EPORT CRPL-4-2 ISSUED JANUARY 13, 194

of cta^idrii'ds

Llbr^^nf, W* Bl'Jj. NOV 1 6 1949 U. S. DEPARTMENT OF COMMERCE

NATIONAL BUREAU OF STANDARDS

CENTRAL RADIO PROPAGATION LABORATORY

WASHINGTON, D. C.

THE TECHNICAL FACTORS INVOLVED IN THE

CHOICE OF A CARRIER FREQUENCY FOR A

WORLD-WIDE LOW FREQUENCY LORAN SYSTEM

BY KENNETH A. NORTON

SUMMARY Am COMCLUBI*

This paper deals with the propagational aM ©th^'s* teeimi'eal problesss

involved in L. F, Loran operation on carrier frequencies between 100 ke

and 450 kc. The optimal freqiienej is d©tei*iBia©d when the distase® be™

tween transmitting stations is a maxiimai, consistent fdth reliable

chronisaation, sine© this maxian® separation provides the highest acc'ur*acj

and results in a miniimm number of stations for world»wid© coverage. For

over-land synchronization paths 9 the lower frequencies in this band pro¬

vide the greater separations between transmitting stations, but for

over-sea synchronization paths® the higher frequencies are better^ pro-

vided surface-wave synchronization is employed. Assuming that moat of

the synchronization paths will be over the sea, it is concluded that the

higher frequencies in this band are more favorable with surfaee-wav®

synchronization than the lower frequencies for L. F. Loran, operation.

It is shown in the report that the us© of sky-wave synchronization

should provide a navigational system with the sam® effective accuracy as

surface-wav© synchronization with the ua© ©f considerably fewer transmit¬

ting stations, ©specially in arctic regions. The optiimm frequency for

this more efficient sky-wave synchronization system is probably also in

the higher part of the tend under consideration. It should be pointed

out that a sky-wave synchronization system operating at high latitudes

will be subject to interruptions duo to ionospheric storms| such ©ffeete

have not been evaluated in this re^rt beeaus© of a lack of adequate data

and, as a consequence, the 99^ reliability shown for a sky-wave synchroni¬

zation system in arctic regions may b© optimistic| sine® ionospheric storms

I

cause a greater degradation in service on the higher frequencies, this

factor should also he kept in aind when evaluating the conclusions

reached in this report®

A discussion of the problem of sky-wave synchronization with J, A,

Pierce of Harvard University brought out the fact that surface-wave

synchronization using cycle matching techniques will probably be more

accurate than sky-wave synchronization with envelope matching only when

the surface waves are comparable in intensity to the sky-waves* Hius, the

maximum range of a surface-wave synchronization system is effectively that

at which fading becomes objectionable and cannot be increased beyond that

point ty*the use of higher powered transmitters. This lends further weight

to the suggestion that the longer range sky-wave synchronization system be

used I the range of sxich systems can be increased considerably with the use

of higher powered transmitters.

Mention is made of the very large improvements in accuracy which are

expected with the use of a wider channel than that used in the present

experimental system. It is hoped that the use of this wider channel will

permit a sufficiently rapid time of rise on the pulse so that the slightly

longer delayed ionospheric waves will not contaminate the surface-wave

with a surface-wave synchronization system, or, alternatively, so that

ionospheric-wave pulses arriving at the synchronization point after one,

two, etc® reflections at the ionosphere will not overlap and thus greatly

increase the s3mchronization errors in a sky-wave synchronization system.

In selecting a frequency for a world-wide Low Frequency Loran System, it

is considered to be very important that this possible requirement for a

very wide channel be kept in mind.

II

It is important to emphasize that, although the conclusions reached

in this report are based upon the best data and theories of propagation

available to the author, nevertheless, this available information is not

considered adequate to form a proper basis for deciding such an important

question as the choice of an optimum frequency for L, F, Loran, This is

due to the fact shown in the report that either night or day propagation

may limit the range of a sky-wave synchronization system under particular

circumstances and this leads to the requirement of a more precise knowledge

of the various propagation factors in order to determine the optimum fre¬

quencies; thus, we must know with accuracy (l) the absolute values of

the day and night sky-wave field intensities, as well as their variations

with frequency, (2) the magnitude of the ratio (determined by measurements

to be approximately equal to 10) between day and night required signal-to-

average-atmospheric-noise ratios, and (3) the influence of latitude on the

intensities of sky-waves (an influence well established in the standard

broadcast band). This requirement of an accurate knowledge of these

three factors arises since a small change in the magnitude of any one of

them can shift the limitation from night to day sky-wave propagation. In

view of these limitations, the conclusions reached should be considered to

be tentative and, if the report serves to guide the efforts of research

workers in this field, it will have served its purpose.

(The present report is a slightly revised edition of a preliminary report originally distributed in January 1947; most of the revisions are in the summary and conclusions and were made on May 29, 1947)

ACKNOWLEDGMENTS

The author wishes to acknowledge the advice and assistance of his

colleagues, William Q, Crichlow and Jack W, Herbstreit, who made m.any

valuable suggestions during the course of this investigation. The typing

and drafting were done by Mrs. Marian D. Adams and Mr, John C, Harman,

III

I f

1

THE TECHNICAL FACTORS IWOLVED IN THE CHOICE OF A CARRIER FI^QUENCI

FOR A WORLD-WIDE LOW FREQUENCY LORM SYSTEM

I RANGE WITH SURFACE-WAVE SYNCHRONIZATION

The factors involved in determining the expected performance of an

L, F, Loran system were discussed in great detail in the reports ”The

Range Reliability and Accuracy of a Low Frequency Loran System*!*^ It is

the purpose of this report to extrapolate the results of that study to

other frequencies within the range 100 kc to 450 kc with th© object of

choosing an optimum frequency for a world-wide L. F» Loran system.

