse is processed througha filter that compresses the echo bycreating a time lag that is inversely

proportional to the frequency.Therefore, the low frequency that

arrives first is slowed down the mostand the subsequent higher frequenciescatch up producing a sharper echo sig-nal and improved echo resolution.

Another method of echo compres-sion uses binary phase modulationwhere the transmitted signal is special-ly encoded with segments of the pulseeither in phase or 180° out of phase.The return echoes are decoded by a fil-ter that produces a higher amplitudeand compressed signal.

The name ‘Chirp’ radar comes fromthe short rapid change in frequency ofthe pulse which is analogous to thechirping of a bird song.

The above methods of radar detec-tion are used widely in long range dis-tance or speed measurement. In thenext chapter we look at which of thesemethods can be applied to the uniqueproblems involved in measuring liquidor solid levels within process vesselsand silos.

Pulse compression of chirp radar echo signal

Long frequency modulated echo pulse

Tim

e la

g

Frequency

Filter

Compressed signal

Fig 3.12 Pulse compression of chirp radar echo signal

45

Part III

Radar level measurementRadar antennas

Radar level installations










Contents

The benefits of radar as a level mea-surement technique are clear.

Radar provides a non-contact sensorthat is virtually unaffected by changesin process temperature, pressure or thegas and vapour composition within avessel.

In addition, the measurement accu-racy is unaffected by changes in densi-ty, conductivity and dielectric constantof the product being measured or by airmovement above the product.

The practical use of microwaveradar for tank gauging and process ves-sel level measurement introduces aninteresting set of technical challengesthat have to be mastered.

If we consider that the speed of lightis approximately 300,000 kilometresper second. Then the time taken for

a radar signal to travel one metreand back takes 6.7 nanoseconds or0.000 000 006 7 seconds.

How is it possible to measure thistransit time and produce accurate ves-sel contents information?

Currently there are two measure-ment techniques in common use forprocess vessel contents measurement.They are frequency modulated continu-ous wave (FM - CW) radar and PULSEradar

In this chapter we explain FM - CWand PULSE radar level measurementand compare the two techniques. Wediscuss accuracy and frequency consid-erations and explore the technicaladvances that have taken place inrecent years and in particular two wire,loop powered transmitters.

47

4. Radar level measurement

radar_applied_to_level_rb.qxd 15.01.2007 18:46 Seite 47

The FM - CW radar measurementtechnique has been in use since the1930's in military and civil aircraftradio altimeters. In the early 1970's thismethod was developed for marine usemeasuring levels of crude oil in super-tankers. Subsequently, the same tech-nique was used for custody transferlevel measurement of large land basedstorage vessels. More recently, FM -CW transmitters have been adapted forprocess vessel applications.

FM - CW, or frequency modulatedcontinuous wave, radar is an indirectmethod of distance measurement. Thetransmitted frequency is modulatedbetween two known values, f1 and f2,and the difference between the trans-mitted signal and the return echosignal, fd, is measured. This differencefrequency is directly proportional to the

transit time and hence the distance.(Examples of FM - CW radar leveltransmitters modulation frequencies are8.5 to 9.9 GHz, 9.7 to 10.3 GHz and 24to 26 GHz).

The theory of FM - CW radar issimple. However, there are many prac-tical problems that need to beaddressed in process level applications.

An FM - CW radar level transmitterrequires a voltage controlled oscillator,VCO, to ramp the signal between thetwo transmitted frequencies, f1 and f2.It is critical that the frequency sweep iscontrolled and must be as linear as pos-sible. A linear frequency modulation isachieved either by accurate frequencymeasurement circuitry with closed loopregulation of the output or by carefullinearisation of the VCO output includ-ing temperature compensation.

48

FM-CW, frequency modulated continuous wave fr

eque

ncy

f2

f1t1

∆ t

fd

time

Transmitted signal

Receivedsignal

Fig 4.1 The FM - CW radar technique is an indirect method of level measurement.fd is proportional to ∆∆t which is proportional to distance


49


��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�� !

��"��

��

��!

��!

��#��!

��!

��#��!

Fig

4.2

Typ

ical

blo

ck d

iagr

am o

f F

M -

CW

rada

r. A

very

acc

urat

e lin

ear

swee

p is

req

uire

d

FM - CW block diagram (Fig 4.2)The essential component of a fre-

quency modulated continuous waveradar is the linear sweep control cir-cuitry. A linear ramp generator feeds avoltage controller which in turn rampsup the frequency of the VoltageControlled Oscillator. A very accuratelinear sweep is required. The outputfrequency is measured as part of theclosed loop control.

The frequency modulated signal isdirected to the radar antenna and

hence towards the product in the ves-sel. The received echo frequencies aremixed with a part of the transmissionfrequency signal. These difference fre-quencies are filtered and amplifiedbefore Fast Fourier Transform (FFT)analysis is carried out. The FFTanalysis produces a frequency spec-trum on which the echo processing andecho decisions are made.

Simple storage applications usuallyhave a large surface area with very lit-tle agitation, no significant false echoesfrom the internal structure of the tankand relatively slow product movement.These are the ideal conditions forwhich FM - CW radar was originallydeveloped.

However, in process vessels there ismore going on and the problemsbecome more challenging.

Low amplitude signals and falseechoes are common in chemical reac-tors where there is agitation and lowdielectric liquids.

Solids applications can be trouble-some because of the internal structureof the silos and undulating product sur-faces which creates multiple echoes.

An FM - CW radar level sensortransmits and receives signals simulta-neously.

50

Pic 2 Typical glass linedagitated processvessel. A radarmust be able tocope with variousfalse echos fromagitatior bladesand baffles


51


In an active process vessel, the vari-ous echoes are received as frequencydifferences compared with the frequen-cy of the transmitting signal. These fre-quency difference signals are receivedby the antenna at the same time. Theamplitude of the real echo signals aresmall compared with the transmittedsignal. A false echo from the end of theantenna may have a significantly high-er amplitude than the real level echo.The system needs to separate and iden-tify these simultaneous signals beforeprocessing the echoes and making anecho decision.

The separation of the variousreceived echo frequencies is achievedusing Fast Fourier Transform (FFT)analysis. This is a mathematical proce-

dure which converts the jumbled arrayof difference frequencies in the timedomain into a frequency spectrum inthe frequency domain.

The relative amplitude of each fre-quency component in the frequencyspectrum is proportional to the size ofthe echo and the difference frequencyitself is proportional to the distancefrom the transmitter.

The Fast Fourier Transform requiressubstantial processing power and is arelatively long procedure.

It is only when the FFT calculationsare complete that echo analysis can becarried out and an echo decision can bemade between the real level echo and anumber of possible false echoes.

Fig 4.3a FM - CW radar level transmitters in an active process vessel

Transmitted signal

f2

fd1, -f-fd2d2, -fd3, -fd4, -fd5

f1t1

Real echo signal

False echo signals


52

Fig 4.3b combined echo frequencies are received simultaneously

Fig 4.3c The individual frequencies must be separated fromthe simultaneously received jumble of frequencies

Sig

nal

am

plit

ud

eS

ign

al a

mp

litu

de

Mixture of frequencies received by FM - CW radar

Combination of mixed difference frequencies received by FM - CW radarIndividual difference frequencies fd1, ffd2d2, fd3, are shown

fd1, fd2, fd3, fd4, fd5 etc combined


Complex process vessels and solidsapplications can prove too difficult forsome FM - CW radar transmitters.Even a simple horizontal cylindricaltank can pose a serious problem. Thisis because a horizontal tank producesmany large multiple echoes that arecaused by the parabolic effect of thecylindrical tank roof. Sometimes theamplitudes of the multiple echoes are

higher than the real echo. The proces-sors that carry out the FFT analysis areswamped by different amplitude sig-nals across the dynamic range all at thesame time. As a result, the FM - CWradar cannot identify the correct echo.

As we shall see, this problem doesnot affect the alternative pulse radartechnique.

53


Fig 4.4 FM - CW frequency spectrum after Fast Fourier transform. The Fast Fouriertransform algorithm converts the signals from the time domain into the frequencydomain. The result is a frequency spectrum of the difference frequencies. Therelative amplitude of each frequency component in the spectrum is proportional tothe size of the echo and the difference frequency itself is proportional to thedistance from the transmitter. The echoes are not single frequencies but a spanof frequencies within an envelope curve

Frequency spectrum echoesEach echo is within an envelope curve of frequencies

amp

litu

de

frequency


PULSE radar level transmitters

54

Pulse radar level transmitters pro-vide distance measurement based onthe direct measurement of the runningtime of microwave pulses transmittedto and reflected from the surface of theproduct being measured.

Pulse radar operates in the timedomain and therefore it does notrequire the Fast Fourier transform(FFT) analysis that characterizes FM -CW radar.

As already discussed, the runningtime for a distance of a few metres ismeasured in nanoseconds. For this rea-son, a special time transformation pro-

cedure is required to enable these shorttime periods to be measured accurately.The requirement is for a ‘slow motion’picture of the transit time of themicrowave pulses with an expandedtime axis. By slow motion we meanmilliseconds instead of nanoseconds.

Pulse radar has a regular and period-ically repeating signal with a high pulserepetition frequency (PRF). Using amethod of sequential sampling, theextremely fast and regular transit timescan be readily transformed into anexpanded time signal.

Fig 4.5 Pulse radar operates purely within the time domain. Millions of pulses aretransmitted every second and a special sampling technique is used to produce a‘time expanded’ output signal


A common example of this principleis the use of a stroboscope to slowdown the fast periodic movements ofrotating or reciprocating machinery.

Fig 4.7 shows how the principle of

sequential sampling is applied topulse radar level measurement. Theexample shown is a VEGAPULS trans-mitter with a microwave frequency of5.8 GHz.

55


To illustrate this principle, considerthe sine wave signal in Fig 4.6. It is aregular repeating signal with a periodof T1. If the amplitude (voltage value)of the output of the sine wave is sam-pled into a memory at a time period T2

which is slightly longer than T1, then atime expanded version of the originalsine wave is produced as an output.The time scale of the expanded outputdepends on the difference between thetwo time periods T1 and T2.

Fig 4.6 The principle of sequential sampling with a sine wave as an example.The sampling period, T2, is very slightly longer than the signal period, T1. Theoutput is a time expanded image of the original signal

Fig 4.7 Sequential sampling of a pulse radar echo curve. Millions of pulses per secondproduce a periodically repeating signal. A sampling signal with a slightly longerperiodic time produces a time expanded image of the entire echo curve

Periodic Signal(sine wave)

Periodic Signal(radar echoes)

Samplingsignal

Samplingsignal

Expandedtime signal

T1

T2

T1

Emission pulse

Echo pulse

T2


ExampleThe 5.8 GHz, VEGAPULS radar level transmitter has the following pulse repeti-tion rates.

Transmit pulse 3.58 MHz T1 = 279.32961 nanosecondsReference pulse 3.58 MHz - 43.7 Hz T2 = 279.33302 nanoseconds

56

Therefore the time expansion factoris 81920 giving a time expanded pulserepetition period of 22.88 milliseconds.

There is a practical problem in sam-pling the emission / echo pulse signalsof a short (0.8 nanosecond) pulse at 5.8GHz. An electronic switch would needto open and close within a few picosec-onds if a sufficiently short value of the5.8 GHz sine wave is to be sampled.These would have to be very specialand expensive components.

The solution is to combine sequen-tial sampling with a ‘cross correlation’procedure.

Instead of very rapid switch sam-pling, a sample signal of exactly thesame profile is generated but with aslightly longer time period between thepulses.

Fig 4.9 compares sequential sam-pling by rapid switching with sequen-tial sampling by cross correlation witha sample pulse.

This periodically repeating signalconsists of the regular emission pulseand one or more received echo pulses.These are the level surface and anyfalse echoes or multiple echoes. Thetransmitted pulses and therefore thereceived pulses have a sine wave formdepending upon the pulse duration. A5.8 GHz pulse of 0.8 nanosecond dura-tion is shown in Fig 4.8.

The period of the pulse repetition isshown as T1 in Fig 4.7. Period T1 is

the same for the emission pulse repeti-tion as for any echo pulse repetition asshown.

However, the sampling signalrepeats at period of T2 which is slight-ly longer in duration than T1. This isthe same time expansion procedure bysequential sampling that has alreadybeen described for a sine wave. Thefactor of the time expansion is deter-mined by T1 / (T2-T1).

Fig 4.8 Emission pulse (packet).The wave form of the 5.8GHz pulse with a pulseduration of 0.8nanoseconds


Instead of taking a short voltagesample, cross correlation involves mul-tiplying a point on the emission or echosignal by the corresponding point onthe sample pulse. The multiplicationleads to a point on the resultant signal.All of these multiplication results, oneafter the other, lead to the formation ofthe complete multiplication signal.

Fig 4.10 shows a short sequence ofmultiplications between the receivedsignal (E) and the sampling pulsesignal (M). The resultant E x M curvesare shown on page 58.

Then the E x M curve is integratedand represented on the expanded curveas a dot. The sign and amplitude of the

signal on the time expanded curvedepends on the sum of the area of theE x M curve above and below the zeroline. The final integrated value corre-sponds directly to the time position ofthe received pulse E relative to thesample pulse M.

The received signal E and samplesignal M in Fig 4.10 are equivalent tothe periodic signal (sine wave) andsample signal in Fig 4.6. The result ofthe integration of E x M in Fig 4.10 isdirectly analogous to the expandedtime signal in Fig 4.6.

57


Fig 4.9 Comparison of switch sampling with ‘cross correlation’ sampling. The pulseradar uses cross correlation with a sample pulse. This means that rapid ‘picosec-ond’ switching is not required

Sampling with picosecond switching

Sample signal

Emission / Echo pulse

Sampling by cross correlation with asample pulse


The pulse radar sampling procedureis mathematically complicated but atechnically simple transformation toachieve. Generating a reference signalwith a slightly different periodic time,multiplying it by the echo signal andintegration of the resultant product areall operations that can be handled easi-ly within analogue circuits. Simple, butgood quality components such as diodemixers for multiplication and capaci-tors for integration are used.

This method transforms the highfrequency received signal into an accu-rate picture with a considerablyexpanded time axis. The raw valueoutput from the microwave module isan intermediate frequency that is simi-lar to an ultrasonic signal. For examplethe 5.8 GHz microwave pulse becomesan intermediate frequency of 70 kHz.The pulse repetition frequency (PRF)of 3.58 GHz becomes about 44 Hz.

