Rebecca Mk8

Rebecca Mk8
Rebecca was the airborne equipment used during and after WWII to find a small beacon (Eureka),
usually dropped earlier to mark a dropping site for parachutists or equipment.
Mk8 was the last one of the Rebecca series of airborne receiver/transmitters, made in 1956 , probably
by Murphy Radio. The main unit is the TR 8193.
The airborne set (Rebecca) transmits 170 pulses per second on one of 8 channels between 214-235
MHz, with a peak power of 1kW, and a pulse width of 5 us.
The Eureka beacon responds each pulse on a different frequency, usually 5MHz above the Rebecca
frequency, with a 20W pulse of 5us width.
These pulses are received by the Rebecca unit on two aerials, one on each side of the cockpit. Their
relative strength gives the direction, the delay from transmit to receive gives the range.
The T.R 8193 Rebecca Mk8 airborne unit has ( 20 or 200 nm) range and course
measuring capability to Eureka or BABS beacons.
6 November 2014, Koos Bouwknegt
Rebecca Mk8
"The Vampire 11 had a Rebecca 8 set, as did the Jet Provost and Hunter. In 'range only' mode, if you were lucky
it had a range of about 40 nm; in 'homing' mode about 20....
The control box was a large, clunky thing with large
bakelite rotary knobs. One control selected the letter and the
other the number of the associated Eureka 7 channel; for
example, Cranwell was C4 and Cottesmore D5 (I think).
The morse coding rate for the Eur7 was very student
friendly - about 1 word per week! You had to check it as it
was quite common for the Rebecca to lock onto a different
station to the one you'd selected.
Somehow we flew radio navigation exercises using nothing
more than Eur7 and UDF, then flew a DME let down at
base using this contraption. At least we had the left/right
indicator, rather than the CRT screen the guys who flew
Pigs had! You went outbound in the 'DME safety lane', then
faffed with L/R and the DI to establish the correct inbound
approach course....:uhoh: Fortunately we'd all had several
sessions in the Link trainer before trying it for real.
The 'swingometer' range needle used to hunt around the dial unless the Rebecca could sniff out a station; once I
was just pulling off the target at Pembrey when the needle went clockwise at precisely the same rate as the gmeter's needle normally moved - except that it kept
going...:ooh: For a moment I eased off the gas a reflex
in case I was overstressing....fortunately I was already
climbing. We used to pull out of the dive and
immediately check the mirror for the 25 lb bomb smoke
- having the distraction of the Rebecca needle moving
round the dial out of the corner of your eye was most
unwelcome!
The bandwidth of Eureka/Rebecca was colossal - about
4 MHz if I recall correctly. When the JP5A came in
with VOR/DME, it seemed like the space age. This was
about 1974! But the wonderful Gnat with its offset
TACAN was truly magical! We had non-offset TACAN
in the GT6 Hunters at Valley, but all bar 2 of the jets at
Brawdy had the old Rebecca. The Mk9s had ADF, but
no-one taught us how to use it - we just used to listen to
music on it. A chum, Dick 'Whizzbang' was a bit of a culture vulture and had a memorable time rocketing with
SNEBs at Pembrey whilst listening to 'Ride of the Valkyries' on BBC Radio 3! "
The first technician courses on Rebecca 8 were in June 1956
a. Transmitter - receiver TR8193 (10D/19594);
b. Range and heading indicator (10AF/530);
c. Contröl unit
type 8197 (10L/16264);
Ground beacon RT-44 / PPN1 (In UK known as Eureka, 1943 )
Ground beacon RT-37 / PPN2 (In UK known as Eureka, 1946 )
Circuit description
Connectors
Front: The pink coax connector on the front connects the transmitting aerial on the bottom side of
the fuselage of the aircraft. Impedance is 50 ohm I assume.
The yellow connector on the front is probably intended to mute other radio systems during the transmit
pulse.
Rear : On the rear side of the TR8193 is a small coax connector to the antenna switch.
Further a rectangular 28-pin connector for the 28V power supply and the control box.
Power supply
The complete unit is powered by 27.5Vdc.
The heaters of the 29 valves are series-parallel connected, together 2.65A at 27.5 Vdc.
The control panel mode switch first applies 27V to the heaters (7A inrush current), and after a few
seconds the switch can be turned further to one of the operational modes. This starts the dynamotor,
mine required at least 10A initial to start running. Once running, the dynamotor takes 1.6A for
mechanical losses, fan and field , plus 2A for its 300V-150mA load.
So, the total input current of the TR8193 is 6.2A at 27.5V.