It was shown in the report “Proposed Antenna Design for L, F, Loraif’'

that a steel tower with a height in excess of 500 feet and with umbrella

loading would provide a satisfactory radiator for an L. F. Loran system

operating on 180 kc. Using the data presented in Figure 1(a) of that

report and assuming that the radiated fields for a given inpit power will

be the same for a given electrical length of antenna throughout the fre-

quency range under consideration, we may readily obtain the expected

unabsorbed field intensities at one mil© for an input power of 100 kw at

frequencies throughout this range. These values are given in Table Ip

iJ William Q, Crichlow, Jack W, Herbstreit, Earl M, Johnson, Kenneth A, Norton and Carl E. Smith, Report No, 0RS-P-23si “Range Reliability and Accuracy of a Low Frequency Loran System,” January 1946, prepared in the Operational Research Staff, Office of the Chief Signal Officer, The Pentagon, Washington 25, D. C,

^ Earl M. Johnson and Carl E. Smith, Report No, 0RS-P«22'*2, "Proposed Antenna Design for L, F, Loran," prepared in the Operational Research Staff, Office of the Chief Signal Officer, The Pentagon, Washington 25^ D. C,

- 2 -

the values above 250 kc were estimated by using an unpublished antenna

efficiency study mad© for the Federal Communication^ Commission by-

Kenneth A. Norton and Ross Bateman.

TABLE I

UNABSORBED FIELD INTENSITY AT ONE MILE TO BE EXPECTED FOR A

TRANSMITTER PEAK PULSE POWER OF 100 KW DELIVERED TO A 625 FOOT STEEL

TOWER WITH OPTIMUM UMBRELLA LOADING

kc ffiv/m

100 1230

150 1600

200 1750

250 1850

300 1900

350 1950

400 2000

450 2050

Using the above estimates of radiated power and the methods of

calculation given in a recent paper^ by the author, the curves of surface-

wave“field“int8nsity versus distance shown on Figures 1 and 2 are obtained.

It was shown in the report ORS-P-2 3!/ that the maximum accuracy over

the largest coverage area can be obtained when the L. F. Loran transmitters

are located on the four comers of a square with the distance separation

y Kenneth A, Norton, "The Calculation of Ground Wave Field Intensity Over a Finitely Conducting Spherical Earth," Proc. I.R.E., Vol 29, pp 623-639, December 1941.

» 3 “

between transmitters at the maximum value consistent with reliable syn^

chronization. This inaxifflura distance is determined when the pulse signal

intensities become so weak that they are no longer discernible In the

atmospheric noise for a sufficiently large percentage of the time. With

surface-wave synchronization, this maximum distance is determined by the

summer night noise at low latitudes and the winter night noise in the

arctic since the surface-wave-pulse-signal to atmospheric noise ratio is

a minimum at these times. In the arctic the atmospheric noise level is

higher in the winter than it is in the summer because the ionospheric

propagation of the noise from its point of origin at lower latitudes

involves less absorption in the winter-time, i.e., the change from summer

to winter in the arctic has somewhat the same effect as a change from day

to night conditions at lower latitudes. Figures 2 and 4 of the ORS

Report^ give the required median-peak-pulse-field-lnteasity in microvolts

per meter to permit pulse matching at night for the percentages of time

shown in areas of various noise grades while show the

corresponding distributions of noise grades throughout the world. The

curves of Figures 2 and 4 were derived from the experimental data of the

L, F, Loran survey and thus correspond to a situation where pulses of

varying intensity are observed in noise of varying intensityj they will

thus be applicable to the surface-wave synchronization problem only at the

50% point since the surface-wave pulses will not vary appreciably with

time. However, Figure 50 in ORS-P-23-S^ gives the time distributions

4/ William Q, Crichlow, Jack W, Herbstreit, Earl M. Johnson and Carl E, Smith, Report No. ORS-P-23-S, "Measurement Technique and Analysis of a Low Frequency loran System," prepared in the Operational Research Staff, Office of the Chief Signal Officer, The Pentagon, Washington 25,

- 4 »

of th© noise alone as measured at Dayton, Ohio, and at Galveston, Texas,

and we may use the slopes of these curves rather than those of Figures

2 and 4 to extrapolate from th© 50% point on Figures 2 and 4 to any

higher or lower percentages of the time. If we assume that synchronization

for 99% of the total time is satisfactory, then, since synchronization will

be possible for nearly 100$ of the time during the low latitude winter day

at the same distance where synchronization is possible for 98$ of th©

time during the low latitude summer night, we may achieve the 99$ goal by

rsqt^iring tliat the pulse field intensities be sufficiently strong to per¬

mit synchronization for only 98$ of the summer night hours. The above

argument contains a factor of safety since surface-wave synchronization

will be possible practically 100$ of the daytime hours in both winter and

summer. Using the data of Figures 2 and 4 of ORS-P-23 and of Figure 50

of 0RS-P-23-S, we find that a pulse field Intensity of 157 microvolts per

meter is required on 180 kc in areas of noise grade 1 while a pulse field

intensity of 1290 microvolts per meter is required on 180 kc in areas of

noise grade 3,5 Noise grad© 1 is the lowest to be encountered anjwhere

in the world? while noise grad© 3,5 is not the highest in the world, it

is believed that synchronizing locations can be so chosen that areas of

higher noise grades than 3.5 can be avoided. Thus, all synchronization

paths throughout the world may be expected to exhibit characteristics

intarmeaiate between those snowm for noise grades 1 and 3,5

The next problem is the extrapolation of these required field inten¬

sities to other frequencies in th© band 100 kc to 450 kc. This question

" 5 -

was discussed by the author in a report^ in which it was shown that the

variation of radio atniospheric intensities with frequency depends upon

two factors; (a) the variation at the source and (b) the superimposed

variation due to propagation effects. At the source, atmospheric noise

6/ intensities are known to vary inversely with the frequency."' At night

the propagation of the atmospheric noise is principally by ionospheric

waves, the intensities of which are known to be approximately independent

of the frequency throughout this band. u Thus, the expected atmospheric

noise at night may be expected to vary inversely with the frequency. The

solid curves on Figure 3 are based on the above discussion and show as a

function of frequency the pulse field intensities required on arctic

winter nights or low latitude summer nights, for surface-wave synchronisation

98^ of the time. It should be noted that the required fields shown on

Fig\ire 3 are based on the assumption that the character of the atmospheric

noise is the same at all frequencies throughout the band under consideration,

i.e,, that the same signal-to-noise ratio is required for a given relia¬

bility at all frequencies. This may not be the case. For example, it is

known^ that the required peak-pulse-field-intensity-to-average-atmospherie-

noise ratio for a given degree of reliability on ISO kc is 10 times as high

Kenneth A, Norton, "Frequency Distribution of the Intensities of Radio Atmospherics," paper presented at the U.R.S.I, meeting held in Washington, D. C. on April 27, 1934 and later that same year at the U.R.S.I. meeting held in London.

