January 1939 Radio-Craft
Wax nostalgic about and learn from the history of early electronics.
See articles from Radio-Craft,
published 1929 - 1953. All copyrights are hereby acknowledged.
In 1938, Bell Telephone Laboratories, Western Electric Company,
United Air Lines, and Boeing worked together to developed the
first practical microwave radio altimeter for use in commercial
aircraft. This is not a radar unit in that the distance is not
determined solely by emitting a signal and measuring the time
taken to the target (the ground in this
case) and back again. Rather, the radio altimeter relies
on a heterodyned beat frequency generated between a reference
signal and that of the transmitted and received ground-directed
signal. Author Washburn does a nice job explaining the process,
so I needn't add to it. It is interesting to note the statement
about the 500 MHz used being the "highest frequency ever
to be used for practical purposes."
New Radio Altimeter Increases Air Safety
Last month radio helped immeasurably to promote aviation
safety. A great laboratory, a great industrial plant, and a
great airline combined to develop, build and test an altimeter,
for use on airplanes, that accurately indicates height above
land-level (including buildings and bridges).
R. D. Washburne
News flash: "Wreckage of plane found on side of mountain
by searching party. All aboard apparently instantly killed."
How many times have you read some such tragic report as the
above, in connection with an airplane crack-up on a mountain?
Published figures place this general type of plane casualties
at about 50 per cent; imagine it - one-half of all plane crashes
in which, due mainly to poor visibility (fog, etc.) , the "ship"
smacked head-on into obstructions of which the pilot was unaware.
Too frequently its passengers and crew have passed into Eternity.
How could such a catastrophe occur? The motor was working
perfectly; the pilot was experienced and trustworthy; the radio
equipment and flying instruments were in perfect working order
- except that the altimeter locked in position by the impact
indicated a height of several thousand feet!
Ah!-There is the answer. Several thousand feet - above sea
level... but the plane was flying in mountainous country
and the altimeter did not, could not, indicate height above
The aeronautic-type altitude indicator operates on the principle
of the aneroid barometer. This type of instrument indicates
height (altitude) with respect to a reference level having constant
barometric pressure - a condition which is found only at sea
Engineers have worked unceasingly to develop an altimeter
that would indicate height above land with the same or greater
accuracy now possible over water (or land at water-level).
Last month radio as shown in the heading illustration solved
the problem. The radio-operated "Terrain Clearance Indicator"
proved its ability in direct comparisons with the Standard Altimeter;
the panel meters of both systems were mounted side-by-side as
shown in Fig. A (insert, in heading illustration).
Bell Telephone Laboratories developed this microwave radio
altimeter, Western Electric Co, made the equipment, and United
Air Lines installed the apparatus in a special Boeing twin-engined
airliner. In cooperation, these 3 groups demonstrated the efficiency
of the new radio altimeter in a test flight over New York.
Despite weather conditions or poor visibility, the pilot
could read his height directly and accurately whether he was
several thousand feet high or merely skimming a few feet above
So keen is the sensitivity of the new altimeter that, from
an altitude of several hundred feet, the presence of the George
Washington Bridge was clearly indicated by the meter as the
test ship flew down the Hudson River, far above the actual obstruction
itself. Over the Hudson River both altimeters indicated a height
of about 800 ft.; when the plane swung over the Palisades, the
standard altimeter continued to show a height of 800 ft. but
the Radio Altimeter indicated the fact that the ground was only
250 ft. below the plane!
(No comment has been forthcoming as yet concerning provisions
for indicating conditions dead-ahead - an abruptly-rising plateau,
for instance - but undoubtedly means will be found to take care
of this condition. -Author)
Fig. 1 - Signal C and reflection E, are shown
beamed, only for purposes of illustration. Actually, they are
only semi-directional. Antennas B and F are each 1 ft. long.
This radio altimeter is effective from heights of 100 to over
Extended flight tests of the new development are being made
over regular airways by engineers of Bell Telephone Laboratories
in a special Boeing twin-engined airliner assigned by United
for service testing.
The principle of operation, shown in a general way in Fig.
1, is as follows:
A frequency-modulated, 500-megacycle (0.6 meter) signal - highest
frequency ever to be used for practical purposes, it is said-from
a transmitter (A) is radiated from antenna No. 1 (B) as signal
C to ground (D). The reflected signal (E). is picked up by antenna
No.2 (F) connected to a receiver (G).
Although antenna No.1 is directive toward the ground a considerable
amount of energy "slops over" to antenna No. 2 (note direction
of arrow, H).
This "slop-over" signal combines with the reflected signal
and, in receiver G, produces a beat frequency which is segregated
by an "interference measuring device" (I) and indicated on a
unit, on the instrument-panel in front of the pilot, called
a "terrain clearance indicator" (J).
The beat frequency is a function of the height, .and directly
proportional to it. Therefore the radio altimeter (J) is calibrated
for direct-reading in feet. In order to understand how this
radio altimeter functions it may be well to mention some of
the principles previously employed.
(1) The idea of measuring the capacity between airplane and
ground is not considered practicable due to the fact that capacity
variation is large for low altitudes but very small for high
altitudes. (2) The principle of measuring the phase difference
between a radiated wave and its reflection is difficult to realize
in practice. (3) The scheme of sending an impulse and then determining
the distance as a function of the elapsed time before the reflection
(echo) is received although useful in measurement of ionosphere
heights is not applicable for short distances because of the
minute time element involved. (4). The frequency-modulator radio
wave principle - and the one upon which the new system is based
- affords a continuous and linearly proportional indication
Merely to mix, at the receiver, original and reflected signals
of the same frequency (since there would be no change in frequency
due to reflection from the ground) would not produce a beat.