Transmitter
Only the transmitter module is placed in a reinforced box, pressurized to 5 psi via a standard tyre
valve. This box has a desiccator cartridge with pressure indicator.
Two cold-cathode thyratrons (CV2349) fire when the voltage on the 0.5uF storage capacitor has
reached 250Vdc. This discharges the capacitor into the HV transformer, providing a 3kV peak pulse to
the balanced oscillator (2x CV1759). The RF output is 1kW during 5 us, 170 times per second.
The average RF output is 1kW x 5us/6ms = 0.8W
Remote tuning is done from the control box with 8 preset variable resistors.
The frequency range is 200 - 235 MHz, probably in 5 MHz steps.
The tuning motor is controlled by two relays ( forward and reverse)
which are controlled by a very sensitive moving coil relay. A
current of only 0.4mA is sufficient to activate one of the two motor
drive relays, yet the coil can withstand 8mA in either direction.
Such relays were made in 1954 by Elliott (model R350B) or
Sangamo Weston ltd (type S.I 15)
Receiver
The receiver is a single superhet with 45 MHz IF, the same as used in the H2S radar.
One grounded-grid RF amplifier, followed by a triode mixer and oscillator. The oscillator is remotely
tuned in the range 169 -190 MHz
The IF amplifier has 100dB gain. Bandwidth is > 3 MHz at -3dB, yet <10MHz at -50dB
Sensitivity is -88 dBm for equal noise and signal.
Remote tuning is done from the control box with 8 preset variable resistors. The same moving coil
relay is used. Now, after 60 years, the moving coil is a bit sticky, with 30% deadband.
What is received
Because there may be up to 75 aircraft interrogating the same beacon, a great many pulses may be
received at this same frequency. Only one of them is synchronous with the transmitted pulses of this
Rebecca unit and should be found in order to measure the distance. The solution is based on a
stroboscopic effect: only the response to its own transmitted pulse arrives each cycle at the same
delay from the transmitted pulse.
Strobe unit
The strobe unit has to find the range and direction to the (eureka) beacon.
The span of the indicator is 20 or 200 nautical miles, which corresponds to 50 + 20 x 12.36 = 297 µs
or 50 + 200 x 12.36 = 2522 µs (time of flight plus beacon delay). This is about one-third of the time
between the transmitted pulses.
To find the reply to the transmitted pulse, in the strobe unit a short pulse is generated at a variable
delay between 10 and 300us resp. 0.1 and 2.5ms after each transmitted pulse. Only received pulses
during this gate time are considered. The delay itself is controlled by a slow ramp, which gradually
increases the delay until at least 5 successive pulses have passed the gate. From then on, the delay no
longer ramps but is controlled to maintain the gate at the expected time of arrival of the received
response pulse.
Strobe gates
Eureka beacons normaly respond with short pulses (5us). However, every few minutes a beacon ID is
transmitted as a two-letter morse code by widening the response pulse to 35us.
All received pulses are applied to the control grid of 4 pentodes; their suppressor grid is normally held
at a negative voltage and the valves are cut-off. Positive "strobe" pulses are applied to this g3 to give
a negative anode pulse if the strobe pulse coincides with a received pulse.
The variable delay generator is followed by two one-shot circuits, to provide four strobe pulses (A,B,C
and D) at or just after the expected time of arrival of the response.
The following pulses are made 10 to 300us resp. 0.1 to 2.5ms after each transmitted pulse :
A TONE-ID (V7)
B BABS dash ( V17)
A
B
C AGC (V14)
D Range (V4)
All pulses are 30Vpp, and 5us wide.
Only pulse C is 8us wide.
D is the "expected time of arrival"
pulse. The wider pulse C is used for
automatic gain control, and
for the course deviation indicator.
Strobe pulse B is used to detect the
longer "dash" pulse in BABS mode,.
C
D
0
12 us
33 us
Pulse A checks if the received pulse is
longer than 25us, so a 1 kHz tone must be made during a dot or dash of the beacon identifier (ID).
The lower two traces are the 10us and 30us one shots. During the positive slope the current in the
anode inductor is cut off, resulting in an overshoot above the +300V HT line. This overshoot is the
strobe pulse, coupled to the suppressor grid of the "gate" pentodes CV2208
Range
The maximum range is either 20 or 200 nautical miles. The choice is made on the control panel, and
selects the maximum delay of the strobe pulse, and the search time :
Range
Max delay
Search time Flyback time
20
0.3 ms
2 sec ?
0.2
200
2.8 ms
20 sec ?