^ R. K, Potter, "An Estimate of the Frequency Distribution of Atmospheric Noise," Froc, I.R.E., Vol 20, pp 1512-1518, Septeml:>er 1932.

2/ Kenneth A. Norton, "Supplementary Report on Wave Propagation for the CCIR,” Part III, A Theoretical Determination of the Intensity of Sky Waves at Intermediate Frequencies," 1937, Federal Communications Commission, Mimeo No, 84810,

- 6 -

at night as during the day. This factor of 10 is not very well*understood

at the present time but is probably due to the difference in character of

the day and night atmospheric noise, and there is thus some reason to

expect that this factor may vary with frequency throughout the band under

consideration. Using the required field intensities on Figure 3 in con-

j*\xnction with the expected surface-wave field intensities shown as a

function of distance on Figures 1 and 2, we obtain the surface-wave syn¬

chronization ranges shown as solid curves on Figure 4 for transmission

paths over land of average conductivity and on Figure 5 for sea-water

paths. It is evident from these curves that the higher frequencies are

slightly better for over-sea-water propagation while the lower frequencies

are better for over-land propagation,

II RANGE WITH SKI-WAVE SYNCHRONIZATION

During the course of this frequency study, it was discovered that

greatly extended base lines could be used by synchronizing on the sky waves

which are quite strong in these frequency bands at large distances both day

and night. The use of this method of synchronization makes possible much

larger service areeis within which the distance errors will be less than 10

miles for 99^ of the fixes made.

Figures 1 and 2 show the expected median values of the daytime sky

waves. The values shown for 200 kc were estimated from the measurements

made on 180 kc and reported in detail in ORS-P-23-So^ The values shown

on Figure 1 for over-land propagation on the other frequencies were inter¬

polated and extrapolated from this 180 kc data and a series of measurements

» 7 »

made on Broadcast Station WLW operating on 700 kej these latter measiire-

ments were reported in a memorand'um to the Chief Engineer of the Federal

Communications Comirdssion prepared hgr the author. The daytime sky-»wa¥@

field intensities shown on Figure 2 for over^sea propagation on the other

frequencies were Interpolated and extrapolated from the 180 kc data and

the data obtained from a continuous field intensity record of Broadcast

Station KFI operating on 640 kc mad© on a iroyage of the M. S. Jeff Davis

from San Pedro, California, to Honolulu. 1/

At night the sky-wave field intensities may be estimated by means of

the theory in the Federal Communications Commission report^/ already

mentioned. This theory was checked experimentally at 180 kc by means of

the data obtained in the L. F, Loran and was originally derived

to explain standard broadcast band sky-wave field intensities! it would thus

be expected to give reliable results in the band 100-450 kc, Tliis theory-

does not include an allowance for a latitude effect in night sky-wav©

propagation which has recently been established as being of considerable

importance in the standard broadcast band be^ng, in fact, included in the

recently proposed Federal Communications Commission Standards of Good

Engineering Practice for Standard Broadcast Stations? consequently, oui'

‘expected sky-wave synchronization ranges in the arctic may be somewhat

optimistic and may not increase with frequency at a sufficiently rapid rate.

The required median sky-wave fields at night may be determined by in¬

creasing by a factor of 2.55 the surface-wave fields required at night as

given by the solid curves on Figure 3* This factor of 2.55 allows for the

^ Testimony of Kenneth A. Norton, Ross Bateman, and Charles A, E13ert at the Ship Power Hearing, November 14, 1938, before the Federal CoMnuni- cations Commission, FCC Mimeo No, 30539.

8

fading of the sky wa^^es ejad ensures that the received pulses will be

sijfficiently strong to permit synchronization for 9S% of the time for

arctic winters or low latitude summers. Alternatively, the required

sky“wave fields may be determined on 180 kc directly from Figures 2 and

4 of the OBS report,'^

The sky-wave field required in the low latitude suimner daytime on

180 ke may be determined by reference to the 9S% point on the 3.5 noise

grade curve of Figure 1 of 0RS“P*»23^ which shows that a median field

intensity of 165 microvolts per meter is required| the corresponding value

for arctic winter days is 2,6 microvolts per meter as given hy Figure 3

of ORS-P-23.^ During the daytime, since most of the atmospheric noise

arrives at the receiver by surface waves, the intensities of which decrease

rapidly with increasing frequency, the received atmospheric noise intensitie

also decrease rapidly with increasing frequency. This is shown by the

dashed curves of Figure 3. These atmospheric noise intensity variations

with frequency were obtained from the report ^Minimum Required Field Inten¬

sities For Intelligible Reception of Radiotelephony in the Presence of

9/ Atmospheric or Receiving Set Noise," It should be noted that the fre¬

quency distribution in the low noise grade area is steeper than that in the

higher noise grade area; this is due to the fact that the atmospheric noise

is propagated over longer paths in the former case and thus is subject to

5/ more surface-wave attenuation,*^

Using the above data on the expected and required sky-wave pulse field

2/ Radio Propagation Unit Technical Report No. 5, December 1945, prepared under the direction of the Chief Signal Officer by the Radio Propagation Unit, 9463rd TSU, Holabird Signal Depot, Baltimore 19, Maryland,

I

“ 9 »

intensities, w© obtain the day and night sky»wav© synchronisation ranges

shown as dashed curves on Figure 4 for transmission paths over land of

average conductivity and on Figure 5 for sea^water paths* W© see by

Figure 4 that the night sky-wave propagation controls the range over land

at the lower frequencies while the daytime sky-wave propagation controls

the rang© over land at the higher end of the band. Thus, the lower of

the two dashed cumres for a given noise grade on Figure 4 controls xne

effective sky-wav© synchronization range if we assume that service must

be maintained loth day and night. We see Figure 5 that the effective

sky-wave synchronization range over the sea is determined by the expected

rang© at night for all frequencies in this bend. Thus, the highcir fre¬

quencies are superior both for over-sea-water paths and for over-land

paths when sky-wave synchronization is employed.