58

Fig 4.10 Cross correlation of the received signal E and the sampling M.The product E x M is then integrated to produce the expanded time curve. Thetechnique builds a complete picture of the echo curve

E

IntegralE x M

max

min

0

M

E x M


59

4. Radar level measurementam

plitu

de

transmit pulse

t1 tt22 t3 t4 t5

time

Pulse radar operates entirely withinthe time domain and does not need thefast and expensive processors thatenable the FM - CW radar to function.There are no Fast Fourier Transform(FFT) algorithms to calculate. All ofthe pulse radar processing is dedicatedto echo analysis only.

Part of the pulse radar transmissionpulse is used as a reference pulse thatprovides automatic temperature com-pensation within the microwave mod-ule circuits.

The echoes derived from a pulseradar are discrete and separated in time.This means that pulse radar is betterequipped to handle multiple echoes andfalse echoes that are common inprocess vessels and solids silos.

Pulse radar takes literally millions of‘shots’ every second. The return echoesfrom the product surface are sampledusing the method described above. Thistechnique provides the pulse radar withexcellent averaging which is particular-ly important in difficult applicationswhere small amounts of energy arebeing received from low dielectric andagitated product surfaces.

The averaging of the pulse techniquereduces the noise curve to allow small-er echoes to be detected. If the pulseradar is manufactured with welldesigned circuits containing good qual-ity electronic components they candetect echoes over a wide dynamicrange of about 80 dB. This can makethe difference between reliable andunreliable measurement.

Fig 4.12 With a pulse radar, all echoes (real and false) are separated in time. This allowsmultiple echoes caused by reflections from a parabolic tank roof to be easilyseparated and analysed

Pulse echoes in a process vessel are separated in time


60

Fig

4.1

1 B

lock

dia

gram

of

PU

LSE

rad

ar m

icro

wav

e m

odul

e


61


Pulse block diagram - (Fig 4.11)The raw pulse output signal (inter-

mediate frequency) from the pulseradar microwave module is similar, infrequency and repetition rate, to anultrasonic signal.

This pulse radar signal is derived inhardware. Unlike FM - CW radar,PULSE does not use FFT analysis.Therefore, pulse radar does not needexpensive and power consumingprocessors.

The pulse radar microwave modulegenerates two sets of identical pulseswith very slightly different periodictimes. A fixed oscillator and pulse for-mer generates pulses with a frequencyof 3.58 MHz. A second variable oscil-lator and pulse former is tuned to a

frequency of 3.58 MHz minus 43.7 Hzand hence a slightly longer periodictime. GaAs FET oscillators are used toproduce the microwave carrier fre-quency of the two sets of pulses.

The first set of pulses are directedto the antenna and the product beingmeasured. The second set of pulses arethe sample pulses as discussed in thepreceding text.

The echoes that return to the anten-na are amplified and mixed with thesample pulses to produce the raw, timeexpanded, intermediate frequency.

Part of the measurement pulse sig-nal is used as a reference pulse thatprovides automatic temperature com-pensation of the microwave moduleelectronics.

Pic 3 Two wire pulse radar level transmitter mounted in a process reactor vessel


Process radar level transmittersoperate at microwave frequenciesbetween 5.8 GHz and about 26 GHz.Manufacturers have chosen frequenciesfor different reasons ranging fromlicensing considerations, availability ofmicrowave components and perceivedtechnical advantages.

There are arguments extolling thevirtues of high frequency radar, low

frequency radar and every frequencyradar in between.

In reality, no single frequency isideally suited for every radar levelmeasurement application. If we com-pare 5.8 GHz radar with 26 GHz radar,we can see the relevant benefits of highfrequency and low frequency radar.

62

Choice of frequency

Fig 4.14 Comparison of 5.8 GHz and 26 GHz radar antenna sizes. These instrumentshave almost identical beam angles. However this is not the full picture when itcomes to choosing radar frequencies

2.6 GHz

5.8 GHz


The higher the frequency of a radarlevel transmitter, the more focused thebeam angle for the equivalent sizeantenna.

With horn antennas, this allowssmaller nozzles to be used with a morefocused beam angle.

For example, a 1½" (40 mm) hornantenna radar at 26 GHz has approxi-mately the same beam angle as a 6"(150 mm) horn antenna at 5.8 GHz.

However, this is not the completepicture. Antenna gain is dependent onthe square of the diameter of the anten-na as well as being inversely propor-tional to the square of the wavelength.

Antenna gain is proportional to:-

Antenna gain also depends on the aper-ture efficiency of the antenna.Therefore the beam angle of a smallantenna at a high frequency is notnecessarily as efficient as the equiva-lent beam angle of a larger, lower fre-quency radar. A 4" horn antenna radarat 6 GHz gives excellent beam focus-ing.

A full explanation of antenna gainand beam angles at different frequen-cies is given in Chapter 5 on radarantennas.

63


Focusing at different frequencies

5 GHz 10 GHz 15 GHz 20 GHz 25 GHz

Fig 4.13 For a given size of antenna, a higher frequency gives a more focused beam

Antenna size - beam angle

diameterwavelength

2

2


A 26 GHz beam angle is morefocused but, in some ways, it has to be.

The wavelength of a 26 GHz radar isonly 1.15 centimetres compared with awavelength of 5.2 centimetres for a5.8 GHz radar.

The short wavelength of the 26 GHzradar means that it will reflect off many

small objects that may be effectivelyignored by the 5.8 GHz radar. Withoutthe focusing of the beam, the high fre-quency radar would have to cope withmore false echoes than an equivalentlower frequency radar.

Antenna focusing and false echoes

Fig 4.15 a Low frequency radar has a wider beamangle and therefore, if the installationis not optimum, it will see more falseechoes. Low frequencies such as5.8 GHz or 6.3 GHz tend to be moreforgiving when it come to false echoesfrom the internal structure of a vesselor silo

Fig 4.15 b High frequency radar has a muchnarrower beam angle for a givenantenna size. The narrower beam angleis important because the shortwavelength of the higher frequencies,such as 26 GHz, reflect more readilyfrom the internal structures such aswelds, baffles, and agitators.The sharper focusing avoids thisproblem

64


High frequency radar transmittersare susceptible to signal scatter fromagitated surfaces. This is due to the sig-nal wavelength in comparison to thesize of the surface disturbance.

The high frequency radar willreceive considerably less signal than anequivalent 5.8 GHz radar when the liq-

uid surface is agitated. The lowerfrequency transmitters are less affectedby agitated surfaces.

It is important that, whatever the fre-quency, the radar electronics and echoprocessing software can cope with verysmall amplitude echo signals. As dis-cussed, pulse radar has an advantage inthis area no matter what the frequency.

65


Agitated liquids and solid measurement

Fig 4.16 High frequency radar transmitters aresusceptible to signal scatter fromagitated surfaces. This is due to thesignal wavelength in comparison to thesize of the surface disturbance. It isimportant that radar electronics andecho processing software can cope withvery small amplitude echo signals.By comparison, 5.8 GHz radar is not asadversely affected by agitated liquidsurfaces. Lower frequency radar isgenerally better suited to solid levelapplications

Condensation and build upHigh frequency radar level transmit-

ters are more susceptible to condensa-tion and product build up on the anten-na. There is more signal attenuation atthe higher frequencies, such as 26 GHz.Also, the same level of coating or con-densation on a smaller antenna natural-ly has a greater effect on the perfor-mance.

A 6" horn antenna with 5.8 GHz fre-quency is virtually unaffected by con-densation. Also, it is more forgiving ofproduct build up.

Steam and dustLower frequencies such as 5.8 GHz

and 6.3 GHz are not adversely affectedby high levels of dust or steam. Thesefrequencies have been very successfulin applications ranging from cement,flyash and blast furnace levels to steamboiler level measurement.

In steamy and dusty environments,higher frequency radar will suffer fromincreased signal attenuation.


FoamThe effect of foam on radar signals

is a grey area. It depends a great dealon the type of foam including the foamdensity, dielectric constant and conduc-tivity. However, low frequencies suchas 5.8 GHz and 6.3 GHz cope with lowdensity foam better than higher fre-quencies such as 26 GHz.

For example, a 26 GHz radar signalwill be totally attenuated by a very thindetergent foam on a water surface. A5.8 GHz radar signal will see throughthis type of foam and continue to seethe liquid surface as the foam thicknessincreases to 150 mm or even 250 mm.

However, the thickness of foam willcause a small measurement errorbecause the microwaves slow downslightly as they pass through the foam.

When foam is present, it is impor-tant to provide the radar manufacturerwith as much information as possibleon the application.

Minimum distanceHigher frequency radar sensors have

a reduced minimum distance whencompared with the lower frequencies.This can be an additional benefit when measuring in small vessels and stillingtubes.

Fig 4.17 Focusing and radar frequency

Summary of the effects of radar frequency

focu

sing

Better focusing at higher emitting frequency means:

. higher antenna gain (directivity). less false echoes. reduced antenna size


frequency range

66



Fig 4.19 Signal strength from agitated and undulating surfaces and radar frequency

redu

ced

sign

al c

ause

d by

dam

ping

refle

ctio

n fr

om m

ediu

m

Reduced signal strength caused bydamping at higher emitting frequencycaused by:

. condensation. build - up. steam and dust

Higher damping caused by agitatedproduct surface

. wave movement. material cones with solids. signal scattered


frequency range


frequency range

67

Fig 4.18 Signal damping and radar frequency


68

AccuracyThere is no inherent difference in

accuracy between the FM - CW andPULSE radar level measurement tech-niques.

In this book, we are concernedspecifically with process level mea-surement where ‘process accurate’ andcost effective solutions are required.

The achievable accuracy of aprocess radar depends heavily on thetype of application, the antenna design,the quality of the electronics and echoprocessing software employed.

The niche market for custody trans-fer level measurement applications isoutside the scope of this book. Thesecustody transfer radar ‘systems’ areused in bulk petrochemical storagetanks. Large parabolic or planar arrayantennas are used to create a finelyfocused signal. A lot of processingpower and on site calibration time isused to achieve the high accuracy.Temperature and pressure compensa-tion are also used.

Range resolution andbandwidth

In process level applications, bothFM - CW and PULSE radar work withan ‘envelope curve’. The length of thisenvelope curve depends on the band-width of the radar transmitter. A widerbandwidth leads to a shorter envelopecurve and therefore improved rangeresolution. Range resolution is one of anumber of factors that influence theaccuracy of process radar level trans-mitters.

Pulse radar bandwidthThe carrier frequency of a pulse

radar varies from 5.8 GHz to about26 GHz.

The pulse duration is importantwhen it comes to resolving two adja-cent echoes. For example, a onenanosecond pulse has a length of about300 mm. Therefore, it would be diffi-cult for the radar to distinguish betweentwo echoes that are less than 300 mmapart. Clearly a shorter pulse durationprovides better range resolution.

An effect of a shorter pulse durationis a wider bandwidth or spectrum offrequencies.

For example, if the carrier frequencyof a pulse is 5.8 GHz and the durationis only 1 nanosecond, then there is aspectrum of frequencies above andbelow the nominal carrier frequency.The amplitude of the pulse spectrum offrequencies changes according to a

curve.The shape of this curve is shown in

Fig 4.21. The null to null bandwidth BWnn of

a pulse radar is equal to

where τ is the pulse duration. It is clear from the curve that the

amplitude of frequencies reduces sig-nificantly away from the main pulsefrequency.

sin xx

2τ



Fig 4.20 Pulse radar range resolution.The guaranteed rangeresolution is the length of thepulse. A shorter pulse has awider bandwidth and betterrange resolution

shorter pulse

better range resolution

bandwidth BW nn,equal to

pulse frequency5.8 GHz

6.8 GHz4.8 GHz

Fig 4.21 The null to null bandwidthBWnn of a radar pulse is equalto 2 / ττ where ττ is the pulseduration. Example a 5.8 GHzradar with a pulse duration ofone nanosecond has a null tonull bandwidth of 2 GHz

Fig 4.22 Envelope curve with pulse radar

Pulse radar envelope curveFig 4.22 shows how a pulse radar

echo curve is used in process levelmeasurement.

A higher frequency pulse with ashorter pulse duration will allow betterrange resolution and also better accura-cy because the leading edge of theenvelope curve is steeper.

2τ

Fig 4.23 A shorter pulse duration gives better range resolution. The combination ofshorter pulse duaration and higher frequency allows better accuracy because theleading edge of the envelope curve is steeper

High frequency, shortduration pulse

Lower frequency pulse withlonger duration

69


70

FM-CW radar bandwidthThe bandwidth of an FM - CW radar isthe difference between the start andfinish frequency of the linear frequencymodulation sweep.Unlike pulse radar, the amplitude of theFM - CW signal is constant across therange of frequencies.

A wider bandwidth produces narrower difference frequency ranges for each echo on the frequency spectrum. Thisleads to better range resolution in thesame way as with shorter duration puls-es with pulse radar.This is explained in the following dia-grams and equations.

frequency

amplitude

fd

∆fd

fd

frequency

timeTs

∆F

Fast Fourier Transform

fd =∆F x 2RTs x c

[Eq. 4.1]

∆F bandwidthTs sweep timeR distancefd difference frequencyc speed of light

The FAST FOURIERTRANSFORM produces afrequency spectrum of all echoessuch as that at fd.There is an ambiguity ∆fd for eachecho fd.

∆fd

[Eq. 4.2]

=2Ts


71


From equation 4.3, it can be seenthat with an FM - CW radar the rangeresolution ∆R is equal to:-

Therefore, the wider the bandwidth, thebetter the range resolution.Examples:

A linear sweep of 2 GHz has a rangeresolution of 150 mm whereas a 1 GHzbandwidth has a range resolution of300 mm.

In process radar applications, eachecho on the frequency spectrum isprocessed with an envelope curve. Theabove equations (Equations 4.1 to 4.3)show that the Fast Fourier Transforms(FFTs) in process radar applications donot produce a single discrete differencefrequency for each echo in the vessel.Instead they produce a difference fre-quency range ∆fd for each echo withinan envelope curve. This translates intorange ambiguity.

amplitude

distance∆R

R

Fig 4.24 to 4.26 - FM - CW range resolution

The ambiguity of the distance R,is ∆ R

∆RR =

∆fdfd

∆RR =

2

∆F x 2 RTs

Ts x c

∆RR =

c∆F x R

∆R = c∆F

[Eq. 4.3]

c∆F


Other influences on accuracyAs we have demonstrated, FM - CW

and PULSE process radar transmittersuse an envelope curve for measure-ment. A wider bandwidth produces bet-ter range resolution. The correspond-ingly short echo will have a steep slopeand therefore a more accurate measure-ment can be made. Other influences onaccuracy include signal to noise ratioand interference.

A high signal to noise ratio allowsmore accurate measurement whileinterference effects can cause a distur-bance of the real echo curve leading toinaccuracies in the measurement.

Choice of antenna and mechanicalinstallation are important factors inensuring that the optimum accuracy isachieved.

