Section II
Paragraphs 
2-25 
to 2-28
drain current a 
function 
of 
gate-source 
voltage. 
With
the 
input 
(gate-source) 
circuit reverse-biased, 
the
FET 
presents 
a 
high impedance  to 
its 
signal sources
(as 
compared 
with the 
low 
impedance 
of the 
forward-
biased 
junction 
transistor  base-emitter 
circuit). 
Be-
cause there 
is no input current, 
FETrs have 
less
noise than 
junction 
transistors. 
Figure 2-4 shows'
the schematic 
symbol 
and biasing for N channel 
and
P 
channel 
field 
effect 
transistors.
Figure 
2-4. Field 
Effect 
Transistor 
Operation
2-25. BASIC 
OPERATION 
OF DEGIIIAT
COUNTER OR 
DECADE DIYIDER
2-26. 
INTRODUCTION. 
Operation of 
the 
decimal
counter circuit and 
the 
decade 
divider circuit 
is 
simi-
Iar.  The difference between 
the two 
is in function.
Decimal 
counter circuits divide the 
input 
signal 
by 
10
and actuate the digital display 
tubes, whereas 
the
decade divider circuits are used 
to divide 
the 
input
2-4
Model 
5245L
signal 
or the output of 
the 
internal 
oscillator 
into 
the
frequencies  to 
be 
counted or 
frequencies 
to 
provide
the 
various 
gate 
times. 
Throughout  the following  dis-
cussion, circuits are referred to 
as 
"counters"though
the description 
applies  equally 
to decade dividers.
Paragraphs  2-25 
through 
2-29 
cover 
general 
opera-
tion of 
the counters 
with 
emphasis 
on counting 
logic;
Paragraphs 
2-30 
through 
2-32 
discuss readout cir-
cuits; and 
Paragraphs 
2-73 
through 
2-86 
discuss
specific 
decimal 
counter 
assemblies 
and 
the 
readout
assembly.
2-27.INPUT 
ANDOUTPUT 
FROM BINARY. 
Figures
2'5Aand 
2-5B 
show 
a 
flip-flop connected 
for 
operation
as a binary circuit 
(basic 
flip-flop operation 
is 
dis-
cussed in Paragraphs 
2-15 
and 
2-16). 
Positive 
input
pulses go 
to the bases 
of both transistors 
and cause
switching by 
cutting off the 
conducting 
transistor.
Negative  reset 
pulses go 
to the base of onetransistor
and 
turn 
it 
on. Note the 
letter 
"A" 
near 
one transis-
tor and 
"4" 
(read 
as 
"A 
bar" 
or 
"not 
A") near 
the
other._ The 
positive-going 
transition 
at the collector
of 
the 
A transistor 
(while 
switching 
from 
A conducting
to A 
conducting) 
provides 
the 
input 
to the next binary
circuit.
2.28. 
CIRCUIT 
ARRANGE 
MENT 
AND 
COUNT
NOTATION.  Figure 
z-SC 
is 
a block 
diagram 
of a
blnicat 
four-binary 
decimal counter. Notice 
that 
the
B 
output 
i_s 
applied to the 
D, D, 
and 
C transistors 
and
that 
the 
D 
output is 
applied  only to 
the C 
transistor.
Each 
input 
pulse 
produces 
a 
different 
combination 
of
conducting and 
cut-off 
stages;there  are only 10 allow-
able combinations  and 
each combination  represents 
a
decimal 
digit, Decimal 
weighting 
is 
the 
decimal value
assigned,  arbitrarily,  to the output of 
a 
pair 
when 
the
plain-Ietter 
transistor 
is conducting.
a. Decimal Count. Decimal 
weighting 
used in 
the
Model 5245L counter  is 
shorvn 
in Figure 
2-5C,
immediately above 
each 
of 
the 
four 
binary stages.
The 
decimal 
weight 
each 
pair 
represents 
is 
present
only 
when 
the 
plain-letter 
side 
(A, 
Br 
I, 
jr 
_9) 
is
conducting; 
when 
the barred-Ietter 
side 
(A, 
B, D, or
C) 
is 
conducting, the 
decimal 
weight 
is 
zero. 
The
decimal count can be 
determined 
by 
adding 
the 
deci-
mal 
weighting 
of the 
four 
stages. 
For 
example, 
if 
the
4, 
B, D, andC 
transistors 
are conductingrwhereA=1,
B 
= 
0r 
D 
= 
4t 
C 
= 
2t 
the 
output 
is 1 + 0 + 4 
+ 2 
= 
1.
b. Binary-coded Decimals. In binary-coded deci-
mal 
notation, 
the 
output is 
either 
1 
(when 
the 
plain-
Ietter 
transistor 
is conducting) 
or 
0 
(when 
the 
barred-
Ietter  transiqtor 
is 
conducting). 
In binary-coded
decimal 
notation, the 
order 
of 
the 
binaries 
is 
given
so that 
binary-coded 
decimals 
can be written with 
the
least 
significant 
digit to the 
right. 
Thus 
in 
the system
used in the 
Model 
5245L, the binary-coded 
decimal
notation 
normally 
is 
given 
in the 
order 
DCBA. 
(Coun-
ter 
binaries 
are 
shown 
intheABDCorderonthe
schematics 
and 
in Figure 
2-5C 
to 
increase clarity 
in
shoring 
signal 
florv.) 
For 
the 
decimal 
count 
of 
7 used
as 
anexatopte 
inParagraph 
arwith 
D 
= 
1r C 
= 
1rE 
= 
0,
A 
= 
1, the 
binary-coded-decimalnumberwould 
be 
1101.
02349-1
ooooo-c-39
A. 
FET Amplifier 
Characteristics
CHARACTERISTIC
COMMON  SOURCE
COMMON 
DRAIN
(Source 
Follower 
)
inpul lmpedonce
Outpul 
lmpedonce
Voltoge 
Goin
Power Goin
lMo-t5Mo
50KO-TOOKO
r0- 200
60dB- 
toodB
lMo-l5Mo
lKo-roKo
<l
40dB-80d8
B. 
FET Biasing
TYPE
N-CHANNEL
DRAIN
GATE
G
o
CONTROL
VOLTAGE
MAXIMUM
CURRENT
FLOW
INCREASING
DECREASES
CURRENT
FLOW
CONTROL
VOLTAGE
I
o
CONTROL
VOLTAGE
MAXIMUM
CURRENT
FLOW
INCREASING
DECREASES
CURRENT
FLOW
CONTROL
VOLTAGE