Transmitters with the capability of handling a pulse peak power of

lOOC kw are now being designed and it is thus desirable to show tL© expected

ranges for such transmitters. The expected field intensities in this

case will b© simplyaTiO times those shown on Figures 1 and 2 and the

corresponding expected distance ranges are given on Figures 6 and 7, It

is interesting to note that the use of one megawatt transmitters shifts the

optimum sky-wave synchronization frequency for over-land paths from about

400 ke (using 100 kw) down to about 250 kc.

10 »“

III ACCURACY OF NAVIGATIONAL FIXES WITH SURFACS-WAVE AW SKI»WA7E

SYNCHRONIZATION

We see "by Figiires 4 and 5 the use of sky-ware synchronisation

increases the expected range substantially in many cases. It thus becomes

of considerable importance to investigate the accuracy to be expected with

a sky-wave synchronization system. It has been showni^ that the variance,

(X g , for a surface-wave synchronized system may be expresseds

(1)

where 0^ and 0^ are the standard deviations of the times of propagation

from the two Loran transmitters to the receiving point and 0^ is the stan¬

dard deviation of the operator errors. In the above, we have omitted the

(usually small) term due to correlation between the times of propagation

from the two transmitters. For sky-wave synchronization, we need only add

to (l) a term 0^ corresponding to the variance of the synchronization path

propagation time,, in order to obtain an estimate of the expected variance,

O' , for delay measurements made on a system employing sky-wave synchroni- O

zation. Thus,

If we assume that the variances for all three propagation paths are nearly

equal, then CT^ = ^2 ~ ^3 estimate the value of 0^ from the known

values of 0^ and Oq by means of the followings

c/ = 0g^+ . 0^^) -- 1.5 (Tg2 - 0.5 O-Q^ (3)

10/ Reference 1, Equations (14), (l5) and (I6)

» 11 -

Table II g3.ves the values of 01 and 0^ as given in Table V of 0RS‘='P‘”23^

together with the estimatea value of (Tg as obtained from (3)s

T.4BLE II

EXPECTED STANDARD DEVIATIONS OF THE DELAY READINGS EXPRESSED IN MICROSECONDS

Surface-Wave Synchronization Sky-Wave Synchronization

CT g % Os %

Day 15.4 9.4 17,7 9.4

Night 22.3 10,9 26,2 10,9

It is evident from the above table that the expected additional random

errors in the delay readings due to the use of sky-wave synchronizatioo are

probably not large and are far outweighed in importance by the much larger

base lines which are made possible thereby.

In addition to these random errors there will be certain systematic

sky-wave delays along the synchronization path but these eauj of coursej

largely be eliminated by calibration.

Figure 8 shows the expected coverage areas based on the data in

Table II for a surface-wave synchronization system on 200 kc compared with

that to be expected using a sky-wave s3mchronized system on 400 kcj results

are given over land and over sea aAd for the two noise grade areas. The

fix accuracy contours shown correspond to regions within which the dis¬

tance error of a fix will be less than the value for which the contours are

labelled for approximately 99% of the time. The methods of calculating such

contours are discussed in ORS-P-2 3I/ and those shown on Figure 6 were

- 12 -

obtained by using a distance error 2,2 times the Yalu© of dj.jj|g expected

at night; this yields a distance error which will be exceeded less than

about 1% of the total time. It is evident from Figure 8 that only about

on©“fourth as many transmitters will be required using sky-wave synchroni¬

zation on AOO kc insteaa of surface-wave synchronization on 200 kc in order

to provide a world-wide L, F. Loran navigation system with a distance

error of less than 10 miles for 99% of the time.

methods are given for calculat¬

ing the expected service areas of all types of radio navigational systems

in terms of the standard deviations of the parameters determining a naviga¬

tional fix. Figure 19 of that report gives the effective service area. A,

of a hyperbolic navigation system consisting of four stations on the foia*

corners of a square, Tliis effective service area is defined to be the

area inside of a contour on which the root mean square distance error,

*^nns* ^ specified value. If we assume that an error of fix less than

10 miles must be attained for 99% of the time, we find by referring to the

results shown on Figure 26 of ORS-P-23^ that the root mean square distance

error, must be less than (l0/2,2) =4.5 miles in those cases where

the ratio of major to minor axes of the error ellipse is near unity. Inside

the square, as may be seen by reference to Figure 28 of ORS-P-23,^ the

angle of cut is always good and this results in an error-ellipse-axis-ratio

near unity. Using the value of dj,jjjg = 4.5 miles and using the values of

0^ and 0^ given in Table II above, we can determine the expected effective

service areas expressed in terms of square base line units from the relation

11/ William Q, Crichlow, ”The Comparative Accuracy of Various Existing and Proposed Radio Navigation Systems,” prepared in the Central Radio Propagation Laboratory, National Bureau of Standards, Washington 25, D. C/(Report No, CRPL-4-1)

„ 13 »

shown on Figure 19 of Report CRP]>4"1»'^^ Tliese ai-e glTen in Table III.

TABLE III

THE EXPECTED EFFECTIVE SERVICE AREAS OF AN L. F, LORAN SYSTEM

(Optimum arrangement of four transmitting stations on the four eorners of a squarej required accuracy throughout the service area corresponds to a distance error less than 10 miles for 99/^ of the navigational fixesj pulse widths and receiver band widths assumed to be the same as those used in the experimental L. F. Loran system.)