FM - CW frequency spectrum - bandwidth and range resolution

Frequency spectrum - narrow bandwidth of linear sweep

envelope curvesaround echoes

envelope curvesaround echoes

frequency

amplitude

amplitude

Frequency spectrum - wide bandwidth of linear sweep

Fig 4.28 Illustration of envelope curve around the frequency spectram of FM - CWradars. The same four echoes are shown for radar transmitters with differentbandwidths. An improvement in the range resolution is achieved with a widerbandwidth of the linear sweep

frequency

72


73


Fig 4.29 Higher accuracy of pulse radarlevel transmitters can beachieved by looking at the phaseof an individual wave within theenvelope curve. This is onlypractical in slow moving storagetanks

High accuracy radarHigh accuracy of the order of

+ 1 mm is generally meaningless in anactive process vessel or a solids silo.For example, a typical chemical reactorwill have agitators, baffles and otherinternal structures plus constantlychanging product characteristics.

Although custody transfer levelmeasurement applications are not in thescope of this book, this section discuss-es how a higher accuracy can beachieved.

Pulse radarFor most process applications, mea-

surement relative to the pulse envelopecurve is sufficient. However, if the liq-uid level surface is flat calm and theecho has a reasonable amplitude, it ispossible to look inside the envelopecurve wave packet at the phase of anindividual wave.

However, the envelope curve of ahigh frequency radar with a short pulseduration is sufficiently steep to producea very accurate and cost effective leveltransmitter for storage vessel applica-tions.

FM - CW radarThe fundamental requirement for an

accurate FM - CW radar is an accuratelinear sweep of the frequency modula-tion.

As with the pulse radar, it is possibleto look inside the envelope curve of thefrequency spectrum if the applicationhas a simple single echo that is charac-teristic of a liquids storage tank. This isachieved by measuring the phase angleof the difference frequency. However,this is only practical with custodytransfer applications where fast andexpensive processors are used withtemperature and pressure compensa-tion.

Fig 4.30 It is essential that the linearsweep of the FM - CW radar isaccurately controlled

frequency error

f2

f2t1


74

Microwave powerRadar is a subtle form of level mea-

surement. The peak microwave powerof most process radar level transmittersis less than 1 milliWatt. This level ofpower is sufficient for tanks and silosof 40 metres or more.

The average power depends on thesweep time and sweep repetition rate ofFM - CW radar and on the pulse dura-tion and pulse repetition frequency ofpulse radar transmitters.

An increase in the microwave powerwill produce higher amplitude echoes.However, it will produce higher ampli-tude false echoes and ringing noiseas well as a higher amplitude echoesoff the product surface. The averagemicrowave power of a Pulse radar canbe as little as 1 microWatt.

Processing powerFM - CW radars need a high level of

processing power in order to function.This processing power is used to calcu-late the FFT algorithms that producethe frequency spectrum of echoes.The requirement for processing powerhas restricted the ability of FM - CWradar manufacturers to make a reliabletwo wire, intrinsically safe radar trans-mitter.

Pulse radar transmitters work in thetime domain without FFT analysis andtherefore they do not need powerfulprocessors for this function.

SafetyThe low power output from

microwave radar transmitters meansthat they are an extremely safe methodof level measurement.

Pulse radarThe low energy requirements of

pulse radar enabled the first ever twowire, loop powered, intrinsically saferadar level transmitter to be introducedto the process industry in mid-1997.The VEGAPULS 50 series of pulseradar transmitters have proved to bevery capable in difficult process condi-tions. The performance of the two wire,4 to 20 mA, sensors is equal to the fourwire units that preceded them.

The pulse microwave module onlyneeds a 3.3 volt power supply witha maximum power consumption of50 milliWatts. This drops down to5 milliWatts when it is in stand-bymode. The difference between the twowire pulse and the four wire pulse isthat the two wire radar sends out burstsof pulses and updates the output aboutonce every second. The four wire sendsout pulses continuously and updatesseven times a second.

With high quality electronics, thecomplete 24 VDC, 4 to 20 mA trans-mitter is capable of operating at only14 VDC. This allows it to directlyreplace existing two wire sensors.

Pulsed FM - CWThe low power requirements of

pulse radar have allowed two wireradar to become sucessful. FM - CWradar requires processing power andtime for the FFT's to be calculated.Power saving has been used to producea ‘pulsed’ FM - CW radar. However,this device is limited to simple storageapplications because the update time istoo long and the processing too limitedfor arduous process applications.

Power Two wire, loop powered radar


75


Summary of radar level techniques

FM - CW (frequency modulated continuous wave) radar · Indirect method of level measurement · Requires Fast Fourier Transform (FFT) analysis to convert signals into a fre-quency spectrum

· FFT analysis requires processing power and therefore practical FM - CWprocess radars have to be four wire and not two wire loop powered

· FM - CW radars are challenged by large numbers of multiple echoes (causedby the parabolic effects of horizontal cylindrical or dished topped vessels)

PULSE radar· Direct, time of flight level measurement· Uses a special sampling technique to produce a time expanded intermediatefrequency signal

· The intermediate frequency is produced in hardware and does not require FFTanalysis

· Low processing power requirement mean that practical and very capable twowire, loop powered, intrinsically safe pulse radar can be used in some of themost challenging process level applications











Contents

The function of an antenna in a radarlevel transmitter is to direct the maxi-mum amount of microwave energytowards the level being measured andto capture the maximum amount ofenergy from the return echoes foranalysis within the electronics.

Antennas for level measurementcome in five basic forms:

Horn antennas and dielectric rodantennas are already commonly usedwithin process level measurement. Wewill be discussing how these designshave been developed for increasinglyarduous process conditions and howantenna efficiencies have beenimproved. The horn antenna and ver-sions of the dielectric rod antenna arealso used in measuring tube applica-tions within the process industry.

Parabolic antennas and planar arrayantennas have been applied to fiscalmeasurement radar systems rather thanfor level measurement within processvessels. We will discuss the design ofthese antennas although at present theiruse in process vessels is limited.

Antenna basicsAn important aspect of an antenna isdirectivity. Directivity is the ability ofthe antenna to direct the maximumamount of radiated microwave energytowards the liquid or solid we wish tomeasure.

No matter how well the antenna isdesigned, there will be somemicrowave energy being radiated inevery direction. The goal is to max-imise the directivity.

Fig 5.1 shows the pattern of radiatedenergy from a typical horn antenna.This is a 250 mm (10") horn antennaoperating at a frequency of 5.8 GHz.

The measurements are made somedistance from the antenna in what iscalled the far field zone. It is clear thatmost of the energy is contained withinthe main lobe, but also there is a rea-sonable amount of energy containedwithin the various side lobes.

Technical information and sales lit-erature on radar level transmittersquote beam angles for different anten-nas. Clearly there is not a tight beam.The convention is to measure the angleat which the microwave energy hasreduced to 50 percent of the value atthe central axis of the beam. This is quoted in decibels:- the - 3dB point.

77

5. Radar antennas

· Horn (cone) antenna· Dielectric rod antenna· Measuring tube antenna

(stand pipes/ bypass tubes etc.)· Parabolic reflector antenna· Planar array antenna

78

Extent of measured microwave energy showingmain lobe and side lobes

The - 3 dB point is the beam angle i.e. the energyhas reduced to 50%

Side lobe energy

Fig 5.1 Typical radiation pattern from a radar level transmitter

0

30

60

90

120

150

30

60

90

120

150

180

main lobe directionangular width (3dB)side lobe suppression

Max.:

Farfield E_Abs (Theta); Phi=90,0 deg.

:::

0 10 20 30

0,0 deg.14,9 deg.21,6 dB

20,4 dB

Radiation patterns of different antennas and radar frequencies are compared at theend of this chapter.

A measure of how well the antennais directing the microwave energy iscalled the ‘antenna gain’.

Antenna gain is a ratio between thepower per unit of solid angle radiated

by the antenna in a specific direction tothe power per unit of solid angle if thetotal power was radiated isotropically,that is to say, equally in all directions.

5. Radar antennas

79

Fig 5.2 Illustration of antenna gain

Isotropic equivalent with total powerradiating equally in all directions

Directional power from antenna

Antenna gain ‘G’ can be calculated as follows:

The aperture efficiencies of radarlevel antennas are typically betweenη = 0.6 and η = 0.8.

It is clear from equation 5.1 thatthe directivity improves in proportionto the antenna area. At a given fre-quency, a larger antenna has a narrow-er beam angle

Where η = aperture efficiency

D = antenna diameter.*

A = antenna area.*

λ = microwave wavelength *

* must be same units

isotropic power

directional power

G = =2

2η x η xπ x D 4π x Aλ λ

[Eq. 5.1]

( )

Also, we can see that the antennagain and hence directivity is inverselyproportional to the square of the wave-length.

For a given size of antenna the beamangle will become narrower at higherfrequencies (shorter wavelengths). Forexample the beam angle of a 5.8 GHzradar with a 200 mm (8") horn antennais almost equivalent to a 26 GHz radarwith a 50 mm (2") horn antenna. This

means that a 26 GHz antenna is lighterand easier to install for the same beamangle. However, as discussed inChapter 4, this is not the whole storywhen choosing the right transmitter foran application.

For a standard horn antenna thebeam angle φ, that is the angle to theminus 3 dB position, can be calculatedusing equation 5.2.

80

Fig 5.3 Graph showing relation between horn antenna diameter and beam angle for5.8 GHz, 10Ghz and 26GHz radar

The following graph shows horn anten-na diameter versus beam angle for the

most common radar frequencies,5.8 GHz, 10 GHz and 26 GHz.

050 75 100 125 150 175 200 225 250

5.8 GHz

10 GHz

26 GHz

20

40

60

80

φ = 70° xλD

[Eq. 5.2]

Antenna beam angles (diameter / frequency)

beam

ang

le in

deg

rees

(-3

dB)

antenna diameter, mm

Beam angle

The metallic horn antenna or coneantenna is well proven for process levelapplications. The horn is mechanicallyrobust and in general it is virtuallyunaffected by condensation and prod-uct build up, especially at the lowerradar frequencies such as 5.8 GHz.

There are variations in the internaldesign of horn antennas. Themicrowaves that are generated withinthe microwave module are transmitteddown a high frequency cable for encou-pling into a waveguide. The metalwaveguide then directs the microwavestowards the horn of the antenna. A lowdielectric material such as PTFE,ceramic or glass is often used withinthe waveguide.

At the transition from the wave-guide to the horn of the antenna the lowdielectric material is machined to apointed cone. The angle of this conedepends on the dielectric constant ofthe material. For example, ceramic hasa sharper angle than PTFE.

The microwaves are emitted fromthis pointed cone in a controlled wayand are then focused towards the targetby the metal horn.

After reflection from the productsurface, the returning echoes arecollected within the horn antenna forprocessing within the electronics.

5. Radar antennas

81

Fig 5.4 The transition ofmicrowaves from the lowdielectric waveguide into themetallic horn where they arefocused towards the productbeing measured

1. Horn antennas

82

In this first design of horn antennathe HF cable signal coupling is into anair filled waveguide with a rectangularcross section. The microwaves aredirected towards the antenna. There is atransition from rectangular to circularcross section. At this point the wave-guide changes to PTFE with a ¼ wave-length step design. The waveguide isthen glass filled until it reaches theinside of the antenna horn where itchanges to a PTFE cone for the imped-ance matching into the vapour space inthe horn

This PFTE cone in combination withthe metallic horn focuses themicrowaves towards the target.

An antenna of this design is capableof withstanding process temperaturesup to 250° C and up to 300 Bar.

A potential problem with the designis the sealing between the PTFE andglass on the process side. The thermalexpansion of glass and PTFE are differ-ent and it is possible for condensationto get between the glass and PTFE andto affect the transmission and receipt ofthe microwave signals.

The explosion proof design requiresmetallic grid around the glass of thewaveguide at the joint between thehousing casting and the flange casting.

Horn antenna design 1Fig 5.5

1. HF Cable

2. Signal coupling

4. PTFE transition

5. Glass waveguide

6. Metallic grid

7. Seal between glassand PTFE

8. PTFE cone

9. Metal horn antenna

3. Waveguide (air filled)

Transition rectan-gular to circularcross section

1

234567

8

9

5. Radar antennas

83

With this antenna design, the HFcable is encoupled into the PTFE mate-rial inside the waveguide. The metalwaveguide is welded to the flange andthere are two process seals between themetal waveguide and the PTFE. Theseseals protect the signal coupler fromthe process. This seal material can beViton for stainless steel horn antennasor Kalrez for Hastelloy C horn anten-nas.

There is a continuous transition forthe microwaves within a single piece ofPTFE which is machined into a cone

form for the transition into the hornantenna. The PTFE cone and the metal-lic conical horn focus the microwavesand collect the return signals in theusual manner.

An antenna of this design is capableof withstanding a process temperatureof 200° C + and a process pressure of40 Bar.

This antenna design can also be usedon very high temperature, ambientpressure applications with air or nitro-gen gas cooling of the antenna.

Horn antenna design 2Fig 5.6

1. HF cable

2. Signal coupling

3. Waveguide(PTFE filled)

5. PTFE cone

4. Process seals Vitonor Kalrez

6. Metallic hornantenna

1

234

5

6

84

Horn antenna design 2aFig 5.7 Very high temperature, ambient pressure applications.

Air/nitrogen cooling through flange

This adaptation of the previousantenna allows the antenna to be cooledwith air or nitrogen gas.

This is achieved by drilling twoholes, 180° apart, laterally from theflange edge into the horn antenna nextto the PTFE cone. The flow of air ornitrogen prevents hot gases fromaffecting the PTFE and the viton sealand it effectively cools the entire flangeand horn area.

This technique has been used suc-cessfully with very high temperatures,including 1500° C + in the steel indus-try with applications such as blast

furnace burden level and molten ironladle levels. The microwaves are unaf-fected by the air movement within thehorn area.

In addition to cooling, this air purg-ing technique is also used for solidsapplications where very high levels ofconductive dust, such as carbon, heavi-ly coat the inside of the horn and causesignal attenuation.

Water purging has also been usedwhere heavy product build up isexpected.

1. HF cable

2. Signal coupling

3. Waveguide(PTFE filled)


4. Tappings forair/nitrogen keepsantenna area cool

1

2

3

4

Air / N2

5

5. Radar antennas

85

Horn antenna design 3

This antenna is also a developmentof the antenna design in Fig 5.6.

The waveguide, PTFE transitioncone and process flange are standard.The face of the flange is all PTFE.

The difference is in the applicationof a special enamel (glass) coated hornthat provides excellent process materi-als compatibility without resorting tomore expensive metals such asTantalum.

The external dimensions of theantenna represent a simple cylinder.The internal dimensions of the antennaare identical to a standard horn antenna(150 mm (6")) is illustrated. At the bot-tom of the antenna there is a gradual lip

between the external cylinder and theinternal horn.

