comparison of performances · for the coupling loss a mobile measurement unit was used which...

RADIO FREQUENCY SYSTEMS

R F S C o n n e c t w i t h t h e b e s t ®

COMPARISONOF

PERFORMANCESOF DIFFERENT LEAKY FEEDERS IN A METRO TUNNEL

RFS kabelmetal

H.-D. Hettstedt,

B. Herbig,

G. Klauke,

R. Nagel

Reprint from the proceedings of the

ITC Conference Basel, November 1994

A4 RZ COMPARISON orange 1.FE 08.07.1999 9:46 Uhr Seite 2

R F S C o n n e c t w i t h t h e b e s t ®

Innens. A4 -Orange 1-3 08.07.1999 9:49 Uhr Seite 1

COMPARISON OF PERFORMANCES

OF DIFFERENT LEAKY FEEDERS IN A METRO TUNNEL

1. ABSTRACT

This contribution deals with the results of a measure-

ment campaign in which the influences of a tunnel on

the electrical characteristics of Leaky Feeders were

investigated. Four different types were tested, two

broadband cables working in the coupling mode from

30 to 2000 MHz, and two cables working in the

radiating mode up to 900 MHz.

2. INTRODUCTION

Leaky Feeders have already been in use for some time in

multiband systems in tunnels and other confined areas

under various conditions. Mostly they are in com-

petition with narrow banded antenna solutions under

the aspect of costs. In train and metro tunnels, cables

seem to be the only solution because of the masking

effects of the trains.

Though a long tradition of successful applications of

Leaky Feeders can be demonstrated by very positive

experiences under very different and sometimes extreme

conditions, often a very diverse opinion on confidence

in Leaky Feeders is met among system engineers. One of

the reasons may be related to the situation that

technical specifications, especially data of coupling loss,

have been gained under very different conditions, thus

the specs could not be compared directly.

In recent years, an international standardisation on

measurement procedures has been achieved, [1]. Thus

coupling loss figures, the most critical parameter, have

become reproduceable from supplier to supplier.

However, these specs are derived from measurements

on testranges in the free space, and one of the oldest

questions remains: how do these specs represent the

performance of Leaky Feeders in confined areas?

Our experiences has been that most of the complaints

have been resulted from incorrect installation or failing

to use the recommended accessories. Intensive investi-

gations on environmental tests confirmed the rule that

the greater the proportion of uncovered sections in the

outer conductor, the higher the sensitivity of the

Feeders is to humidity. Thus it was conclusively shown

that continuously slotted cables show an extremly high

sensitivity, and that RLF-type cables are virtually

unaffected. These new tests and test procedures and

longterm applications have led to a position where it is

possible to find the proper type of Feeder for the actual

purpose and to make a relatively secure prediction of

performance even under extreme conditions.

At least some of the important questions remain. The

first one is, how to qualify and quantify influences of a

tunnel on the electrical characteristics. Another question

is, are the influences on cables in the radiating mode

higher than on those in the coupling mode. This could

lead to the understanding that the newer cables in the

radiating mode lose their regular field characteristics of

the open air and of course their importance, [2]. This

paper shall help to clarify the situation, but cannot give

universal explanations, because only one type of tunnel

was investigated.

1

Text COMPARISON 08.07.1999 9:52 Uhr Seite 1



3. MEASUREMENTS

3.1 Tunnel

In Hanover a metro tunnel has been available for an

intensive measurement campaign. At the Waterloo

Station one line is not used, thus two closed tunnel

sections can be reached via the platform. Fig. 3.1 shows

one of two test areas at a maximum length of

approximately 130 m each in the tunnel end sections.

Fortunately no gravel and rails obstructed the measu-

rements. The tunnel has a rectangular size which differs

in height and width, presenting the opportunity to

investigate different tunnel types. Only the last part of

the sections shows a constant size. The walls are of

smooth concrete, the area dry, but very dusty.

The cables were installed using special clamps on the

outer side wall at a height of 3.3 m, which corresponds

to the window height of a train. Other cable locations,

e.g. under the ceiling, were not permitted. In Fig. 3.1

three different guide lines I,II,III are outlined. Line I was

reserved for an older ALF-type cable only, whose results

are not considered in this report. The synthesiser was

always positioned on the platform side, the load near by

the end wall.

3.2 Equipment

For the coupling loss a mobile measurement unit was

used which consists basically of a computer controlled

ESVD receiver from R&S and a tracking system which

records the relationship between data and location by

40 samples per wavelength, see Fig. 3.2. The antenna

orientations for the different polarisation planes are

illustrated in Fig. 3.3. The software computes diagrams

of coupling loss including the curve of reception

probability with characteristic values of 5%, 50% and

95%. Furthermore it is possible to zoom sections of the

whole measurement run in order to consider the

different tunnel sections separately. For the

measurements of

reflection and cable

losses standard

equipment was used.