Effective Service Area Expressed in Square Base Line Units

Surface-Wave Synchronization Sky-Wave Synchronisation

CT/d ' rms a/b^ cr /d ' rms

1

>

1 >

M

! ' 1

Day 3.39 1.70 3.89 1.55

Night 4.91 1.27 5.76 1,025

The following example will illustrate the us© of the data given in

Table III, Figure 6 indicates that an over-land base line length

B = 815 miles is feasible in the arctic with 1000 kw power using surface-

wave synchronization on 200 kc| thus, a service area A= 1,7 x 815^ =

1,130,000 square miles can be expected in the daytime and © service area

A ■= 1,27 X 815^ = 844,000 square miles can be expected at night. On the

other hand, if sky-wave synchronization with 1000 kw power on 400 kc is

employed, we see by Figure 6 that an over-land base line length B ” 1975

miles is feasiblei thus a service area A = 1,55 x 1975^- 6,050,000 square

miles can be expected in the daytime and a service area A - 1.025 x 1975^™

4,000,000 square miles can be expected at night.

u -

IV THE ITIEQUENCY BAND PEQUIBED BI L. F. LORAN

The above discussions of the expected ranges and accuracies of an

L, F. Loran system were based on the experimental system in use during

1945 in the Caribbean and Eastern United States, The time of rise of the

pulses at the receiver output for this system was so large that delayed

pulses propagated via the ionosphere arrived at the output of the receiver

before the surface-v;ave pulse had risen to its full amplitude and this

contarriinaticn resulted in a very large decrease in the accuracy with

which the total propagation delays could be determined. With the stan-

dai'd loran system, operating on medium frequencies, the time of rise at

the receiver output is sufficiently short and the ionospheric propagation

delays enough longer (due to a slightly higher ionosphere virtual height at

this frequency) so that the pulses arriving via the surface wave and the

waves arriving after one, two, etc. reflections at the ionosphere do not

appreciably over-lap in the receiver output. This made possible a much

higher degree of accuracy in the determination of propagation delays via

either the surface wave or a single ionospheric wave since a match can be

made on either one or the other without the deleterious effects of mi;tual

contamination. However, with the M. F, Loran system, the components of the

transTidtted pulse occupied a larger part of the spectrum than used by the

experimental L, F, system, and thus caused interference throughout a

wider band and the receiver with its wider band accepted noise over a

greater range of frequencies.

This comparative situation for the present M, F. Loran system and

the experimental L. F, loran system is outlined in Table IV, The data for

- 15 -

this table were obtained froBs an IRPL report,"*^ The first line of this

table gives an estimate of the minimuTn propagation delay to be expected

between the first sky-wave pulse and the surface-wave pulse or between

successive sky-wave pulses. The two values of 35 and 55 /Js given for

L, F. Loran are lower and upper bounds to the values that might be encoun¬

tered on frequencies belov; 500 kc, the higher value corresponding, in

general, to higher frequencies. The 62 ps value has been determined with

accuracy at 1950 kc.

If pulse rise times and band widths were obtainable with L, F, Loran

as shown in the last column of Table IV, then it seems quite likely that

the higher accuracy corresponding to the uncontaminated pulses could be

obtained. In fact, it seems likely that pulse rise times twice as wide

as shown in this last column irJght stil] resolve L. F, Loran pulses since

the time of rise in the receiver output would be only 46 ^s and this is

less than the required 55 ps delay; however, this maximum usable time of

rise should be determined experimentaily. The required band width is

approximately inversely projortional to the tim.e of rise of the pulse,

and thus it may be found that band widths half as wide as those shown in

this last column could be used.

To the extent that an uncontaminated pulse L, F. Loran system is

feasible at all, it will certainly be iwich easier to achieve in the higher

part of the frequency band under consideration, having regard to the fact

12/ "Relations Between Band V/idth, Pulse Shape and Usefulness of Pulses in the Loran System," Report No, IRPL-R24, prepared by Interservice Radio Propagation Laboratory, National Bureau of Standai'ds, Washington, D. C,

TABLE I¥

CHARACTERISTICS OF TRANSMITTED Al® HECErVED LORiN PUISES

STANDARD M,F.L0RAN

EXPERIMENTAL L,F,L0RAN

CALCULATED L, F. LORAN , WITH DISCRnaNATIOK AGAINST t DELATED PULSES EQUIVALENT TO M. F, LORAN

Lower Estimate Upper Estimat;

Minimum Expected Propagation Delay

62 35 us 55 us j

Time of Rise(Transmitted Pulse) 12,5 iis 72 us 7,1 us 11.1 us

Base Width(Transmitted Pulse) 64 JUS 330 us 36 us 56,8 us '

Transmitted Pulse Sj«etrum 6 db 33 kc 5,7 kc* 58,5 kc 37,2 kc

Width 20 db 112 kc 19 kc* 198 kc 126 kc

50 db 322 kc 56 kc* 570 kc 363 kc

Receiver Band Width 6 db 80 kc 10 kc 150 kc 90.2 kc

Time of Rise(Receiver Output Pulse)

26 us 176 us** 14.7 us 23,1 us

Base Width(Receiver Output Pulse)

75 us 400 us*** 42,4 us . 66,5 us

^Estimated by assuming that the L. Fa pulse occupies a portion of the spectrum less than M, Fa Loran in proportion to the respective times of rise of the two transmitted pulses,

♦^Estimated by using a band width B = 10 kc x 72/12,5 " 57,6 ke on Figure 25 of Report Noa IRPL‘=R24| and increasing the given rise time of 30,5 IJs in the ratio 72/12,5*

***Estimated by using a band width B = 10 ke x 72/12,5 = 57,6 on Figure 27 of Report No IRPL“R24j, and increasing the given base width of 77 in the ratio 330/64.

. 17 »

that the required band will be a smaller percentage of the carrier fre¬

quency in this case.

Assuming that uncontaminated pulses can be obtained with L, F. Loran,

then the expected accuracy of determination of the propagation delays

would be expected to be the same as that obtainable with M, F, Loran within

its surface-wave range and slightly better than that obtainable with S, S„

Loran employing sky-wave synchronization, due in the latter case to the

expected greater stability of the ionosphere heights at the lower fre¬

quencies, With uncontaminated pulses on L, F, Loran, it is feasible to

expect that a navigational fix distance error of 0,385 miles would be

exceeded less than l/^ of the time for fixes made anywhere on a contour

consisting of a square with the four Loran transmitters on its corners

when these are synchronized with surface waves. The corresponding imcon-

taminated pulse accuracy on this contour employing sky-wave sjmchronizatio;

is only 3,74 miles. These figures serve to illustrate the large increases

in accuracy which might be expected with the uncontaminated pulses.

More transmitter power for the same distance range would be necessary

with the uncontaminated pulse system because of the greater receiver band

width required, and the consequent increase in interfering noise.