The top of the cylinder has a flangefor sealing between the PTFE transitioncone and the process flange and alsobetween the glassed antenna and thevessel nozzle. External studs hold theenamel antenna to the process flangeand PTFE seals are used to provideinternal sealing.

The antenna is manufactured fromcarbon steel with blue enamel coatingwhich is identical to the enamel foundin glass lined vessels. It provides theefficiency benefits of a horn antennawith first class materials compatibility.

Fig 5.8 Special enamel coated antenna

2. PTFE waveguide

1. Signal coupling

3. PTFE flange face

5. Lapped flange

4. PTFE seal

6. Steel internals ofhorn antenna

7. Enamelled coating

12

345

6

7

The above antenna has beendesigned with both high temperatureand high pressure in mind. Themechanical strength and sealing abilityof PTFE degrades at elevated tempera-ture and is therefore limited to about200° C.

This special design of radar hasa chemically and thermally stableceramic (Al2O3) waveguide within astainless steel or Hastelloy C hornantenna and flange. The ceramicwaveguide is fused to a ‘vacon’ steelbush using a special brazing technique.‘Vacon’ is used because it has acoefficient of thermal expansion that issimilar to ceramic, whereas normal

stainless steel expands more than twiceas much as ceramic. A double graphiteseal is fitted on the process side of the‘vacon’ bush. The entire waveguideassembly is laser welded to ensure thatthe transmitter is gas tight and thatdifferential thermal expansion isnegligible.

In order to withstand constant pro-cess temperatures of 400° C, the elec-tronics housing of the radar is mechani-cally isolated from the high processtemperature by a temperature extensiontube. The microwave module is con-nected via the HF cable and an aircoaxial tube to the signal coupler in theceramic waveguide.

86

Horn antenna design 4Fig 5.9 High temperature / high pressure antenna with ceramic waveguide

1. Connection to HFcable frommicrowave module

2. Coaxial tube tosignal coupling

3. Signal coupling inceramic waveguide

4.Vacon/ceramicbrazing seal

5. Graphite seal

6. Ceramic waveguidecone

1

2

3

4

5

6

5. Radar antennas

87

1. HF cable (coaxial)

2. Signal coupling

3. Ceramic waveguide

4. Brazing of ceramicto vacon

5. Vacon bush

6. Graphite seal


Fig 5.11 This antenna design is capableof with standing 160 Bar at400° C with dual graphite seals.Graphite seals have proved to besuperior to tantalum seals

Ceramic signal coupling

Vacon/ceramic brazing

Graphite / Tantalum seal

Fig 5.10 Close up of ceramic waveguide assembly

1

23

4

5

6

7

Another possible variation of a hornantenna radar is measurement througha low dielectric window. We have dis-cussed Hastelloy, Tantalum and thespecial enamel coated horn antenna.However, if a liquid is being measured and it is conductive or has a dielectric

constant of more that εr = 10, then it ispossible to measure through a lowdielectric window or lens.

Some antennas are manufacturedwith a PTFE window as part of theconstruction.

88

Fig 5.12 Horn antenna radar is constructed with a metal housing around the antennaand a PTFE process ‘window’

Fig 5.13 Variations of this design include the use of cone shaped windows. The cone canpoint towards the horn or towards the process

Adapting horn antenna radars

a. Measurement through a PTFE window

Antenna housing

Horn antenna

Process flange

PTFE window

b. Horn antenna -waveguide extensionIn the first section of Chapter 6,

Radar level installations, we discusshow horn antenna radars should beinstalled. It is recommended that theend of the antenna is a minimum of10 mm inside the vessel. A 150 mm(6") horn antenna is 205 mm (8") long.

If the nozzle is longer than 200 mm,we should consider a waveguide exten-sion piece between the radar flange andthe horn antenna. Waveguide exten-sions should only be used with highlyreflective products.

c. Horn antenna -bent waveguide extensionsAs well as simple waveguide exten-

sions it is possible to bend waveguideextensions in order to avoid obstruc-tions or to utilise side entry flanges.

A simple 90° bend or an ‘S’ shapedextension tube are possible.

The waveguide extensions should befree from any internal welds and theminimum radius of curvature should be200 mm.

5. Radar antennas

89

Fig 5.15 Waveguide extensionswith bends. The directionof the polarization isimportant

Waveguide extension with 90° bend

Waveguideextension with ‘S’bend

Fig 5.14 Extended waveguide hornantenna to enable measurementin long nozzles or through aconcrete tank or sump roof

The majority of antennas in thischapter are designed for microwavefrequencies of between 5.8 GHz and10 GHz. Later in this chapter, we dis-cuss the use of radar in measuringtubes where there is a minimum criticaldiameter for each frequency. A measur-ing tube is a waveguide. The minimumtheoretical tube diameter for a 5.8 GHzradar is 31 mm.

At a higher frequency the minimumdiameter of a waveguide is smaller.

At this minimum diameter, themicrowaves are established within thewaveguide with a single mode andhence a single velocity.

As the waveguide diameter increas-es in size, more modes become estab-lished for the given frequency.

Measurement problems will beencountered if there are multiple modeswithin an antenna waveguide. This isbecause with different modes themicrowaves travel at different veloci-ties in the waveguide and therefore asingle target will reflect more than onereturn echo. Measurement will becomeinaccurate or impossible.

For this reason, the encoupling of ahigh frequency radar must be made intoa small waveguide. The small wave-guide assemblies of high frequencyradar are susceptible to contaminationby condensation and build up whencompared with lower frequencies suchas 5.8 GHz.

A special patented high frequencyantenna design from VEGA minimisesthe potential problems associated withsmall waveguide assemblies.

The encoupling is made within asmall PTFE waveguide to establish asingle mode. As the microwaves traveltowards the horn antenna, there is acarefully designed transition thatincreases the diameter of the PTFEwaveguide while maintaining the singlemode.

The increased diameter of the PTFEwaveguide reduces the adverse effectsof condensation and build up where thetapered cone of the waveguide entersthe metallic horn of the antenna.

Compare this design with hornantenna design 2, Fig 5.6. The 5.8 GHzradar does not need a transition in thewaveguide diameter and the angle ofthe metallic horn is not as sharp as forthe high frequency radar.

Viton or Kalrez process seals are fit-ted between the PTFE and stainlesssteel body of the waveguide.

Extended versions of the highfrequency antenna design involvelengthening the HF cable within astainless steel extension tube and weld-ing the waveguide assembly to the endof the extension tube.

90

High frequency radar antennas

5. Radar antennas

91

Fig 5.16 High frequency (26GHz) horn antenna design

1

2

3

4

5

6

1. HF cable frommicrowave module

2. Signal coupling intosmaller diameter PTFEwaveguide assembly

4. Viton or Kalrez processseals between PTFE andstainless steel of thewaveguide

5. Cone shape of PTFEwaveguide for thetransition into themetallic horn of theantenna

6. Metallic horn antennaof high frequency radar.It has a sharper anglethan the lower frequencyradars

3. Carefully designedtransition from smalldiameter to largerdiameter withoutaffecting the waveguidemode

92

The dielectric rod antenna is anextremely useful option when applyingradar level technology to modernprocess vessels. Dielectric rods can beused in vessel nozzles as small as40 mm (1½") and they are manufac-tured from PP, PTFE or ceramic wettedparts.

This means that, normally, radarlevel transmitters can be retro-fittedinto existing tank nozzles and theyhave low cost materials compatibilitywith most aggressive liquids includingacids, alkalis and solvents.

The design of dielectric rod antennashas been refined in recent years.Essentially the microwaves are fedfrom the microwave module through anHF cable to a signal coupler in thewaveguide. As with the horn antennathe waveguide can be air filled or filledwith a low dielectric material such asPTFE .

The waveguide feeds themicrowaves to the antenna. Themicrowaves pass down the parallelsection of the rod until they reach thetapered section of the rod. The taperedsection of the rod acts like a lens and itfocuses the microwaves towards theproduct being measured. The size andshape of the dielectric rod depends onthe frequency of the microwaves beingtransmitted.

The reflected echoes are captured ina similar fashion for processing by theradar electronics.

Rod antennas should only be usedon liquids and slurries and not on pow-ders and granular products.

There are some important considera-tions when applying rod antennaradars.

First of all, the tapered section of therod must be entirely within the vessel.

If the tapered section is in a nozzle,it will cause ‘ringing’ noise that willeffectively blind the radar. This isexplained more fully in Chapter 6.

Also, it can be seen from Fig 5.17that the microwaves rely on the rodantenna being clean. If a rod antenna iscoated in viscous, conductive and adhe-sive products, the antenna efficiencywill deteriorate very quickly.

With the horn antenna product buildup is not a particular problem.However, product build up worksagainst the reliable functioning of a rodantenna radar.

2. Dielectric rod antennas

5. Radar antennas

93

The microwaves travel down the inactiveparallel section of the rod towards thetapered section .

The tapered section of the rod focuses themicrowaves toward the liquid beingmeasured .

It is very important that all of the taperedsection of the rod must be inside the vessel

It is not good practice to allow a rodantenna to be immersed in the product

If a rod antenna is coated in viscous,conductive and adhesive product, theantenna efficiency will deteriorate

Fig 5.17 Dielectric rod antenna

94

This rod antenna is a simple and lowcost design that provides a radar leveltransmitter with good materials com-patibility. It is ideal for vented and lowpressure vessels such as acid and alkalitanks. It is designed for use in short1½" BSP / NPT process nozzles. Thenozzle height should not exceed 60 mm(2½").

The process connection is a 1½"PVDF boss and the antenna ispolypropylene (PP) or PTFE.

The HF cable from the microwavemodule is coupled into PTFE/PP insidea metallic tube that acts as a wave-guide. This metallic tube is totallyenclosed within the PTFE/PP parallelsection of the antenna. The microwavespass down the metallic waveguidedirectly to the tapered section of theantenna where they are focusedtowards the product being measured.

Rod antenna design 1

1. HF cable

Fig 5.18 Rod antenna for short process nozzles

2. Process connectionPVDF boss

3. Signal couplingwithin PTFE/PPfilled waveguide

4. Inactive sectionwith metallic wave-guide, PTFE/PPinner and outerparts

5. Solid PTFE/PPactive taperedsection of antennafocuses themicrowaves towardsthe product surface

123

4

5

5. Radar antennas

95

With this design of rod antenna thesignal coupling is into an air filledwaveguide. The microwaves are direct-ed towards the antenna. There is a tran-sition to PTFE via a cone shaped ele-ment. The microwaves continuethrough the PTFE waveguide to thesolid PTFE dielectric rod. The taperedsection of the rod focuses themicrowaves towards the product beingmeasured.

If this type of antenna is to be usedin a long nozzle, the parallel section ofthe solid rod is extended to ensure thatthe tapered section is entirely withinthe vessel.

An extended, solid PTFE rod anten-na can suffer from ‘ringing’ noisecaused by microwave leakage from theparallel section resonating within thenozzle. See Fig 5.20.

Rod antenna design 2Fig 5.19 Rod antenna with solid PTFE extendible rod

1. HF cable

2. Signal coupling

3. Air waveguide

4. PTFE cone

5. Process connection

7. Solid PTFE taperedsection

6. Solid PTFE parallelsection length canbe extended

1

234

5

6

7

96

In theory, the microwaves shouldtravel within the parallel section for theentire length until it reaches the taperedsection. However, in practice, some ofthe microwave energy escapes from theparallel sides.

Some solid PTFE rod antennasare supplied with screw - on extendibleantennas.

In addition to the ‘ringing’ noiseproblem described, this design can suf-fer from condensation forming betweenthe rod sections causing signalattenuation.

Also the PTFE expands at elevatedtemperatures and under certain processconditions it is possible for the rod sec-tions to detach.

The potential problems of solidPTFE rod antennas have been solvedby the latest designs. It is important tohave a completely inactive parallel sec-tion within a vessel nozzle. This isachieved by special screening or signalcoupling beyond the nozzle.

Fig 5.20 Extended rod antenna in solid PTFE. This design can suffer from ‘ringing’noise caused by leakage of microwave energy from the parallel section of thesolid PTFE rod resonating in the vessel nozzle

5. Radar antennas

97

This antenna is designed for use innozzles of either 100 mm length or250 mm length. All wetted parts of theantenna are PTFE. The parallel sectionthat is designed to be within the nozzlehas a PTFE coating on a cast metaltube.

Below this parallel section is theactive, solid PTFE, tapered antenna.

The HF cable from the microwavemodule is fed through the metal castingand the signal coupling is made justabove the tapered rod. The parallel and

tapered sections are sealed together andare designed to withstand a processtemperature of 150° C .

This antenna design is used with1½" BSP (M) stainless steel bosses orwith PTFE faced flanged transmitters.

The flanged version is designed formaximum chemical resistance to acids,alkalis and solvents. The flange face isPTFE with a tight seal between theflange PTFE and the top of the PTFEcovered inactive section.

Rod antenna design 3Fig 5.21 Extended rod antenna with inactive section and signal coupling below nozzle

level

1. HF cable

2. Rod extensioncasting(metal within PTFE)

3. Signal coupling atthe bottom of therod extension

4. Inactive section

5. Solid PTFE tapered‘active’ section ofrod antenna

4

5

1

2

3

98

For less arduous applications a stainless steel extension tube is used instead of thePTFE covered tube. The tapered section of the antenna is made of polyphenylenesulphide (PPS).

Fig 5.22 Extended rod antenna with inactive section and signal coupling below nozzlelevel. All wetted parts are PTFE on the flanged version of this antenna

Extended rod antennafor 250 mm nozzle

Extended rod antennafor 100 mm nozzle

Fig 5.23 Extended rod antenna with stainless steel inactive section and PPS rod antenna.This is for less chemically arduous process conditions

5. Radar antennas

99

This design of dielectric rod antennais for use with flanged process connec-tions.

The HF cable is connected into aPTFE filled waveguide which directsthe microwave energy towards the rodantenna. There is a PTFE male screwedfitting at the end of the waveguidewithin the process flange. The fabricat-ed, one piece, rod antenna screws on tothis connection.

The antenna flange facing and theparallel section of the antenna have car-bon impregnated PTFE wetted parts.

Inside the parallel section of the rodthere is a tubular metallic grid that acts

as an extension to the waveguide.Inside the grid the waveguide is virginPTFE, outside the grid the PTFE is car-bon impregnated.

At the end of the parallel section,there is a transition into a solid PTFEtapered rod which provides the imped-ance matching and focusing of themicrowaves towards the product beingmeasured.

This antenna has the option for100 mm or 250 mm nozzle lengths. Asalready discussed, the tapered sectionmust be entirely within the vessel.