48m68m12m

4.5m

7.5m

4.9m

5.85m

I

IIIII

128m

3.3m

5.2m

Side-View

Radiating Cable

Figure 3.1:

Sketch of the

Waterloo Tunnel

in Hanover

2


ESVD / R&S

Computer

Dipol-Antenna Radiating Cable

3.3 m

2 m

Tracking-System

Tunnel Cross Section

λ /2-Dipole (30-to 900 MHz)

hk

d

ha

hk

d

ha

hk

d

ha

perpendicular vertical horizontal

Logarithmic Antenna (above 900 MHz)

hk

d

ha

hk

d

ha

hk

d

ha

horiz. perp. vertical horizontal

(Load-Transm.) parallel

@



Figure 3.2:

Measurement

Equipment

for Coupling Loss

Figure 3.3:

Antenna

Orientations3




3.3 Measurement program

1. Cable-types:

Coupled mode cables: RLF 9/23 , a cable with large slot spacing

R-LFC 78, a quasi-slotted cable

Radiating mode cables: LK 37, RAY 78

2. Frequency steps : 30, 75. 145, 450, 900 MHz with λ/2-dipole 1800, 2000 MHz,

coupled mode cables only with logarithmic-periodic antenna

3. Cable losses and return losses

4. Coupling loss:

Antenna orientations: horizontal parallel, vertical, perpendicular

(dipole)

horizontal perpendicular, horizontal parallel vertical

(log.-per. antenna)

Antenna heights: 3.3 m, same as the cables, similar to free space measurements,

also 2.0 m and 4.0 m

Antenna distances: 2.0 m standard, where possible 2.0 m from opposite wall

The results were investigated for the whole cable runs and for the two different tunnel sections seperately.

5. Field distributions: at heights of 2.0 m, 3.3 m and 4.0 m

Cross section measurements of coupling loss were made in the greater tunnel section perpendicular to the

cable in all antenna orientations.

4. RESULTS

Due to the great quantity of data it is impossible to

report the results in detail in this paper. Thus only an

overview is shown in Fig. 4.1 to 4.8 which seem to be

relevant and representative diagrams.

4.1 Cable loss and return loss

The results do not differ noticeably from the catalogue

values.

4.2 Coupling loss

Fig. 4.1 to 4.4 show the comparison of the coupling loss

in free air and in the tunnel by the relevant values of

reception probability of 5%, 50% and 95%. These

considered measurements were made in the height of

the cable in order to have comparable conditions to the

standardised in free space. Only below 145 MHz greater

differences became clear. Though the tendency of

increasing coupling loss by decreasing frequencies is

normal, measurement errors have not been clarified till

now. Surprising is the fact, that the general behaviour

of each investigated cable type in free air can also be 4




50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz

Antenna Orientation: horizontalFree Air 100m

95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz

Antenna Orientation: horizontalTunnel 0-125m

95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz

Antenna Orientation: verticalFree Air 100m

95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz

Antenna Orientation: verticalTunnel 0-125m

95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz

Antenna Orientation: perpendicularFree Air 100m

95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz

Antenna Orientation: perpendicularTunnel 0-125m

95%

5%

50%

Reception probability at discrete frequencies

Antenna height: 3.3 m, antenna distance: 2.0 m

5

Figure 4.1:

Coupling Loss of RLF 9/23 Cable




observed in the tunnel. Above 145 MHz the differences

are typical in the range of ± 5 dB which seems to be

related to the standing waves in the cross section, see

paragraph 4.3. Figs. 4.5 to 4.7 show the coupling loss

of the RLF 9/23 cable at 900 MHz where the two tunnel

sections are considered separately. The differences are

quite low and the mean values approximately 2 dB

lower, only at parallel antenna orientations the

behaviour is reversed.

These considered characteristics are typical in this type

of tunnel. Results from other locations do not differ

significantly. On the whole, the field characteristics can

be explained as a superposition of free air characteristics

and influences from the standing waves.

4.3 Cross section measurements

Fig. 4.8 shows the coupling loss in the cross section of

the large part of the tunnel. It was measured at an

antenna height of 3.3 m at 900 MHz at all antenna

orientations. Evidently the RLF-type cable excites an

electrical field of a standing wave. At greater distances

than 2.0 m a field distribution with a regular periodicity

of one wavelength can be observed of which the mean

value is quite similar to that of measurements along the

cable at equivalent antenna positions, compare Fig. 4.1.

The deep and sharp minima should be noted. These

field characteristics are typical for other antenna heights

and frequencies as well. They demonstrate that small

deviations in distance from the cable can lead to large

changes in field strength.

6

5. CONCLUSIONS

This particular rectangular metro tunnel influences the

field characteristics of the investigated cables installed

on a side wall. The relevant values of reception pro-

bability are comparable with those measured in the free

space. The maximum differences are in a typical range

of ± 5 dB above 145 MHz. The influences on cables in

the radiating mode are not greater than on cables in the

coupled mode. The coupling loss measured in the

tunnel section of smaller size is approximately 2 dB

lower. A field of a standing wave in the cross section

leads to a constant mean level which is similar to that

measured at 2.0 m distance along the cable. The

deviations are periodic with the wavelength.