FIE

LD

INT

EN

SIT

Y

IN

MIC

RO

VO

LT

S

PE

R

ME

TE

R

EXPECTED LF LORAN FIELD INTENSITIES OVER LAND FOR lOOKW IN A 625 FOOT STEEL TOWER WITH UMBRELLA LOADING

dr=5Xlo''‘^ E.M.U. e ^ \5

10,000

6,000

4,000

2,000

1,000

600

400

200

100

60

40

20

10

6

4

2

1

0.6

0.4

0.2

0.1

0.06

0.04

0.02

0.01

0.006

0.004

0.002

0.001 4,000 6,000 10,000 2,000

pirHrf^ iLirir pT r = 100 kc|H ■'~±rY “tt F r-

uz-ibuj trbi-

lisoF - vfF? . -TT-.-lLJ.;"

L!.: .-l.l.T iL• SF I,'- j '. '.■*

t: ^ 1 1

Ife 200)i -4 IFF'H ^"■;:- ‘: ■ Lk -I -If' tel tf II:

\.aJ - xl

r'H

250li ■ Pfp il;":! n;TiaF 1 1

kip] igfe

if If;

'^'f. ; ■ L Un

500|ffP

350|-rr.F-

ifHM ■ btbi.i.F

-.:.::Pk,:

■•.i.b

n

Tfff

II'

ill

TTT~.

L;:.L::iL-,;,i:

Ppif"

p —K:-ht

Htr-yM

pf

1x41

fii-PiF

m .lL4L:,

k( y\( .-1, . __

f-lp.'-.

I450F 1'.. .-■<■■ -;;:':'-i:'n.F;: Sip ft!;- m MM

T ' - - 1 . ■ 1\’ -V'- A’i^ii ^ rN:TrTT'fT:-H::4;; p-’-rm ' •; n i^i-s 'tH Eli till

Eil" E

m yir

- siHyi

L.'iiiS F.i

A ■-tt’

i.; • ' , [ * t- -- - . ' . ."'

1 ■ ■; 1 •

jk: Hi.Ji.'l

FiTtrft

Iffll

■iFFi .fIf

p te

IfE

St

Ie-

ip

it

iH M

1

g

!§

yi?

1

r:;T’-,Tr~ tFT'T itT77-F.r:. IE: trrpbip

!■ •--Tvrr ’r~::-;-Tn.[:ij\r \J\ FA

: A k;i''p.:4!4p4 Fvf :'.! y[ P; I'-Lj

Jlis

liTUt •! l:- •if it-'i ip-ifet 4Ttfi

TTi;; ̂ ■.hr:F.::,:f,-J; - ,. V- r: • u Hi. \\ ifli

L-iVK.P\P-~“\ rtriit rH •-'. ■"btj;|ir.F- =^atet ’tapr

;pnF j..:i...:,j |''t7

pfi

‘ \: pqTjtHH Hr" l\

■■'Y':. % IH ifiPH -’Ft i-ir;*! pSSiii

•:it.‘:;I'l rH •.-;'-4r--i> ..S-mir ■#li

pltll

TmirSferr- CHi |i

- [ir riiifp v\ V i4jh.iV4 4i%f

fe:p:

P A- HE iE f TiffT iirrnr

ft::!:

L5

|iji-

= ——— kpy.v 1*^ -afiF-TF T- -— :br5q3brr.-;: -•!^TLp l4FL4t4\\ 7p! v\

pjxjLr: Hililll

M;;- pr \ , ip Vi -1 W AA A' HifF HlSf ji.n[^ t4 W-' - : .v ,L^4~L!:L. i;~7i .tn^rr:.^

-Hi- rfSi i.-j'J.ir. V kV' ^T.m kii ''*'M7th7

■ ; r-i ... a.q‘.

rr ?ISi! V-l Wm iM fif Hh:t!

fiH PP

r’^f*"‘r #

% t'St -S-i Htiit bt: Hb ■’tf

;a; Rnl

1—ik.. si

: I- MSu li V-Hf- --f-H-l -\i\’\- ifT-Tr- iiii fetei ifipJ te tiitM

-1 H- r ■ '1 ru.li„: \+V -ftlii V“\H‘ '^\ ' ;YX:4 ixrtTtfk '^T“

pfe tej TTTrZT-'^d ' ■ j—r-y-r.--r4--p-

l-> ! ■ J-M piTTb

» =p.4

TO tyW H

=fcr±t

-:ht if pi tI

ili^ —i+.r44

4-h;'|

‘ • :-r; i Ft "r.-,:.: JiLip iff

a.b \\ : \f\p

r\S

#P e|s

IpIF •H iiii

jii

w jlsj^ ii

Mij

l-M

III

ti HE

1 IK ■H

1“ *n

, pui-jir 2K: Ft fii. if: hp iSir: 4 Jin Sf

r.. ... -r: : . , 1 . i. i.i

b|i ii HiiT .iF..;:'.

L'- ini s 4-H-t f:

ritt' In iF fi]

ttitt. fU

Pii kH*’ rrjjhj i 55

r-T"!—^ . L; L :t .If t:\yrVl 1 1 -1' Ziilg!IP;ii Slim

- zLHi. . ;-vj . ]# ^ni-: t-Up;'.

rTu^iii Fa i'Li;:!:

ifc

T A mti 7UT\T

TT’—n 'iEff

a pi

?:::4( J it im

a

^Tfff D-■ ' ;, i: 1

; r; lifHp

fF.-f: 'V v\- t

"z" V -V-Ji-'•’’h

■7\r“' i aXV \ vM 1.1 \ \\\ 1 1 iVW.iL. 1

Firiijtjffia j:;: pFilifFif

Sri

PF

Wi

PfiHf- T’. .jj'.;',;'. pfuTT.]

■' :;■■ ibrr Tb i.p

r ^ LTitJ tiifl

fw >V,'

-■'ix' PiiPiF

f 'i- :-l P! F:

— ~ : 1 ii-H: y. ly^ — ~

—L.^ :rr'.;:4 -H *~ — —^ i— — •--41-^- .1- I.H. . _ . T- —f——

— —:—r'- -i -r... - -14 - i ■ M' ■ I “' 1 ■ ■ r:.T.'z^ ...— -- —1 r’Y" T" T 1 _..i ; — -4—1 —

IT:

^\h\i

- .•! ^ I i

SURFACE WAVE ^ i

l\ F \ . ..