Rod antenna design 4Fig 5.24 Extended rod antenna with metallic grid waveguide extension within carbon

impregnated PTFE inactive rod. Tapered active section of virgin PTFE

1. HF cable

2. Signal coupling

3. PTFE waveguide

4. Screwed connection

5. Carbon impregnatedPTFE antenna parallelsection and flange face

6. Internal metal grid actsas extended waveguideand prevents microwaveleakage from theparallel section of theantenna

7. PTFE waveguide

8. Virgin PTFE taperedantenna

1

2345

6

7

8

100

Rod antenna design 5Fig 5.25 This is a high temperature ceramic rod antenna design. There is temperature

separation between the electronics and the signal coupling (similar to the hightemperature horn antenna Fig 5.10). The ceramic rod has a sharper taper thanthe equivalent PTFE rod

Rod antennas are available with thedielectric rod manufactured fromceramic (Al2O3).

Ceramic has good chemical andthermal resistance. However, care must

be taken when installing ceramic rodsbecause they are brittle and prone toaccidental damage.

1

2

3

4

1. Signal coupling

2. Ceramic waveguide

3. Process seal (graphite ortantalum)

4. Active tapered ceramicrod

5. Radar antennas

101

As discussed, conical horn antennasand dielectric rod antennas are usedwidely within the process industry.

In general horn antennas aremechanically more robust and do notsuffer as much from build up or heavycondensation.

On the other hand, dielectric rodsare smaller, weigh less and can be con-structed from low cost but chemically

resistant plastics such as PTFE andpolypropylene.

However, there are applicationswithin the process industry where theinstallation of an antenna directly with-in a vessel is not suitable for reasons ofvessel design or radar functionality. Inthese cases a measuring tube (bypasstube or a stand pipe within the vessel)may be an alternative.

· Highly agitated liquid surfaces -a stilling tube ensures that theradar sees a calm surface withno scattering of the echo signal

· Low dielectric liquids such asliquefied petroleum gas (LPG) -a stand pipe concentrates andguides the microwaves to theproduct surface giving themaximum signal strength fromliquids with low levels ofreflected energy

· Toxic and dangerous chemicals -a stand pipe installation makes asmall antenna size possible.This can be used to look througha full bore ball valve into thestand pipe.The instrument can be isolatedfrom the process formaintenance

· Small vessels - stand pipes orbypass tubes can be used formeasurement in very smallprocess vessels such as vacuumreceivers. There may not beenough head space for a rodantenna or a suitable connectionfor a horn antenna. A small boretube can be used with a radar

· Foam - a stilling tube can oftenprevent foam affecting themeasurement

· Replacing existing floats anddisplacers - radar can beinstalled directly into existingbypass tubes

3. Measuring tube antennas

Bypass tube and stand pipes are used for the following reasons:

102

Horn antenna radars are most com-monly used in measuring tube levelapplications. Stilling tube internaldiameters can be 40 mm (1 ½"), 50 mm(2"), 80 mm (3"), 100 mm (4") and 150mm (6"). Larger tubes are possible.

Normally, the 40 mm and 50 mmtubes do not require a horn. The PTFEor ceramic waveguide impedancematching cone can be installed directlyinto the tube.

For 80 mm and above, the appropri-ate horn antenna is attached and this isdesigned to fit inside the tube.

As discussed in Chapter 2, Physicsof radar and Chapter 6, Radar levelinstallations, the linear polarization ofthe radar must be directed towards thetube breather hole or mixing slots, ortowards the process connections in thecase of a bypass tube.

Measuring tube radar 1 - horn antennas Fig 5.26 Installation of horn antenna radars into stand pipes or bypass tube

DN50

∅ 50 ∅ 80 ∅ 100 ∅ 150

DN80 DN100 DN150

5. Radar antennas

103

The standard length dielectric rodantennas should not be installed withinmeasuring tubes. There is a high levelof ‘ringing’ noise which severelyreduces the efficiency of the antenna.

However, a special design of short,offset rod antenna can be used on smalldiameter tubes (50 mm and 80 mm).

This design is similar in constructionto rod antenna design 3. All wettedparts are in PTFE and the short antennais off centre. This asymmetric designproduces improved signal to noiseratios within a measuring tube.

Measuring tube radar 2 - offset rod antennasFig 5.27 Offset rod antenna for use on 50 mm and 80 mm measuring tubes

1. HF cable

2. Signal coupling

3. PTFE faced flange

4. Offset short solid PTFErod antenna

1

23

4

The speed of microwaves within ameasuring tube is apparently slowerwhen compared to the velocity in freespace. The degree to which the runningtime slows down depends on the diam-eter of the tube and the wavelength ofthe signal.

The microwaves bounce off thesides of the tube and small currents areinduced in the walls of the tube. For acircular tube, or waveguide, thevelocity change is calculated by thefollowing equation :

Fig 5.28 The transit time of microwavesis slower within a stilling tube.This effect must be compensatedwithin the software of the radarlevel transmitter

cwg is the speed of microwaves inthe measuring tube / waveguide

co is the speed of light in freespace

λ is the wavelength of themicrowaves

d is the diameter of the measur-ing tube

Microwave velocity within measuring tube

cwg 1 -=2

2co xλ

1.71d( }{ )[Eq. 5.3]

104

5. Radar antennas

105

There are different modes of propa-gation of microwaves within a wave-guide. However, an important value isthe minimum diameter of pipe that willallow microwave propagation.

The value of the critical diameter,dc , depends upon the wavelength λ ofthe microwaves: The higher the fre-quency of the microwaves, the smallerthe minimum diameter of measuringtube that can be used.

Equation 5.4 shows the relationshipbetween critical diameter and wave-length. For example, 5.8 GHz has awavelength λ of ~ 52 mm. The mini-mum theoretical tube diameter isdc = 31 mm

With a frequency of 26 GHz, awavelength of 11.5 mm, the minimumtube diameter is dc = 6.75 mm. In prac-tice the diameter should be higher. Thediameter for 5.8 GHz should be at least40 mm.

Fig 5.29 Graph showing the effect of measuring tube diameter on the propagation speedof microwaves

Higher frequencies such as 26 GHzwill be more focused within largerdiameter stilling tubes. This will min-imise false echoes from the stilling tubewall.

The installation requirements ofradar level transmitters in measuringtubes are covered in the next chapter.

% s

peed

of l

ight

, c

Tube diameter / wavelength, d / λ

0.60

20

40

60

80

100

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0

dc =λ

1.71

[Eq. 5.4]

106

The subject of this book is radarlevel measurement in process vessels.

Although they are usually applied tocustody transfer applications and notprocess vessel applications, the subjectof antennas would not be completewithout discussion of parabolic anten-nas.

The parabolic antenna is well knownto all. The parabolic form is widelyseen from satellite television dishes andradio telescopes to car headlights andtorch beams.

The main structure of a parabolicantenna is the parabolic reflector dish.This is usually of stainless steel con-struction and is designed to focus themicrowaves as accurately as possible.

The microwaves are fed through thecentre of the dish to the primary anten-na that is in front of the dish at thefocus. The microwave energy is trans-mitted from the primary antenna backtowards the parabolic dish, the sec-ondary antenna, which reflects theenergy and focuses it towards the prod-uct being measured.

Fig 5.30 Typical parabolic antenna

4. Parabolic dish antennas

1

2

34

1. Feed from microwavemodule

2. Parabolic reflector -secondary antenna

3. Primary antenna

4. Focus of parabolicreflector

5. Radar antennas

107

The reflected energy is captured bythe dish and focused back to the prima-ry antenna for echo analysis.

Parabolic antennas are used widelyin custody transfer applications and arewell proven in large storage tanks.

The benefits of parabolic antennas inthese applications are clear. The goodfocusing of the paraboloid shapeensures high antenna gain or directivi-ty. Also this narrow beam angle resultsin higher sensitivity.

However, parabolic antennas arelarge, heavy, relatively complex andexpensive to manufacture. These fac-tors limit the use of parabolic antennasin most process level applications.

The central feed to the primaryantenna at the focus of the dish causesa blind area directly in front of the

antenna. This can reduce the antennaefficiency.

Parabolic antennas have beenapplied to bitumen storage tanks wherebuild up on the parabolic dish is said tocause minimum signal attenuation. Ifthe primary antenna was coated in vis-cous product, this would cause a majorproblem to the signal strength.

In conclusion, the parabolic antennahas a niche application in fiscal mea-surement of large, slow moving prod-uct tanks, but is not suitable for thearduous conditions that are prevalent inthe wide variety of vessels within theprocess industries.

Pic 1. Parabolic antennas have beenaround since the beginning ofradar

108

Planar array antennas were original-ly designed and built for aerospaceradar applications. When the nose coneof a modern jet fighter is removed, itreveals a flat circular disk faced withdielectric material and covered withsmall slots instead of the more ‘tradi-tional’ parabolic metal dish. This flatdisk is typical of the planar array anten-nas which have been developed for useon radar level transmitters.

Planar array antennas have theadvantage of being relatively small andlight in weight especially when com-pared with parabolic antennas.

The construction of a planar arrayantenna for a radar level transmitter isquite complex. The antenna is backedwith a round stainless steel disk thatprovides rigidity and strength to theassembly. The steel disk is faced with amicrowave absorbing material. Thismaterial ensures that the microwaveenergy is directed towards the processand that there is no ‘ringing’ noiseinterference from microwave energybouncing off the steel back plate.

Fig 5.31 Planar antenna - side view

5. Planar array antennas

1. Electronics housing

2. Process flange

3. Antenna feed

4. Stainless steel back

6. Microwave patches

7. PTFE process seal

5. Microwave absorbingmaterial

1

2

3

4

567

5. Radar antennas

109

The microwaves pass in a commonfeed from the microwave modulethrough the stainless steel and absorp-tion material to a feed network acrossthe area of the planar antenna. A patternof microwave patches are fed from thisnetwork.

There is a pattern of microwave ele-ments across the area of the antenna.Each element is built up of three ormore microwave patches with dielec-tric material between. This forms amultiple microwave array with manyindividual elements transmitting fromthe face of the planar antenna.

Finally, the microwave elements andthe bonding materials that form thestructure of the planar antenna are pro-tected by a PTFE process seal coveringthe face of the antenna. Additional anti-static material is used for hazardousarea applications.

Planar antennas can be designedwith good focusing of the microwavesand minimal side lobes. As well asapplications within vessels, they can beused for measuring tube applications.

Fig 5.32 Cut away of planar array antenna for radar level transmitter

1. Stainless steel back toantenna provides rigidity

5. PTFE process seal withanti-static elements

2. Microwave feed throughantenna back into feednetwork to microwavepatches

3. Microwave absorbingmaterial preventsringing from stainlesssteel back

4. Microwave patches withlow dielectric layersbetween them focus themicrowaves from eachelement of the array

12

3

4

5

At the beginning of this chapter westated that the definition of ‘beamangle’ is the angle at which themicrowave energy measured at the cen-tre line of the radar beam has reducedto 50% or minus 3 dB.

We discussed directivity and antennagain and stated that even the bestdesigned antennas have side lobes ofenergy. The aim is to maximize the

directivity and minimise the effect ofside lobes.

The metallic horn (or cone) antennaand the dielectric rod antenna are themost practical for process level mea-surement. The following pages showantenna radiation patterns for differentantenna types, frequencies and sizes.These can be summarised as follows :

1. Comparison of horn antenna beam angle with horndiameterThe following diagrams show the comparison of 100 mm, 150 mm and 250mm (4",6" & 10") horn antennas at 5.8 GHz

0

30

60

90

120

150

30

60

90

120

150

180


Max.:


:::

-10 0 10 20

0,0 deg.32,1 deg.16,9 dB

14,3 dB

Fig 5.33 Horn antenna100mm (4"),frequency 5.8GHz,beam angle 32°

· Larger horn antennas have more focused beam angles

· Dielectric rod antennas have more side lobes than hornantennas

· For a given size of horn antenna - the higher the frequencythe more focused the beam angle

Antenna energy patterns

110

5. Radar antennas

0

30

60

90

120

150

30

60

90

120

150

180


Max.:


:::

-10 0 10 20

0,0 deg.27,9 deg.20,9 dB

15,4 dB

0

30

60

90

120

150

30

60

90

120

150

180


Max.:


:::

0 10 20 30

0,0 deg.14,9 deg.21,6 dB

20,4 dB

Fig 5.34 Horn antenna150mm (6"),frequency 5.8GHz,Beam angle 27.9°

Fig 5.35 Horn antenna250mm (10"),frequency 5.8GHz,Beam angle 14.9°

111

112

The following show a 5.8 GHz hornantenna compared with a 5.8 GHz rodantenna.

Although the beam angles aresimilar, the rod has more significantside lobes.

2 Comparison of dielectric rod antenna with horn antenna

0

30

60

90

120

150

30

60

90

120

150

180


Max.:


:::

-10 0 10 20

0,0 deg.32,0 deg.14,6 dB

15,2 dB

0

30

60

90

120

150

30

60

90

120

150

180


Max.:


:::

20100-10

0,0 deg.27,9 deg.20,9 dB

15,4 dB

Fig 5.36 Dielectric rodantenna, 5.8 GHz.Beam angle 32°

Fig 5.37 150mm (6"), hornantenna, 5.8 GHz.Beam angle 27.9°

5. Radar antennas

113

The following diagrams show thebeam angle of 26 GHz radar with a40 mm (1½" ) and 80 mm (3") horn

antenna. These should be comparedwith the previous 5.8 GHz hornantenna patterns.

3 Frequency differences and beam angles

0

30

60

90

120

150

30

60

90

120

150

180


Max.:


:::

20100-10

0,0 deg.18,2 deg.17,2 dB

19,3 dB

0

30

60

90

120

150

30

60

90

120

150

180


Max.:


:::

0 10 20 30

0,0 deg.9,4 deg.

22,1 dB

24,3 dB

Fig 5.38 40 mm (1½") hornantenna, 26 GHz.Beam angle 18.2°

Fig 5.39 80 mm (3") hornantenna, 26 GHz.Beam angle 9.4°

Vorwort ixDanksagung xiEinleitung xiii

Teil I1. Geschichte des Radars 12. Physikalische Grundlagen des Radars 133. Radartypen 33

1. CW-Radar 332. FMCW-Radar 363. Pulsradar 39

Teil II4. Radar-Füllstandmessung 47

1. FMCW-Radar 482. Pulsradar 543. Frequenzwahl 624. Genauigkeit 685. Leistung 74

5. Radarantennen 771. Hornantennen 812. Dielektrische Stabantennen 923. Standrohrantennen 1014. Parabolantennen 1065. Planarantennen 108Richtcharakteristik von Antennen 110

6. Installation 115A. Mechanischer Einbau 115

1. Flüssigkeitsanwendungen - Hornantenne 1152. Flüssigkeitsanwendungen - Stabantenne 1173. Allgemeine Einbauhinweise 1204. Standrohre und Bypass-Rohre 1275. Messung durch Behälterwand und Radarfenster 1346. Messung von Schüttgütern mit Hornantennen 139

B. Elektrische Anschlussvarianten 1411. Nicht-Ex-Anwendungen 1412. Geräte für Ex-Anendungen 144

Inhalt

Der richtige Einbau ist für dieFunktion eines Füllstandradars vonsehr großer Bedeutung. Obwohl dieSignalverarbeitungssoftware modernerGeräte inzwischen auch schlechte

Echoverhältnisse zuverlässig auswertenkann, ist dies immer noch die wichtig-ste Vorausetzung für eine funktionie-rende Messung.