2
The control voltage for the variable delay generator varies between 20 and 200V. This voltage drives
the 1mA, 270 deg rangemeter via a 150 kOhm resistor. There are separate adjustments for the zero
range for the 20nm, 200nm and BABS mode of operation. BABS beacons have a different internal
delay ?
Automatic Gain Control
The AGC is based on the average value of the selected received pulse, not on all other received pulses
that are responses from interrogation by other aircraft.
This automatic action can be switched-off from the older control panels, in which case a manual gain
setting can be set..
ID tone
Every few minutes, an Eureka type beacon transmits its ID as a two-letter morse code by widening
the response pulse from 5 to 12us. This is detected in the Rebecca strobe unit using a second strobe
10us after the main gate. If still a pulse is recieved there for 5 successive cycles, then a collpitts
oscilator is started and a 1kHz tone is produced at the ID-tone output.
Course (L/R) indicator
The Eureka beacon signal is received by two aerials, one on each side of the cockpit. A vibrator type
relay switches between these aerials 15 times per second.
The vibrator relay has auxiliary contacts, connected to the Rebecca unit. In the strobe unit, these
auxiliary contacts connect the AGC signal alternating to the left or right terminal of the Course
indicator in the same rhytm as the antenna switching. If the signal from the left antenna was stronger,
then the course indicator turns to left, and the pilot should make a left turn to fly to the beacon.
BABS (Blind Approach Beacon System )
These beacons are placed at the far end of the runway, and retransmit a pulse on a different frequency
just like the Eureka beacon does. However, the beacon has a special antenna that switches the main
lobe 15 times per second, between just left or just right of the runway. When left, the response pulses
are 5us, when right they are 12.5 us.
The Babs ground antenna was a metal box
around a central antenna with two switched
slots to alternate the main lobe between left
and right of the runway. The unmanned
mobile beacon was placed in a corner
reflector on a van at the end of the runway.
BABS circuit in Rebecca Mk8
In BABS mode, the auxiliary contact from the antenna switch is replaced by a relay contact which
position depends on whether short or long pulses are received. The short/long pulse detection is done
in V15-V16-V17. These valves and associated relays are only fitted in Rebecca Mk8 units with blind
approach support .
Connector
Both the strobe unit and the main chassis of the
T.R.8193 have the same rectangular, 28-pole PYE
connector.
The pinout was probably made such that the remote
control cable could be connected to the strobe unit
without to much modifications.
pin
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
on main unit
gnd
-150 V
Tx pulse
200 nm range
Range only ?
BABS mode
Course meter
Course meter
Range meter
Antenna switch L
Antenna switch R
Antenna switch return
ID tone
OFF flag ?
Rx signal
+ 300V
AGC link to 18
AGC link to 17
Rx tuning
Tx tuning
Off flag ?
Var DC range offset
Fix DC range offset
Var DC range offset
Var DC range offset
+28V dynamotor
+28V heaters and relays
on strobe unit
gnd
-150 V
Tx pulse `TRIG`
Rx signal `SIG `
Range only ?
BABS mode (Relay B)
Course meter
Course meter
Range meter
Antenna switch left position
Antenna switch right position
Antenna switch return
ID tone
Full AGC voltage
200 nm range (Relay A)
+300 V
heaters
AGC ( -1 ... -4V)
heaters
+28 V heaters & relays
Off flag ?
Range offset (20nm)
Fix DC range offset
Range offset (200nm)
+85V
Range offset (BABS mode)
heaters
Front panel -yellow coax plug
BABS
aerial
+300V
IF AMPLIFIER 45 MC
type 8349
V1 .
CV417
15k
47p
27p 12p
47p
100
100
100
560p
4
10k
22p
27p
27p
47k
vibrator
100
100
100
V6 .
CV137
82p
15k
180
V5 .
CV138
100
100
43.25 MC
V4 .
CV138
100
100
214-235 MC
type 8545
V3 .
CV138
V2 .
CV138
V1 .
CV138
V2 .
CV858
33k
2 uF
3
47
47
47
15k
47
47p
10k
180
27p
47k
10k
100
470
non value caps are 1 nF
1
100
18
169-190 MC
5
120
AGC
25uF
tx blanking
17
4k7
18
10nF
6
250mA
Rx
500
M1
3
1k5 50mA
28
7
+28V
33k
20
47
Receiver RF
Transmitter
V4
TRANSMITTER type 8545
10
10nF
1
2
10nF
2 uF
+300V
M2
POWER SUPPLY UNIT
type 8320
Strobe unit
4
V2
V9
V5
V7
V10
1.4 Ω
1.4 Ω
19
V6
V8
V11
+300V /175 mA
- 150V / 5mA
17
V13
V12
V14
120k
0.5uF
200-235 MC
100k
G
V3 .