-PF T" ““

— — DAYTiME SKY WAVE

EXCEEDED 50% OF THE TIME : J ■ ““H yf

V • \

\ mUi'V'

:M

'ii : ■ : J

tef 1:

-n

1

.....

ill £4 -HIM

L-::..,l . \\'U i S

\ V '\"vi\ V T

1 1 J-

..•■ 1 "•fi .;: 1

:;i! V

MILES

FIG. I

FIE

LD

INT

EN

SIT

Y

IN

MIC

RO

VO

LT

S

PE

R

ME

TE

R

EXPECTED LF LORAN FIELD INTENSITIES OVER THE SEA FOR 100 KW IN A 625 FOOT STEEL TOWER WITH UM8RELLA LOADING

^7" = 5X I0“" E.M.U. 6* =80

0.006

0.004

0.002

0.001 400 600 1,000

MILES

2,000 4,000 6,000 10,000

Flfi ?

FiE

LD

INT

EN

SIT

IES

IN

MIC

RO

VO

LT

S

PE

R

ME

TE

R

PULSE FiELD ii^TENSITiES REQUIRED FOR SYNCHRONiZATiON

98% OF THE TIME FOR ARCTiC WINTERS OR LOW LATITUDE SUMMERS

(BASED ON THE PULSE WIDTH AND RECEIVER BANDWIDTH USED IN THE EXPERIMENTAL L. F LORAN SYSTEM)

FREQUENCY IN KILOCYCLES PER SECOND

FIG. 3

DIS

TA

NC

E

IN

ST

AT

UT

E

MIL

ES

THE MAXIMUM DISTANCE AT WHICH L.F. LORAN TRANSMITTERS MAY BE SYNCHRONIZED FOR 99% OF THE TIME WITH THE

TRANSMISSION PATH OVER LAND

100 KW POWER IN A 625 FOOT STEEL TOWER WITH UMBRELLA LOADING

tr = 5X id'"^ E.M.U. ^' = 15

SKY WAVE SYNCHRONIZATION

] /y,?/5(? Grade / Arctic Regions

jNiGHT

Noise Grade 3.5 Low Latitudes

Noise Grade / Arctic Regions

^ NIGHT Noise Grade 3.5 Low Latitudes

1,800

1,600

1,400

1,200

1,000

800

600

400

200

2,000

0 100 150 200 250 300 350 400 450

SURFACE WAVE SYNCHRONIZATION

CARRIER FREQUENCY IN KILOCYCLES PER SECOND

DIS

TA

NC

E

IN

ST

AT

UT

E

MIL

ES

THE MAXIMUM DISTANCE AT WHICH L F. LORAN TRANSMITTERS

MAY BE SYNCHRONIZED FOR 99% OF THE TIME WITH THE TRANSMISSION PATH OVER THE SEA

100 KW POWER IN A 625 FOOT STEEL TOWER WITH UMBRELLA LOADING

Cr =5x lO’" EMU £■ = 80

100 150 200 250 300 350 400 450

CARRIER FREQUENCY IN KILOCYCLES PER SECOND

FIG. 5

DIS

TA

NC

E

IN

ST

AT

UT

E

MIL

ES

THE MAXIMUM DISTANCE AT WHICH L.F LORAN TRANSMITTERS

MAY BE SYNCHRONIZED FOR 99% OF THE TIME WITH THE

TRANSMISSION PATH OVER LAND

TRANSMITTER POWER 1,000 KW DELIVERED TO A 625 FOOT STEEL TOWER WITH UMBRELLA LOADING

cr = 5X id"^ E.M.U e = \ 5

4,000

3,600

3,200

2,800

2,400

2,000

1,600

1,200

800

400

w S- PI

:::; pi 51 xxt 1 X P :g

lip w Pgi g gg Ws^ pgigp

h-t rl- -it !'■ - M-rr ..it. -4i. li" ■■gr|- ---i g $ g --Uii miLi isLiLL^ LtiXid

: g mTI f-'- :g: ;g ::4

g I

tz 1 — SURFACE WAVE SYNCHRONIZATION T-r

irmTjT g ;g; z:^z:z:zi g — ;g g H ± :1 cufv u//iii'c CYur.uankiiTATinAi g

1 1 P -iz

P tFF

iki:.

gjij

m

n:i^

p-n

ik -'-■-ni - f- r

1^ ■4 t i 1

4

1 gtj pip

g

i 44

g

Ppr

4f g gfX

g

ES

t

-4 :a-

1

g

g 1 4ii 4+y

i g gg

trg

1 P

i Ff g --

g

pf

w iV\ t*' Pi

lTt]Ti'.T'

g

rkl.^4 ■+^ 4:

gg-

4-

# tf 4 4

I f 1 i gg 44.

4

-g--

xi; sWz gffi

rl-H-H

P 1 g

•gg i 44 Xg4:

m itli = --:t g # g g

g

14 1 4

n ft 1 r

li s 4 XL FFj-

fp

pdt

fp= pp i 4 i m

Hm i|; 1 pi P 4f p 1 P- :p :i g g

a.

Pig i XftX

xg

ft g Pi

g M

pI -L -

^' j LIMITED BY g

NIGHT SKY WAVE

PROPAGATION li

1 gg

g

pF

4 It 4 g i-r-T-*-

pp

tpp:

4-1-

tp n 5-

g

f LIMITED 8Y 4

1 DAY SKY WAVE 4

+ PROPAGATION 4

p g g

it

Hr!