6. Installation

Mechanischer Einbau

Abb. 6.1 Abb.6.2

KorrekterEinbau

FalscherEinbau

1. Flüssigkeitsanwendungen - HornantenneStutzen / Muffen

Üblicherweise werden Radarsen-soren auf einem Behälterstutzen odereiner Muffe installiert. Referenzpunktfür die Messung ist die Unterseite desGeräteflansches.

Die Vorderkante der Hornantennesollte immer mindestens 10 mm ausdem Stutzen heraus in den Behälterragen.

Die Hornantenne eines Gerätes miteinem Flansch DN 150 (6") ist z.B. 205 mm lang. Ist der Montagestutzendeutlich länger als 195 mm, sollte eineHohlleiterverlängerung verwendet wer-den. So kann garantiert werden, dassdas Ende der Hornantenne über denStutzen hinausragt.

115

10 mm

116

Eine Hohlleiterverlängerung sollteverwendet werden, wenn ein Radar-gerät mit Hornantenne in einem langenStutzen installiert wird. Hierfür wirdein Edelstahlrohr zwischen den PTFE /keramischen Hohlleiter im Flansch undder Hornantenne montiert. Es ist auchmöglich, die Hohlleiterverlängerungfür einen seitlichen Einbau des Gerätesabzubiegen. Der minimale Biegeradiusfür diesen Antennentyp ist 200 mm, derWinkel sollte nicht über 90° betragen.

Bei der Verwendung eines geboge-nen Hohlleiters ist die Ausrichtung derlinearen Polarisation des Radarswichtig. Die Polarisationsrichtung desRadars sollte horizontal sein, wenn dieBiegung nach unten verläuft.

Verlängerte und gebogene Hohlleitersind für Flüssigkeiten mit gutenReflexionseigenschaften geeignet. Siesollten nicht bei Flüssigkeiten mitniedrigen DK-Werten oder beiSchüttgütern verwendet werden.

Mit einer Hornantenne ist es norma-lerweise möglich, flüssige Medien bisan die Unterkante der Antenne zumessen. Dies ist allerdings nurmöglich, wenn die Flüssigkeit guteReflexionseigenschaften hat.

Das Eintauchen der Antennen in dieFlüssigkeit, eventuell sogar mit Anhaf-tungen, verursacht insbesondere bei 6,3 GHz-Geräten kaum Probleme.

Abb. 6.3: Einbau von Geräten mitHornantenne.

Hohlleiterverlängerung undgebogene Hohlleiter

Minimale Messdistanz beiGeräten mit Hornantenne

6. Installation

117

Eine PTFE-Stabantenne eignet sichgut bei chemisch aggressiven Produk-ten wie Säuren und Laugen. Sie wirdoft in der chemischen und pharmazeu-tischen Industrie benutzt, wo Mischun-gen aus Lösungsmitteln, Säuren undLaugen alltäglich sind.

Die PTFE-Stabantennen mit Tri-Clamp und spaltfreier Dichtkonstruk-tion sind speziell für Anwendungen inder Lebensmittelindustrie und für sterile Behälter optimiert.

Die Stabantenne wird für Flüssig-keiten und Schlämme, aber nicht fürSchüttgutanwendungen benutzt. DerSensor ist meistens in einem einfachenStutzen oder in einer Gewindemuffeeingebaut. Radarsensoren mit Staban-tenne werden passend für geschraubte

Verbindungen wie 1½" (NPT oder G),Flanschanschlüsse von DN 50 (2") bisDN 150 (6") oder hygienische Lebens-mittelanschlüsse geliefert.

Beim Einbau ist wichtig, dass derkomplette konische Teil der Antenneaus dem Stutzen in den Behälterragt.

Für den Einbau in langen Stutzensind Stabantennen mit unterschied-lichen inaktiven Längen verfügbar.Typische Längen für diesen inaktivenTeil, und somit die maximale Längedes Stutzens, sind 100 mm und 250 mm.

Abb. 6.4: Typische Einbaueiner Stabantenne: Der aktivekonische Teil der Antenne musskomplett in den Behälter ragen.Für längere Stutzen solltenAntennen mit inaktiver Längeverwendet werden.

2. Flüssigkeitsanwendungen - StabantenneStutzen / Muffen

118

Wenn der konische Abschnitt einerStabantenne in einem Stutzen montiertwird, erzeugen die abgestrahlten Mikro-

wellen ein starkes Rauschen (Klingeln).Dies führt speziell im Nahbereich zueiner Verringerung der Messsicherheit.

Abb. 6.5:

Richtig:Antenne mit angepassteminaktiven Teil für langeStutzen. Normale Rauschkurve mitdeutlichem Echo.

Abb. 6.6:

Falsch: Kurze Stabantenne in einemlangen Stutzen. Produzierthohes „Klingeln“. Im Nah-bereich kann dies sogar dasEcho überdecken.

Falscher Einbau einer Stabantenne

6. Installation

119

Stabantenne direkt auf demBehälter

Radarsensoren mit Stabantenne kön-nen direkt in eine Öffnung in derDecke eines Tanks montiert werden.Dies kann entweder über einen Flanschoder ein Einschraubgewinde gesche-hen.

Maximale Füllhöhe bei einerStabantenne

Wie bereits erklärt, ist es wichtig,dass der konische Abschnitt einer Stab-antenne komplett innerhalb des Behäl-ters ist. Die Gerätesoftware kann das„Klingeln“ bei einem falschen Einbaunicht eliminieren. Eine Erhöhung derVerstärkung würde dies noch weiterverschlechtern.

Die Länge der Stabantenne ab demFlansch bestimmt die maximale Befüll-höhe im Behälter. Im Idealfall solltedas flüssige Füllgut die Stabantennenicht berühren. Allerdings ist diesmanchmal unvermeidlich, hierbei mussFolgendes in Betracht gezogen werden.

Mechanische BelastungEs sollte beachtet werden, dass die

PTFE-Antennen nur beschränktenmechanischen Belastungen widerstehenkönnen. Beim Auftreten einer Quer-kraft kann sie sich biegen und verfor-men oder sogar brechen. Hat dieAnwendung starke Füllgutbewegun-gen? Kann die Biegekraft Schaden amStab verursachen?

Anhaftungen auf der StabantenneWie schon erklärt, werden die

Mikrowellen bei einer Stabantennevom konischen Abschnitt des Stabsausgesandt. Taucht nun der Stab in eineviskose Flüssigkeit ein, und dasProdukt bildet auf der Antenne einenÜberzug, so gefährdet dies dieMessung. Bilden sich starkeAnhaftungen, dann wird das Radarnicht mehr funktionieren.

Berühren niedrigviskose Flüssig-keiten wie z.B. Lösungsmittel oderwasserbasierende Produkte die Stab-antenne, kann dies sogar einenSelbstreinigungseffekt haben und dieMessung bleibt stabil. Bei solchenMedien kann die Antenne bis zurHälfte eintauchen. Jedoch ist auch hierschon mit deutlich verringerterMesssicherheit und Genauigkeit zurechnen.

Nach Möglichkeit sollte einEintauchen der Antenne gänzlich ver-mieden werden.

120

Abb. 6.8: Dieser Effekt tritt auf, wenn dasGerät in der Spitze eines gewölbtenDeckels montiert werden.

Abb. 6.7: Die ideale Position für dasGerät ist bei Behältern mit gewölbtemDeckel bei der Hälfte des Radius.

Montage in Behältern mitgewölbtem Deckel

Ein Radarsensor sollte nicht imZentrum eines gewölbten Deckels oderzu nahe an der Gefäßwand montiertwerden. Die ideale Position ist unge-fähr ½ Radius von der Außenwand ent-fernt. Gewölbte Tankdeckel könnensonst als parabolischer Reflektorwirken.

Ist der Radarsensor im „Brenn-punkt“ eines parabolischen Deckelsmontiert, empfängt er deutlich über-höhte Vielfachechos. Dies wird ver-mieden, wenn der Sensor wie zuvorbeschrieben eingebaut wird.

ParaboleffektWird ein Radarfüllstandmessgerät

im Zentrum eines gewölbten Deckelsmontiert, empfängt der Sensor starküberhöhte Vielfachechos. Der Effektdieser Vielfachechos kann deutlich aufder Echokurve betrachtet werden. Abb. 6.8 zeigt, dass das dritte Vielfacheeine deutlich höhere Amplitudeaufweist als das erste, tatsächlicheEcho. Dieser Effekt kann auch inliegenden Rundtanks vorkommen.Vielfachechos können bei Pulsradardurch die Software erkannt werden, dasie zeitlich deutlich getrennt sind. Wiebereits in Kapitel 4 beschrieben, istdies bei FMCW ein größeres Problem.

3. Allgemeine Einbauhinweise:Horn- und Stabantenne bei Flüssigkeitsanwendungen

r/2

r

Echokurve

Folgendes sollte bei der Montage eines Radargerätes mit Horn- oder Stabantenne auf einem Behälter berücksichtigt werden.

6. Installation

121

Abb. 6.9: Profile mit ebenen Flächenoder scharfen Ecken verursachenstarke Störechos.

Abb. 6.10: Durch diffuse Reflexion an runden Teilen werden deutlich gerin-gere Störechos produziert.

Abb. 6.11: Ein Streublech verteilt dieMikrowellenenergie zur Seite und reduziert damit die Störechoamplitude.

Ebene Flächen, Einbauten z.B. Ver-steifungen oder auch Einbauten mitscharfen Kanten verursachen großeStörechos. An diesen Objekten werdenhohe Störamplituden produziert. RundeProfile hingegen produzieren eine dif-fuse Reflexion und somit nur geringeStörechos. Sie sind deshalb vom Gerätleichter zu verarbeiten als großeStörechos, die von einer ebenen Flächestammen.

Können flache Reflexionsebenen imMessbereich des Radars nicht ver-

mieden werden, sollten diese mit einemzur Seite ablenkenden Streublech ver-sehen werden. Die dann mehrfachgebrochenen Radarsignale sind in derAmplitude deutlich kleiner und deshalbvon der Software leichter zu verarbeit-en.

Diese Maßnahmen müssen umsogewissenhafter durchgeführt werden, jegeringer der DK-Wert des Produkts istund je höher die Genauigkeitsan-forderungen sind.

Störechos

122

Bei der Einbauposition des Radar-gerätes sollte darauf geachtet werden,dass sich keine Streben und keinBefüllstrom im Detektionsbereich desRadars befinden.

Die folgenden Beispiele zeigen typ-ische Messprobleme und wie sie ver-mieden werden können.

Behälterprofile mit flachen Absätzenrechtwinklig zur Hauptstrahlrichtungdes Radars erzeugen starke Störechos.Durch den Einbau eines Streublechskann die Störechoamplitude deutlichreduziert werden, um somit eine zuver-lässige Messung zu ermöglichen.

Einbauten mit einer rechtwinkligenFläche zum Sensor, z.B. Einlässe,Achsen, sollten mit einem „Dach“ ver-sehen werden (Abb. 6.13). Hiermitwird das Radarsignal ebenfalls ge-streut, die übrigen Störechos könnenvon der Signalverarbeitungssoftwareherausgefiltert werden.

Abb. 6.12: Streublech an einemAbsatz im Behälter.

Abb. 6.13: Streublech auf Einbauten.

Vermeiden vom Störechos Absätze

Einbauten wie z.B. Streben, Leitern,Versteifungen und Sonden verursachenoft Störechos. Durch einen gute Wahlder Einbauposition können vieleStörechos bereits im Vorfeld vermiedenwerden.

Auch Schweißnähte im Behälterkönnen Störechos produzieren. Speziellbei höherfrequenten Radargeräten, dienahe an der Wand montiert sind, können diese die Messung bei einem

schlecht reflektierenden Produkt ge-fährden. Durch Anbringen von kleinenBlechen können diese Störechosverkleinert werden. Die Störamplitudesinkt und kann von derSignalverarbeitung besser verwertetwerden. Bei der Herstellung desBehälters können Störechos durchVerschleifen der Schweißnähte minimiert werden.

6. Installation

123

Abb. 6.14: Der Sensor sollte abseitsvon Einbauten, z.B. Leitern, mon-tiert werden.

Abb 6.15: Winkelbleche anSchweißnähten oder Versteifungenkönnen Störechos reduzieren.

Behältereinbauten

Ist der Radarsensor zu nahe an derBehälterwand montiert, können Pro-duktanhaftungen Störechos erzeugen.

Der Sensor sollte deshalb immer etwasAbstand zur Behälterwand haben. Derideale Kompromiss ist ½ Radius.

Wie schon in Kapitel 2 besprochen,sind die Mikrowellen der VEGA -Radargeräte linear polarisiert.

Obwohl die Polarisation eine grö-ßere Bedeutung in Standrohren undBypassrohren hat, kann sie auch beiAnwendung in „normalen“ Behälternvon Bedeutung sein. Die Amplitude

von Störechos, z.B. von Streben oderder Behälterwand, kann oft durchDrehen des Radarsensors um 45º oder90º reduziert werden.

Die Richtung der Polarisation wirddurch das Einkoppelsystem festgelegt,es ist am Gerät durch die Position desTypenschildes erkennbar.

124

Abb. 6.16: Störechos durch Anhaftungen an derBehälterwand sollten vermieden werden.

Anhaftungen

Polarisation

Bei Flüssigkeitsanwendungen mussdas Radar-Füllstandmessgerät mög-lichst senkrecht nach unten zur zumessenden Oberfläche geführt werden.

Wird das Gerät angewinkelt, sinkt dieEchoamplitude und die Gefahr vonStörechos wächst.

Ein Radarsensor sollte nicht direktüber oder in der Nähe einer Befüllungmontiert werden. Dadurch wird ver-

mieden, dass anstelle der Produkt-oberfläche der Befüllstrom gemessenwird.

6. Installation

Ausrichtung des Radargerätes bei Flüssigkeitsanwendungen (Stab- oder Hornantenne)

Abb. 6.17: Bei Messungen vonFlüssigkeiten muss der Sensorsenkrecht ausgerichtet sein.