6. ACKNOWLEDGEMENTS

The authors would like to extend their thanks to the

UESTRA AG, Hanover, for their kind permission to use

the tunnel and to Mr. Witwer for his good collabora-

tion. Furthermore, they wish to thank Mr. Mahlandt

from the Cable Development Department for his helpful

advice and for his supply of software. Mr. H. Cyriacks is

also thanked for preparing the large quantity of data.

7. REFERENCES

[1] IEC: Recommendations of Subcommittee 46 A., 1993

[2] H.-D. Hettstedt: Development and Applications of

Leaky Feeders,

International Seminar on Communications Systems

For Tunnels, London, 1993




50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 500 1000 1500 2000

Frequency / MHz


95%

5%

50%


Antenna height: 3.3 m, antenna distance: 2.0 m

7

Figure 4.2:

Coupling Loss of R-LCF 78 Cable




50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%


Antenna height: 3.3 m

Antenna distance: 2.0 m, λ/2-dipole

Figure 4.3:

Coupling Loss of LK 37 Cable

8




50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%

50

60

70

80

90

100

Co

up

lin

g L

oss

/ d

B

0 200 400 600 800 1000

Frequency / MHz


95%

5%

50%


Antenna height: 3.3 m

Antenna distance: 2.0 m, λ/2-dipole

9

Figure 4.4:

Coupling Loss of RAY 78 Cable




Antenna: 3.3 m height, 2.0 m distance,

horizontal orientation

Figure 4.5:


at the two Tunnel Sections

from 20 to 80 m and 80 to 125 m

10




Antenna: 3.3 m height, 2.0 m distance, vertical orientation


vertical orientation

Figure 4.6:




11





perpendicular orientation

Figure 4.7:




12




Figure 4.8:

Cross Section

Measurements

at 900 MHz,

Antenna Height: 3.3 m13




Distance

between

leakages

d → 0

d << λ

d in order of λ

d >> λ

Geometrical

characteristics

Continuously

slotted

Quasislotted

Special slot

arrangements

Large distance

between slots

Electrical

field

Noncoherent

interferences

Noncoherent

interferences

Coherent

interferences of

spherical waves

predominate

Noncoherent

interferences

Electrical

mode

Coupled mode

Coupled mode

Radiating mode

Coupled mode

Field

characteristics

Very irregular

electrical field

Very irregular

electrical field

Transverse

electrical wave.

Function of

linear array

Very irregular

electrical field

Table 3.1:

Overview of modes

14




15


R F S C o n n

R F S G R O U P M E M B E R SR F S G R O U P M E M B E R S

S A L E S O F F I C E S

2299.050.PM 171 001 55 - 01

Please visit us on the internet at http: //www.rfs-group.com

FRANCE: RFS FranceTel: +33-1 3423 6200Fax: +33-1 3423 6324

BRAZIL: RFS Brasil - kmPTel: +55-11 7961 2433Fax: +55-11 494 2937

USA: RFS CablewaveTel: +1-203 630 3311Fax: +1-203 821 3852

AUSTRALIA: RFS AustraliaTel: +61-3 9761 5700Fax: +61-3 9761 5711

GERMANY: RFS kabelmetalTel: +49-511 676 2520Fax: +49-511 676 2521

INDONESIA: RFS IndonesiaTel: +62-21 797 60 76Fax: +62-21 797 60 77

THAILAND: RFS ThailandTel: +66-2 273 8051-2Fax: +66-2 273 8053

CHINA: RFS BeijingTel: +86-10 659 00370Fax: +86-10 659 06932USA: RFS Cablewave

Western RegionTel: +1-510 661 9620Fax: +1-510 661 9630

USA: RFS CablewaveEastern RegionTel: +1-203 630 3311Fax: +1-203 821 3852

USA: RFS CablewaveSoutheast RegionTel: +1-770 649 9929Fax: +1-770 649 8884

USA: RFS CablewaveMidwest RegionTel: +1-630 516 0106Fax: +1-630 516 0207 UK: RFS UK

Tel: +44-1494 447 110Fax: +44-1494 442 742

HONG KONG: RFS Hong KongTel: +852-2861 4438Fax: +852-2861 4488

RUSSIA: RFS MoscowTel: +7-095 258 0632Fax: +7-095 258 0633

CHINA: RFS ShanghaiTel: +86-21 5836 8417Fax: +86-21 5836 8291

ITALY: RFS ItaliaTel: +39-039 2 14 85 69Fax: +39-039 2 14 85 53

SINGAPORE: RFS SingaporeTel: +65-272 9233Fax: +65-272 9266

A4 RZ COMPARISON orange 1.FE 08.07.1999 9:46 Uhr Seite 1

comparison of performances · for the coupling loss a mobile measurement unit was used which...

Documents