This is easy to understand if we take the example of a person
whistling a single note uninterruptedly. Unless the continuous
whistle is interrupted for a certain length of time it is not
possible to hear an echo (reflection) should one exist. Interrupting
the whistle and noting the lapse of time until the echo is heard
may be taken to represent the impulse method of determining
Let us suppose that instead of a continuous single note,
a continuous sequence of notes is whistled. To make the point
more clear let us suppose that the note is warbled up and down
the scale, considering however, that in warbling only 2 notes
are sent out.
First we whistle note No.1; then, 2nd, note No.2; 3rd, No.1
again; 4th, No.2 again, and so-on. Now, if we are far enough
from a reflecting surface so that note No. 1 reaches our ear
as an echo just as note No. 2 is being whistled we will be able
to hear not only both notes simultaneously but also a 3rd or
beat frequency due to the heterodyning (mixing together) of
notes 1 and 2. Without going into too much additional detail
let us take for granted that a meter arranged to show the presence
of these beats could be calibrated to indicate the number of
feet that note No.1 had to travel before it heterodyned with
note No.2; for maximum indication on the "beat" meter this will
always be a fixed distance.
Fig. 2 - The frequency-modulated signal is
How then can we obtain beats whose readings on the beat-meter
will be equivalent to other distances?
Well, let us consider a second example in which we wobble
the frequency. 5 notes up and down the scale instead of warbling
only 2 notes. The note sequence would then look like this: No.
Now, if, at the time note No. 4, let us say, is being sent
out, note No. 1 is received, it indicates that the sound had
to travel so far to produce an echo that a 4-note lag resulted;
and since each additional note lag means the sound had to travel
just that much further before the echo could be observed, a
"lag meter" indicating this degree of lag - or difference between
radiated and received frequencies - could be calibrated to read
directly in feet distance. This "lag" may be compared to the
"beat" mentioned in the 1st example; the beat frequency (lag)
therefore is a direct indication of the echo distance.
This becomes quite clear upon observing that, when note No.5
is radiated and note No.2 is being received, there i still a
difference of 4 notes; that is, the echo distance has not changed,
hence, the feet-distance (beat frequency) indication remains
Applying this crude simile to the radio altimeter, let us
consider that the 500-megacycle signal is "wobbled" or frequency
modulated. That is, shifted up and down the radio frequency
(wavelength) scale, smoothly and continuously.
The various beat frequencies then required to indicate equivalent
echo distances (or, in the instance of airplane operation, altitude
above ground, or a bridge, a building, or any other reflecting
surface) are as given in Table 1. The figures are from patents
issued to Lloyd Espenschied and assigned to American Tel. &
With this introduction we are prepared to consider a graph,
Fig. 2, which has been made available by Western Electric Co.
This figure illustrates the basic principle of reflecting a
frequency-modulated radio wave (the speed of which is constant)
and indicating the elapsed time as proportional distance.
If an airplane is equipped with a radio oscillator whose
frequency can be "wobbled" according to the sawtooth curve a1
b1 c1 of Fig. 2, and if a corresponding wave is radiated toward
the ground, some of the energy will be reflected back to the
plane where it will set up a current in an antenna. The frequency
of that current will have a similar wobble, but displaced in
time to the position a2 b2 c2 by reason of the delay in travel
to the ground and back. At any instant the frequency of the
received wave will differ from that sent out by a constant amount
equal to p1 p2.
If the ground falls away or the plane rises, the travel time
will be increased and the received current will be displaced
still further; consequently, the frequency difference at any
instant will increase in proportion to the change in clearance.
The difference in frequency between 2 currents can be measured
by passing them through a modulator tube, and measuring the
frequency of the "difference" component in a frequency meter.
The scale of this instrument is then graduated in feet, and
indicates directly the terrain clearance.
Transmitter and Receiver
Some idea of how microwave signals may be generated and frequency-modulated
at the transmitter, and the reflected and "slop-over" signals
received and the beat frequency detected and made to indicate
on a meter calibrated in altitude at the receiver, may be obtained
by reference to Fig. 3. These illustrations are based in part
on figures in an article, by Sadahiro Matsuo (Faculty of Engineering,
Tohoku Imperial University, Sendaishi, Japan), on a radio altimeter,
in a recent issue of Proc., I.R.E.
Referring to Fig. 3A the sawtooth-waveform modulation, shown
in Fig. 2, is obtained from modulator V2 and applied in required
degree to the grid of V1. (Note - Matsuo's transmitter employed
V1 as a Barkhausen-Kurtz triode oscillator. -Author)
In Fig. 3B, the beat-frequency signal is detected, amplified,
limited to a constant value, and applied to an output meter
(altitude indicator). In order to secure satisfactory operation
of an indicator-type frequency meter, with a constant-output-level
beat frequency input, inverse voltage is fed to tubes V4-V5
by connecting transformer secondaries Sec. 1 and Sec. 2 as shown.
Tubes V4-V5 alternate in conductivity.
Fig. 3. These theoretical, composite diagrams
illustrate the sequence of operations. The frequency-modulated
signal results in a beat frequency that may be resolved on an
output meter into terrain clearance indications in feet.
In addition to the regular meter, the device may also be
equipped with a red signal light which will automatically flash
a warning when the plane descends below a safe predetermined
This radio distance indicator may also be used on shipboard.
In foggy weather the presence of icebergs, other ships, promontories,
etc., could be detected and their exact distance indicated.
In commenting on this latest scientific tool for aeronautical
navigation, W. A. Patterson, President of United Air Lines,
said, "Our engineers and pilots regard the development of this
device as one of the most important technical advancements in
the history of air transportation, and a major contribution
to the safety of scheduled flying. Following completion of service
tests now in progress with the device in United's Flying Laboratory,
we will make these devices standard equipment on every airliner
in our fleet."
Posted August 12, 2015