Abb. 6.18: Montieren Sieden Radarsensor abseits vonBefüllströmen.

125

Fließende Produkte

126

Wird der Radarsensor zu nahe an derBehälterwand montiert, kann diesstarke Interferenzen verursachen. DieEchos von Anhaftungen, Nieten oderSchweißnähten überlagern sich mitdem richtigen Echo. Es muss aus-reichend Abstand vom Sensor zurBehälterwand eingehalten werden, umdies zu verhindern.

Abhängig von der Antennengrößehaben verschiedene Radarfüllstand-

messgeräte unterschiedliche Öffnungs-winkel (Kapitel 5: Radarantennen).

Im Allgemeinen sollte darauf ge-achtet werden, dass sich die Behäl-terwand nicht innerhalb des 3dB-Öff-nungswinkels der Antenne befindet.Bei ungünstigen Einbaubedingungenbzw. Störungen durch dieBehälterwand können die Mess-verhältnisse durch Verändern derPolarisation optimiert werden.

Abb. 6.19: Richtdiagramm einer Antenne mit 150 mm Durchmesser bei 6,3 GHz.

Sensor zu nah an der Behälterwand

Radar-Füllstandmessgeräte werdenoft für Messungen in Standrohren oder,Bypassrohren eingesetzt. Diese Art vonInstallation kann bei Messungen mitSchaum, starken Turbulenzen, mecha-nisch komplexen Behältern oder beiFlüssigkeiten mit sehr niedrigem DK-Wert notwendig sein. Radar-Füllstand-messgeräte werden oft auch benutzt,um vorhandene Geräte in Rohren zuersetzen, z.B. Verdränger undSchwimmer.

SchaumbildungEin dichter, leitfähiger Schaum aufdem Produkt kann die Messung stören.Unter diesen Bedingungen ist es wahr-scheinlich, dass der Radarsensor dieOberfläche des Schaums messen wird.Es gibt aber auch Anwendungen mitSchaum geringer Dichte, der vonRadarwellen problemlos durchdrungenwird. Allerdings kann hier keinegenerelle Aussage getroffen werden,deshalb muss bei Messungen mitSchaum stets mit Umsicht undErfahrung vorgegangen werden. LassenSie sich bei solch einer Anwendungvom Sensorhersteller beraten.

Flüssigkeiten mit sehr niedrigerDielektrizitätszahlSelbst nichtleitende Produkte undFlüssigkeiten mit äußerst niedrigerDielektrizitätszahl wie z.B. Flüssiggaskönnen in Standrohren trotzdem genauund zuverlässig gemessen werden. Wieschon in Kapitel 5 erklärt, konzentriertdas Standrohr die Mikrowellen underzeugt so ein starkes Echo von derProduktoberfläche. Produkte mitDielektrizitätszahlen bis zu 1,5 könnenso gemessen werden.

Turbulente ProduktoberflächeStarke Turbulenzen, verursacht durchRührwerke oder heftige chemischeReaktionen, beeinflussen die Radar-messung. Ein Standrohr oderBypassrohr mit hinreichender Größeerlaubt eine zuverlässige Messungsogar mit starken Turbulenzen imBehälter. Voraussetzung hierfür ist,dass das Produkt im Rohr nichtanhaftet. Leichte Anhaftungen verur-sachen jedoch in größeren Rohren, z.B.100 mm Durchmesser, kaum Probleme.

Allgemeine Hinweise zurRadarmessung in Rohren

Ein Standrohr muss unten offen seinund sich über dem vollen Messbereichausdehnen (d.h. von 0 % bis 100 %Füllstand). Zum Druckausgleich mussdas Rohr über dem 100 % Punkt eineBohrung besitzen. Ausgleichsbohrun-gen oder Schlitze müssen auf einerAchse liegen und dürfen maximal aufzwei gegenüberliegenden Seiten desRohrs angebracht werden. Die Ausrich-tung der Löcher zur Polarisation mussbeachtet werden, bei VEGA-Sensorenmüssen diese senkrecht unter demTypschild angebracht sein.

Als eine Alternative zum Standrohrim Gefäß kann ein Radarsensor auchaußerhalb des Behälters auf einemBypassrohr installiert werden. DiePolarisation muss wie in Abb. 6.21dargestellt, zu den Prozessver-bindungen ausgerichtet werden.

6. Installation

127

4. Standrohre und Bypassrohre

128

Abb. 6.20: Position von Entlüftungs-bohrung und Polarisation auf einemStandrohr.

Abb. 6.21: Polarisationsrichtung beieinem Bypassrohr.

Abb. 6.22: Installation auf einemBypassrohr. Radarsensoren könnenVerdrängersysteme und Schwimmerproblemlos ersetzen.

E E E

6. Installation

129

Die Sensorpolarisation muss ineinem Bypassrohr in Richtung der Pro-zessverbindungen und in einem Stand-rohr in Richtung der Ausgleichs-bohrungen oder Schlitze ausgerichtetwerden. Die Löcher oder Schlitzemüssen auf einer Achse liegen.

Eine korrekte Polarisation verbessertdie Messung erheblich. Störechos wer-den dadurch reduziert und somit dasSignal-Rausch-Verhältnis optimiert.

Wie bereits in Kapitel 2 und Kapitel5 erklärt, reduziert sich in einem Stand-rohr, abhängig vom Durchmesser, dermaximale Messbereich. Verursachtwird dies dadurch, dass sich die Mikro-wellen im Rohr langsamer, alsLichtgeschwindigkeit ausbreiten. Ineinem Rohr mit 50 mm Durchmesser(2") verringert sich die Laufzeit um 20 % und die maximale Länge beträgtdadurch noch 16 m. Bei einem Rohrvon 100 mm Durchmesser (4")reduziert sich die nutzbare Länge auf19 m.

Standrohr zur Messung von inhomogenen Produkten

Abb. 6.23: Durch Schlitze wird eine gute Durchmischung von inhomogenen Produktenerreicht. Die Polarisation muss in Richtung der Schlitze ausgerichtet werden.

Polarisation Laufzeitänderung derMikrowellen

130

Um Messprobleme und Messfehlerbei der Messung von anhaftendenProdukten in Standrohren zu vermei-den, sollte das Rohr einen Innen-durchmesser von mindestens 100 mm(4") haben. Sollen inhomogeneProdukte oder Produkte gemessen wer-den die eine Trennschicht ausbilden,muss das Standrohr Löcher oder langeSchlitze haben. Diese Öffnungen stel-len sicher, dass die Flüssigkeit durch-mischt wird und, dass sie sich an denrichtigen Füllstand angleicht. Je inho-mogener das Produkt, desto mehr Öff-nungen müssen vorhanden sein.

Die Löcher und Schlitze müssen ausGründen der Polarisation in zwei um180º versetzten Reihen positioniertwerden. Der Radarsensor muss so aus-gerichtet werden, dass die Polarisationin Richtung der Löcher ausgerichtet ist.

Zur Abtrennung des Rohrs bzw. desMessgeräts vom Prozess kann einKugelhahn verwendet werden. Mit demKugelhahn ist es möglich, Wartungs-arbeiten durchzuführen, ohne denBehälter zu öffnen. Dies ist beiFlüssiggas und giftigen Erzeugnissenbesonders wichtig. Bei geöffnetemVentil sollten möglichst keine Kantenim Durchlass zu sehen sein, dies würdesonst zu Störechos führen.

Abb. 6.25: Mit einem Kugelhahn kannder Radarsensor vom Behälter getrenntwerden, ohne den Behälter zu öffnen,bzw. den Prozess zu stoppen.

Abb. 6.24: Die Polarisation muss inRichtung der Schlitze oder Löcher aus-gerichtet sein.

EE

Anhaftende Produkte Messrohr mit Kugelhahn

6. Installation

131

Diagramm 1 (Seite 132)Für Messung in Stand- oder Bypass-

rohren werden Geräte mit Flansch-größen DN50 (2"), DN80 (3"), DN100(4") und DN 150 (6")benutzt.

Diagramm 1 zeigt die Konstruktioneines Stand- oder Bypassrohrs miteinem Rohrdurchmesser und FlanschDN50.

Das Standrohr muss innen glatt sein(Rauhigkeitswert Rz < 30). Ideal ist eindurchgehendes Rohr ohne Verbin-dungsstellen im Messbereich. Werdengrößere Rohrlängen benötigt, solltendie Teilstücke mit Vorschweißflanschenoder Rohrverschraubungen verbundenwerden. Hierbei ist jedoch darauf zuachten, dass die Stoßstellen möglichstspaltfrei und ohne Durchmessersprungausgeführt werden. Beim Schweißendarf kein Verzug entstehen, dieRohrstärke muss angepasst werden, umnicht durch das Rohr durch-zuschweißen.

Rauhigkeiten und Schweißnähte imRohr müssen sorgfältig entfernt wer-den. Diese würden sonst Störechosverursachen und Anhaftungen begün-stigen. Schlitze und Löcher müssensorgfältig entgratet werden.

Diagramm 2 (Seite 133)Diagramm 2 zeigt die Konstruktion

eines Standrohrs für einen Radarsensormit einem DN100 (4") Flansch.

Radarsensoren mit Flanschen vonDN80 (3"), DN100 (4") und DN150(6") müssen zur Messung im Standrohrmit einer Hornantenne ausgerüstet sein.Der Antennendurchmesser sollte hier-bei möglichst nahe am Innendurch-messer des Rohrs liegen. Zur Messungin Rohren DN50 und DN80 sindspezielle Stabantennen vorhanden. DieFlanschverbindung zum Gerät ist nichtmehr kritisch, da sie hinter derAbstrahlebene der Antenne liegt.

Bei starker Bewegung im Behälter(z.B. Rührwerk) muss das Standrohrentsprechend befestigt werden, dies giltauch für sehr lange Rohre.

Bei der Messung von Flüssigkeitenmit niedrigem DK-Wert kann oft derNullpunkt nicht sicher gemessen wer-den, oder es kommt zu starkenMessfehlern im Bodenbereich.Ausgelöst wird dies dadurch, dass dasEcho des Behälterbodens hinter demRohrende ein stärkeres Echo erzeugt,als das Produkt selbst. In solchenAnwendungen kann der Einbau einesStreublechs am Ende des Rohrs vonVorteil sein. Die Mikrowellen werdenhiermit zur Seite abgelenkt und dasstarke Bodenecho hierdurch vermieden.Allerdings geht dadurch am RohrendeRaum verloren, da außerhalb des Rohrsnicht gemessen werden kann.

Konstruktionsrichtlinien für Standrohre

132

Diagramm 1

Abb. 6.26

~45˚

2.9

150…

500 5…

15

2.9…6100%

0%

RadarsensorVEGAPULS 54

Flansch DN 50

Vorschweißflansch

Schweißungs der Verbindungsmuffe

Schweißung des Vorschweißflansches

Halterung des Standrohres

minimal messbareFüllhöhe (0%)

TankbodenAblenk-platte

Löcher müssen gratfrei sein

Vorschweißflansch

Verbindungsmuffe

0.0…

0.4

1.5…2

Rz ≤ 30

Rohrdruchmesser 50 mm

0.0…

0.4

Alle Abmessungen in mm

~45˚

3.6

150…

500

0.0…

0.4

3.6

100%

0%

RadarsensorVEGAPULS 54

Flansch DN 100

Schweißflansch

Schweißung derVerbindungsmuffe

Schweißung desVorschweiflansches

Halterung des Standrohres

minimal messbare Füllhöhe (0%)

BehälterbodenAblenk-Platte

Löcher müssen gratfrei sein

Vorschweißflansch

Verbindungsmuffe

0.0…

0.4

1.5…2

Rz ≤ 30

5…15

Schweißung desSchweißflansches

Rohrdurchmesser 100 mm

Alle Abmessungen in mm

6. Installation

133

Abb. 6.27

Diagramm 2

134

Abb. 6.28: Gut reflektierende Medien können direkt durch die Behälterwand oder durchein Messfenster gemessen werden.

5. Messung durch die Behälterwand und RadarfensterDie Mikrowellensignale von Radar-

füllstandmessgeräten durchdringendielektrische Materialien wie z.B.PTFE, Polypropylen und Glas. Dies istfür einige Anwendungen sehr wichtig,z.B. bei der Messung von hochreinenFlüssigkeiten in der Pharmaindustrieoder der Halbleiterfertigung, oder beihochaggressiven Produkten in derchemischen Industrie. In diesen Fällenist es aus Sicherheitsgründen und imHinblick auf die Produktqualität vonVorteil wenn der Behälter geschlossenbleibt.

Ein solche Messung ist bei Produk-ten mit guten Reflexionseigenschaftenmöglich, sie können bei geeignetemBehältermaterial direkt von oben,durch die Behälterdecke, gemessenwerden. Produkte mit guter elektrischerLeitfähigkeit und mit einer Dielek-trizitätszahl von mehr als 10 sind dafürgeeignet. Bei Messungen in denen esprozess- oder produktbedingt zu star-ken Niederschlägen oder Kondensationan der Behälterdecke kommt, ist diesesVerfahren mit Vorsicht anzuwenden.

6. Installation

135

Wie Licht folgen auch Mikrowellenden Gesetzen der Reflexion. Obwohlbei geeignetem Behältermaterial dergrößte Teil der Energie durch dieBehälterwand hindurch dringt, wirdimmer ein Teil dort reflektiert. Beiebener Tankdecke und Aufsetzen desRadargerätes auf dem Tank wird dieserTeil der Sendeenergie direkt in dieAntenne zurückreflektiert (Abb. 6.29).Dies führt zu erhöhtem Rauschen imNahbereich.

Die Qualität der Messung wirdverbessert, wenn das Radargerät übereinem schrägen Bereich des Deckels(35º bis 50º) in einem Abstand von ca.400 mm zum Behälter montiert wird.

Der Winkel stellt sicher, dass dieReflexionen von der Tankwand nichtdirekt in die Antenne strahlen und essomit nicht zu Störechos kommt. (Abb. 6.30)

Mit einem Pulsradar kann auchdurch „dielektrische Fenster“ inMetalltanks gemessen werden. DasFenster muss groß genug und sollte im

Idealfall auch angewinkelt sein. Auchhier sollte der Sensor auf Abstand zumFenster montiert werden.

Anmerkung: Prüfen Sie dieBestimmungen für den Einsatz vonRadar-Füllstandmessgeräten außerhalbvon geschlossenen Behältern in ihremLand. Die geltenden Regeln könnensehr unterschiedlich sein.

Abb. 6.29: Eine flache Behälterdeckeproduziert eine Störreflexion direktzurück in die Antenne.