CV2349
G
1nF
20
Tx
M
19
V17
120 Ω
22 Ω
IF Amplifier
V3
V2
V1
V6
V5
V4
TR 8193 Rebecca Mk 8
+28V
0.5uF
Rx
V4
V16
+28V
27
23 october 2014 kb
28
500
500
2
V3
V15
2x 0.75uF
1
V4 .
CV2349
HEATERS
2
1nF
3
22:1
V2 .
CV1759
100mA
2x 0.1uF
22k
27
-150V
180 Hz
50nF
V1 . 47
CV1759
3
2k5
V2
1.1A
2.65A
4
3kV
V1
47 Ω
-150V
1k5
33k
Tx
V3
7
5
500
200mA
6 2
28V dc / 6A
Page 1: RF and power
STROBE UNIT See page 2
RECEIVER LO and MIXER
500
100mA
8
1nF
10k
A
16
2
+300V
100k
10pF
-21V
2nF
-44V
10k
-150V
L5
-21V
4k7
5
7
0.1u
10k
3M3
AGC
14
T1
17 AGC
390k
1nF
V4 .
CV2209
470k
10k
20k
- 44V
Rel B
RV10
40k
RANGE
+300V
47k
470
4M7
V9 .
CV140
+200
V10 .
CV138
0.1uF
+85V
7
RV9
27k
+100
470k
-150V
28
50k
Rel A
20nF
fly back
50k
RV8
27k
47k
270k
10k 4k7
0.1u
Rel C
5M6
5nF
RV6
25uF
Rel D
track
D
4k7
18
V13b
1mA
V13 .
CV137
47k
0.2uF
20nF
+300V
RANGE
27k
Rel C
Tx pulse
TRIG
3
5us
22k
100p
+300V
0
0.25uF
Rel A
5k
15
680k 100k
Rel A
470
6
12k
56p
220p
470
1M
180k
Transitron
220k
18k
Rel B
Variable delay
-150 V
20
+28V
Relays C, D and E are bi-stable relays with
2 coils, each 1500 ohm, that can be used in series.
L2
L3
3M3
27k
56p
4k7
C
1M
4k7
10p
1M
4k7
220k
150k
220k
RV2
5k
BABS
1 nF
V3 .
CV2209
L4
A
12k
+85V
RV1
200 nm 4
4k7
1M
180p
21
12k
V2 .
CV140
+200
-150 V
RV11
30k
D
B
1M
22p
10p
V6 .
CV858
1M
V5 .
CV858
25
+ 85V
39k
23
RV4
4k7
150k
-150 V
30 us delay
Numbers on the main TR8193 connector (
are the same as on this strobe unit ( )
unless otherwise stated.
10k
30k
27k
220k
10 us delay
)
testpoint
V1 .
CV449
56k
24
100k
RV5
RV7
330k
1mA
+20 V
27k
27p
130k
130k
L1
1M5
9
150k
1 0nF
on front
R
47k
5M6
10M
antenna
switch
5nF
-18V
L
4x 10nF
11
10M
3k9
COURSE
47k
5M6
10
12
5M6
V11
CV137
1M
V12 .
CV137
47k
10M
10k
0.1u
0.1u
OFF flag ?
33M
27k
2M2
2
1k
47k
???
10M
V11b
1
0.1u 8k2
Rel B
47k
6
47k
8
11
2nF
5
orange
Rel E
4k7
T5
68k
10
Rel D
V14 .
CV2209
V15 .
CV137
22k
V12b
10k
820k
2nF
C
330k
1M
-150
6k8
330k
+300V
82k
8k2
3p
V16 .
CV137
Rel E
22p
220k 10k
4
3p
2M2
T2
V17 .
CV2209
820k
470pF
100k
-150V
TONE ID
- 44V
1M
ID TONE
+300V
1M
1nF
1k
47k
SIG
Replies from
beacon, interrogated
by <75 aircraft
yellow
V8 .
CV137
1k5
6M8
20nF
-21V
0.1uF
50nF
1M5
4k7
15
V8b
4k7
100p
100k
see heaters
diagram on
page 1
2nF
10k
1nF
B
T3
V7 .
CV2209
-21V
1
19
27
17
BABS
100k
-150V
13
Code tone output
T4
+300V
100k
68k
56k
26
RV3 100k
Rebecca Mk 8 Strobe circuit
10 Nov 2014 kb
22
8k2
+300V
10k
`