Pi 1 ilk _

4

■ xt g xpix 4 44 rP TO ,.. p,.. g-itj rr^ "x^ pP 3g

Sc-- nr M :g

±ti: 4: 4i 4==

nit s ix xxp 4

g 4i-

4p i g g tf i-pttg tg-rnr^ g 1 -i-g- m

gg 'gg -tg

iil-- jj inrd ruiRir fpi- g Xp: [1 P?1 P

ggggn ip 4=4 g g nit'

i: TT'iHt

±p: s if g 4 4- i-i

trni 4±p NIG HT

rs^ g g aF

rnT~[1 P 1.;;; z'.\z-^- T#

-+- g-p- .i:: 1 g n

44 44^ ii-t

TTTT Fppx Xpr'i i-gnj gg

g -L-rr-T-*-

S g nr =g-

:g Ip P=' M il

4 A/o He

fe 'TPT g de

pPp4 iP g i

ip Sr gX

m

;:iP: ::±i‘

i s I IHI: g m

Gro

—Lj 4-

'ctic Ri ̂ gic

g

i/?s" rrrr- :±± gg-l

-t4- Ftf

Si

m :g- gtS

-4 5

ig: g 1 g g

t4 PtF r-H-g

-gp-

g gg 1

■gr g

14+ Pi 1—i-F - -j—i- -

g pit -r 1 ' 1 g fr g ppi jpDAY

-la-; 4-rn- gpr Eg :4t -h4

g kh'i gPjP 4p 4:: :;i±: ipT SH!

g g g~4 A-^\+ g xixt g

-. .44- ztz-r g g-g —f pE gg g Xpi:

ttt iiiii Lx-g.ii. g

■g w m g g i^tf g itjp:

x= 4f rrrt xhx Tgr F+t TXT iiiJt ±g -4- pt: -irr gg-f- g.g ss gI7. xgr xii ng" ptiP F^rTT ', ; g gg g;:' 7x3: g

•ff t 5: hr+p = + W: pF ■g g gg

44 TXp:

g Ti; r i if -ig gg Tg- 'gg g F4

I- M P g-

P±; r Tt

xEix W Pitt

g i PF 1 1 yy g fji- --ap Titr xfgr g

Ti'g ~glT ±fi

g 1 --P+-+

=P

lip

:;±; g g;

|j|f

-T*-

g

g g I

t

s s g

pu- y

ti-Ll ffii

g4p

XIXP

1 g

P

P

g F

tpi w. P -jg ::i: XT' g

4 ip g 4 gj g giSp 4== fg 1

14; pH

fcs gg:i; i

i-i -14

XOT

firm -1—i-i- g-l S

xi —1 g g| pf 4: g- g pp pig -ft-'

Ixn: TTm f Ml A A-. -p.

n ■gj -g ipx

AS ;g Nr' --4- ip F ' tjjt,

g T 11 Pf Jx J

I’l—H .X rre- +4+- -4-h —-rt- rill! •ir+f g[-i- g g i 1 -t-4- -tt- g Mill xtd: ffp; XtPt' 11 i gt' g ggt ±±; •gi /yn/Qfi Ornrlet / Arntir. Rfininnczzux gpT Tip' 44F g

g^-,- 4 g NIGHT g g g g g—ri

*1 = pp-t-S

i: ++ff- W 4f i g

g —g-

tP -rr“' g g| 444- g= p gg gip Pp

fg

f|

Ij 1 i 1 g t'P

■M 1 -Pxr:

g- s=

g g rf--t- :4

Tt 4

4 1 4

g

J 1 A/o

rirr XTXt

ise

g

Gra 'de

g g

3.5

pi

il Low Ll

gg pi

7l/lL

FT

g /dee

i -^-n—1

gg g g

li

)AY

m

g

}| if: 1

gg^

p

■4;*

i fo/st

gj

M ? Gr adi

1 j.

=4

g

5 L

1

.0

4

4 ZX

li w

ffl

Lai Hue § les

-p"

y g

14 ±

t

at

X:

p 144

b

i

g

1 fp

w g

Pp

g

xnx

( 44

E g

g

i'ill

• ^ ■ Ij ggt: lit s 1 4f

•4++

g Pill i f w il ip pi

n= ll I4g tp f

Ps g B M ii s s gg: § M g g ii g n]-ri g g ii gi

100 150 200 250 300 350 400 450

CARRIER FREQUENCY IN KILOCYCLES PER SECOND

DIS

TA

NC

E

IN

ST

AT

UT

E

MIL

ES

THE MAXIMUM DISTANCE AT WHICH L.F LORAN TRANSMITTERS

MAY BE SYNCHRONIZED FOR 99% OF THE TIME WITH THE

TRANSMISSION PATH OVER THE SEA

TRANSMITTER POWER 1,000 KW DELIVERED TO A 625 FOOT STEEL TOWER WITH UMBRELLA LOADING

O' =5X lO'" EMU. S = 80

CARRIER FREQUENCY IN KILOCYCLES PER SECOND

EXPECTED ACCURACY OF L. F LORAN FOR THE PRESENT EXPERIMENTAL SYSTEM

OPERATING ON 200 KC INVOLVING CONTAMINATED PULSES COMPARED WITH A

SKY WAVE SYNCHRONIZATION SYSTEM OPERATING ON 400 KC ( PULSES OCCUPYING A NARROW BAND HAVE A TIME OF RISING SO LONG THAT SKY WAVES ARRIVE AT THE RECEIVER

BEFORE THE SURFACE WAVE ATTAINS ITS FULL AMPLITUDE; THE DISTANCE ERROR TO BE EXPECTED IN THIS CASE WILL BE LESS THAN THE VALUES FOR WHICH THE CONTOURS ARE LABELLED FOR 99% OF THE FIXES;

THE SYSTEM IS ASSUMED TO CONSIST OF THE IDEAL L. F, LORAN ARRANGEMENT OF FOUR STATIONS ON THE FOUR CORNERS OF A SQUARE)

SYNCHRONIZATION PATHS OVER

THE SEA

SYNCHRONIZATION PATHS OVER

LAND

f

I I I I I I I

\ V,

10 MILES

NOISE GRADE I ARCTIC REGIONS

10 MILES

NOISE GRADE I ARCTIC REGIONS

NOISE GRADE 3.5 LOW LATITUDES

0 100 200 300 400 500 1000

NOISE GRADE 3.5 LOW LATITUDES

- SURFACE WAVE SYNCHRONIZATION ON200KC

-WITH SKY WAVE SYNCHRONIZATION ON 400KC FEWER STATIONS ARE REQUIRED FOR WORLD

_WIDE COVERAGE_

1300 2000

MILES

■;>' • lA .