Abb. 6.30: Die Messung über einemangeschrägten Bereich des Behälter-deckels verbessert die Messung deut-

Abb. 6.31: Optimale Installation fürein 6,3 GHz-Radar zur Messungdurch ein dielektrisches Fenster.

Reflexionen an der Behälterwand

Messung durch ein dielektrisches Fenster

136

In einigen Ländern ist es verbotenFMCW-Radar-Füllstandmessgeräteaußerhalb eines Metallgefäßes zubetreiben. In solchen Fällen muss dasGerät, um die Vorteile eines „dielek-trischen Fensters“ nutzen zu können, ineinem metallischen Stutzen über einemKunststoff oder Glasfenster installiertwerden (Abb. 6.32). Dies kann jedocheinen hohen Störpegel verursachen.

Bei Messungen durch ein Fensterkann eine Verbesserung erzielt werden,wenn die Scheibe eine konische Formerhält (siehe Abb. 6.32). Solch eineTrennscheibe kann bei geeigneterDimensionierung als Linse wirken unddie Mikrowellen zusätzlich fokus-sieren. Diese Form begünstigt zusätz-lich das Ablaufen und Abtropfen vonKondensat.

Abb. 6.32

Radar-Sensor

metallischerStutzen

konischeTeflonscheibe

Messung durch ein dielektrisches Fenster

6. Installation

137

Die Wahl der richtigen Material-dicke ist für die Messung durch einFenster sehr wichtig.

Die entstehenden Interferenzendurch das Fenster bestehen aus zweiunterschiedlichen Echos. Das ersteEcho stammt von der äußerenOberfläche des Fenstermaterials, an derdie Mikrowellen ins Fenster eindrin-gen. An dieser erste Oberfläche, demÜbergang von DK = 1 auf den DK-Wert des Fenstermaterials, gibt es eine

180º-Phasendrehung der Mikrowellen.Das zweite Echo, beim Verlassen desFensters, besitzt keine Phasendrehung.Hier geht es von einem dichteren in einweniger dichtes Medium. Durch Wahlder Fensterdicke als λ/2 derMikrowellenfrequenz löschen sichdiese beiden Echos aus (siehe auchKapitel 2).

Abb. 6.33: Die optimale Dicke des Fenstermaterials beträgt λλ/2 der Radarfrequenz.

Dimensionierung des dielektrischen Fensters

GesendeteWelle

Reflexion mitPhasendrehungvon der Oberfläche

Kunststoffdeckel

Reflexion ohnePhasendrehung vonder inneren Oberfläche

D

Sendesignal

Gegenseitige Auslöschung

Reflexion mitPhasendrehung

Reflexion ohnePhasendrehung

{

138

Anmerkung: Die optimale Dicke kann auch durch Aufschichten einiger Lagenidentischen Materials erreicht werden. Die Schichten müssen jedoch ohne Luftspaltaufeinander liegen. Vielfache der optimalen Dicke führen ebenfalls zu gutenErgebnissen, jedoch verursacht die Dicke des Fenstermaterials eineSignaldämpfung.

Die Tabelle zeigt die optimale Dicke für die wichtigsten Kunststoffe und Gläserdie zum Durchstrahlen geeignet sind. Es wird die optimale Dicke für 6,3 GHz und 26 GHz gezeigt.

zu durchdringendes Material εεr optimale Dicke D in mm PE Polyethylen 2,3 15,5 (31; 46,5 …) PTFE (Teflon) 2,1 16,5 (33; 49,5 …) PVDF Polyvinyl ~7 9 (18; 27; 36 …) PP Polypropylen 2,3 15,5 (31; 46,5 …) Borosylikat-Glas 5,5 10 (20; 30; 40 …) Rassotherm-Glas 4,6 11 (22; 33; 44 …) Labortherm-Glas 8,1 8,5 (17; 26,5; 34…) Quarzglas ~4 12 (24; 36; 48…) POM Polyoxymethylen 3,7 12,5 (25;37,5; 50 …) Polyester 4,6 11 (22; 33; 44 …) Plexiglas Polyacrylat 3,1 13,5 (27; 40,5; 54 …) PC Polycarbonat ~2,8 14 (28; 42 ...)

zu durchdringendes Material εεr optimale Dicke D in mm PE Polyethylen 2,3 3,8 (7,6; 11,4 ...) PTFE (Teflon) 2,1 4 (8,0; 12,0 ...) PVDF Polyvinyl ~7 1,8 (3,6; 5,4 ...) PP Polypropylen 2,3 3,8 (7,6; 11,4 ...) Borosylikat-Glas 5,5 2,5 (5; 7,5 …) Rassotherm-Glas 4,6 2,7 (5,4; 8,1 …) Labortherm-Glas 8,1 2 (4,0; 6,0; 8,0 …) Quarzglas ~4 2,9 (5,8; 8,7 …) POM Polyoxymethylen 3,7 3 (6,0; 9,0 ...) Polyester 4,6 2,7 (5,4; 8,1 ...) Plexiglas Polyacrylat 3,1 3,2 (6,4; 9,6 ...) PC Polycarbonat ~2,8 3,6 (7,2; 10,8 ...)

Fenstermaterialien für Radarsender: Frequenz 26 GHz

Fenstermaterialien für Radarsender: Frequenz 6,3 GHz

6. Installation

139

Zur Messung von Schüttgütern wer-den fast ausschließlich Hornantennenverwendet. Dies schließt alle pneuma-tisch beförderten Erzeugnisse wiePulver, Granulate und Körner ein. DieStabantenne hat ihre Stärke inFlüssigkeitstanks.

Die Oberflächen von Schüttgütern inSilos und Behältern sind selten flach.Bei Produkten wie z.B. Pulver oderGranulat sieht das Profil bei Befüllungund Entleerung zumeist unterschiedlichaus. Der Winkel des Schüttkegels hängtvom Produkt selbst, der Füll- undEntleermethode und von Form undAbmessungen des Silos ab.

Radar-Füllstandmessgeräte, ebensowie Ultraschallwandler, sollten außer-halb der Mitte zum tiefsten Punkt desBehälters ausgerichtet montiert werden.Auch hier sollte das Ende des Hornsmindestens 10 mm in den Behälterragen.

Der Radarsensor wird angewinkeltmontiert um immer möglichst senk-recht zur Produktoberfläche zu senden.So wird über die gesamte Füllhöhe diebeste Echoamplitude erreicht.

Der Radarsensor sollte abseits vomBefüllstrom und von Einbauten mon-tiert werden um möglichst wenigStörechos zu erhalten.

Abb 6.34: Für Schüttgutanwendungenwerden Hornantennen verwendet. DieAntenne ist außerhalb der Mitte montiertund zum tiefsten Punkt im Silo aus-gerichtet. Dies ergibt bei verschiedenenSchüttkegeln das beste Messergebnis.

6. Messung von Schüttgütern mit Hornantennen

140

Abb. 6.35 und 6.36: Schüttkegel von typischen Schüttgutanwendungen beim Befüllen undEntleeren.

Bei Anwendungen mit hohen Tem-peraturen oder stark anhaftendenStaubablagerungen auf der Antennesollte diese mit Druckluft oder Stick-stoff gespült werden.

Hierzu wird der Flansch von zweigegenüberliegenden Seiten bis zumKonus der Teflonfüllung durchbohrt.An diesen Stellen kann dann die Luft-bzw. Stickstoffspülung angeschlossenwerden.

Abb. 6.37: Luft- bzw. Stickstoffspülungzum Kühlen und Reinigen der Antenne.

Luft- bzw. Stickstoff

Hohe Temperaturen und anhaftende Produkte

6. Installation

141

In den vergangenen Jahren hat sichdie Auswahl an unterschiedlichenRadar-Füllstandmessgeräten erhöht.Zudem haben sich eine Vielzahl vonelektrischen Anschlussmöglichkeitenfür Standard- und Ex-Anwendungenauf dem Markt etabliert. Dieseumfassen 4 … 20 mA- und ver-schiedene Feldbussensoren. Bei derAuswahl eines Radarsensors müssendie entsprechenden Verkabelungs-kosten berücksichtigt werden.

Seit ihrer Markteinführung habensich eigensichere Zweileiter-Radarsensoren als vollwertiger Ersatzfür traditionelle Sensoren wie z.B.Differenzdruckmessumformer oderVerdränger durchgesetzt. FMCW-Radarsensoren benötigen jedoch nochimmer die erhöhte Energie aus einerVierleiterversorgung. In diesemAbschnitt werden die möglichenBeschaltungskonfigurationen für alleArten von Radar betrachtet.

1. Nicht-Ex-Anwendungena. 4 … 20 mA, Zweileiter-Radarsensor

b. Vierleiter-Radarsensor mit 4 … 20 mA Stromausgang

20/250 VAC / VDC4 … 20 mA

Abb. 6.38

Abb. 6.39

4 … 20 mA, 24 VDC

c. HART®-ProtokollDie meisten Zweileiter- und Vierleiter-, 4 … 20 mA Radar-Füllstandmess-geräte sind mit dem HART®-Protokoll, aufmoduliert auf dem Stromsignal, verfügbar. Dadurch wird Folgendes möglich:- Fernparametrierung mit dem HART®-Handheld Programmiergerät- Einspeisung der HART®-Daten direkt in das Prozessleitsystem- multi-drop Betrieb mit bis zu 16 Sensoren parallel an einem Strang

B. Elektrische Anschlussvarianten

142

Abb

. 6.4

0

d.Fe

ldbu

s (V

BU

S)

bis

zu 1

5 Se

nsor

en p

aral

lel a

uf z

wei

Drä

hten

mit

VE

GA

LO

G 5

71 u

nd E

V-E

inga

ngsk

arte

n m

axim

al 2

55 M

essu

ngen

zus

amm

enfa

ssba

r

VB

US

VE

GA

LOG

571

mit

bis

zu 2

55 S

enso

ren ve

rsch

iede

ne In

dust

rie-S

tand

ard-

Kom

mun

ikat

ione

n

bis

zu 1

5 S

enso

ren

anei

ner

Zw

eidr

ahtle

itung

6. Installation

143

e.Fe

ldbu

s (P

rofib

us P

A)

max

. 32

Sens

oren

gem

eins

am a

n ei

nem

Seg

men

tkop

pler

Abb

. 6.4

1

Pro

fibus

PA

Seg

men

tkop

pler

Pro

fibus

DP

144

2. Geräte für Ex-Anwendungena. eigensicher ia, 4 … 20 mA, Zweileiter-Sensoren mit HART®-Protokoll

Abb. 6.42

b. 4 … 20 mA, Zweileiter-EEx-d-ia Sensoren mit Verkabelung in erhöhter Sicherheit.

- Versorgung 12 bis 36 VDC- Zener-Barriere in integriertem Ex-d Gehäuse,

eigensicherer Gehäuseteil für Sensorelektronik und zur Bedienung- keine zusätzliche Trennbarriere erforderlich

Abb. 6.43

4 … 20 mA, 24 VDC

Ex-Bereich

Ex-Bereich

Nicht-Ex-Bereich

Nicht-Ex-Bereich

4 … 20 mA, 24 VDC Ex eEx d

Ex ia

Zener barrier

Bedienung,Display undElektronikeigensicherausgeführt

Zener-barriere

Ex ia

6. Installation

145

c. Vierleiter, EEx d ia Versorgung- Versorgung 24 VDC- eigensicherer 4 … 20 mA Stromausgang

Abb. 6.44

d. Vierleiter, 4 … 20 mA, Ex e Versorgung- Exd-Gehäuse

e. Vierleiter eigensicher (ib) mit Trennübertrager und Datenkoppler

Abb.6.45

Abb.6.46

4 … 20 mA

eigensicher

Zener barrier

Ex d 24 VDC, Ex e

24 VDC, Ex e4 … 20 mA

Ex d

Ex d

Stromversorgung

Stromversorgung &digitale Kommunikation

4 … 20 mA

Ex-Bereich

Ex-Bereich

Ex-Bereich

Nicht-Ex-Bereich

Nicht-Ex-Bereich

Display oder Signalver-arbeitungs-einheit

Nicht-Ex-Bereich

146

Abb

6.4

7

f.Fe

ldbu

s (V

BU

S)

- m

ax. 1

5 Se

nsor

en a

n zw

ei L

eitu

ngen

in E

x e,

Ver

drah

tung

in e

rhöh

ter

Sich

erhe

it-

sepa

rate

Ene

rgie

vers

orgu

ng d

er S

enso

ren

in e

rhöh

ter

Sich

erhe

it

VB

US

bi

s zu

15

Sen

sore

n an

ein

erZ

wei

drah

tleitu

ng, E

x e

vers

chie

dene

Indu

strie

-S

tand

ard-

Kom

mun

ikat

ione

n

VE

GA

LOG

571

mit

bis

zu 2

55 S

enso

ren

Sep

arat

e S

pann

ungs

-ve

rsor

gung

Ex

e

6. Installation

147

Abb

. 6.4

8

g.Fe

ldbu

s (V

BU

S)

- m

ax. 1

5 Se

nsor

en E

x-e

(je

fünf

Sen

sore

n pr

o St

rang

, dre

i Str

änge

) pr

o V

BU

S-K

arte

- V

erdr

ahtu

ng in

erh

öhte

r Si

cher

heit

ohne

ext

erne

Ver

sorg

ung

VB

US

F

ünf S

enso

ren

an je

der

Zw

eidr

aht-

leitu

ng v

erso

rgt d

urch

die

se L

eitu

ng

VE

GA

LOG

571

Ex

eV

erka

belu

ng

vers

chie

dene

Indu

strie

-S

tand

ard-

Kom

mun

ikat

ione

n

148

Abb

. 6.4

9

h.Fe

ldbu

s (V

BU

S)

- m

ax. 1

5 Se

nsor

en, e

igen

sich

er E

x ia

, pro

Aus

gang

skar

te-

je 5

Sen

sore

n pr

o Z

wei

leite

r- S

chle

ife,

max

. 3 S

chle

ifen

pro

EV

-Kar

te

VE

GA

LOG

571

VB

US

Fün

f Sen

sore

n an

ein

er Z

wei

drah

tleitu

ngei

gens

iche

r

VB

US

vers

chie

dene

Indu

strie

-S

tand

ard-

Kom

mun

ikat

ione

n

6. Installation

149

Abb

. 6.5

0

i.Fe

ldbu

s (P

rofib

us P

A)

- E

x ia

eig

ensi

cher

, max

. 8 S

enso

ren

pro

Zw

eiad

er-S

chle

ife

- V

erbi

ndun

g üb

er S

egm

entk

oppl

er z

u Pr

ofib

us D

P

Ach

t Sen

sore

n an

ei

ner

Zw

eidr

ahtle

itung

eig

ensi

cher

Seg

men

tkop

plerP

rofib

us P

A

Pro

fibus

DP