Note: Descriptions are shown in the official language in which they were submitted.
``` ~3~3~
The present invention relates to a semiconductor
switching device that has a particular application in
telephone networks.
The invention is used in a circuit system having a
supply line to an apparatus and having a remote isolation
device to disconnect the apparatus in response to a remote
command, the improvement in which the remote isolation device
comprises a semiconductor switching device comprising a PNP
transistor having a P-type emitter region, an N-type base
region, and P-type collector region, an NPN transistor having
an N-type emitter region, a P-type base region, and an N-type
collector region and a reverse breakdown PN diode having a
P-type region and an N-type region, wherein the respective
emitter regions of the transistors are connected in series in
the supply line to serve as terminals of the semiconductor
switching device, the collector region of the NPN transistor
is connected to the base region of the PNP transistor and to
the N-type region of the reverse breakdown PN diode, and the
collec~or region of the PNP transistor is connected to the
base region of the NPN transistor and to the P-type region of
the reverse breakdown PN diode.
The semiconductor switching device behaves as a
voltage sensitive switch that appears as a high impedance to
voltages applied at the emitter regions of the transistors,
(the emitter region of the PNP transistor being positive
relative to the emitter region of the NPN transistor) up to
voltag~s slightly in excess of the reverse breakdown voltage
of the PN diode, that changes to a low impedance once the
applied voltage is large enough to cause reverse conduction,
through the PN diode, the transition from the high to the low
impedance state being by way of a negative ~d
rn/
imped~nce sta~e, and reverts to the high impedance
state, by way of the negative impedance state, on the
removal of the applied voltage.
The semiconductor switching devi.ce, i~ capable of
operating as a remote isolation device (RIDl when
ins~alled in the supply lines to a subscriber's
apparatus, in a telephone networ~, since the line
supply voltage may be used to make it switch between
its high and low impedance status and thereby act as a
voltage controlled switch that will disconnect the
subscriber's apparatus in response to a remote command,
the re~ote command being effected by a reduction in the
line supply voltage. The semiconductor switching device
is designed to go to its low impedance state when
subjected to the normal line supply voltage an~,
therefore, it does not interfere with the use of the
subscriber's apparatus. - .
The reverse breakdown PN diode is, preferably,
formed at the collector-base junction of one of ~he
transistor~ and may be formed either at the collector-
ba~e junction of the NPN transistor or the collector-
base junction of the P~P transi~tor.
Thus the NPN transistor may include an additio~al
N-type regisn, that e~tends across the collector-base
junction of the tran~istor ~ of higher impurity concen-
tration than the collector re~ion of the transistor,
which additional N-type }egion provides a part of
the reverse breakdown P~ diode.
-3~
Alternatively, the P~P transi~tor may include an
additional ~-type region, that extends across the
collector-base junction of the transistor, of higher
impurity concentration than the base re~ion of the
S transistor, which additional N-type region provides a
part of the reverse breakdown PN diode.
In either arrangement, the impurity concentration
of the additional N-type region ~ay be of the order of
102 to 1021 atoms per cubic centimetre~
There may be a resistive element connected between
the base and the emitter regions of the NPN transistor.
Preferably, the semicondu~tor switching device has
the form of a monolithic integrated circuit, in order
to minimise the cost per device and the space occupied
by the device when installed in, say, a telephone
networkl
~n integrated circuit se~iconductor swit~hing
device may comprise an N-type semicondu~tor body that
is both the base region of the PNP transistor and the
collector region of the NPN transistor, a firs~ P-type
i~land, at a surface of the N-type body, that is bo~h
the ba3e region of the NPN transistor and the collec~or ~1
reqion of the PNP tran~istor, a se~ond P-type island, ;¦
at the surface of the semiconductor body, that is the
e~itter region of the P~P transistor, an N-type island,
ln the first P-type island, that is the emitter region
of the NP~ transi~tor, and conductive contact region~
providing re~pective ter~in~ls for the se~ond P-~ype
t
~4~ ~L3~,1'3~'~S
island and the N-type island.
An impurity concentration of bet~een 10l4 and 1016
atoms per cubic centimetre is suitable for the N-type
semiconductor body, and, an impurity concentration of
S b~tween 1017 and 1019 atoms per cubic centimetre is
suitable for the first and second P-type islands.
Alternatively, the integrated circuit
semiconductor switching device may comprise a P-type
semiconductor body that is both the base region of the
NPN transistor and the collector region of the PNP
transistor, an N-type island, at a surface of the
P-type body, that is both the base region of the PNP
transistor and the collector region of the NPN
transistor, a P-type island, in the N-type island, that
is the emitter region of the PNP transistor, an N -type
island, at the surface of the P-type body~ that is the
emitter region of the NPN transistor, and conductive
contact regions providing respective terminals for the
P-type island and the N+-type island.
An impurity concentration of bet~een 1014 and 10l6
atoms per cubic centimetre is suitable for the N-type
island, and, an i~purity concnetration of between
1017 and 1019 atoms per cubic centi~etre, is suitable
for the P-type semiconductor body and the P-type
island.
The integrated~circuit semiconductor switehinq
device ba~ed on an N-type body may include a
~3~3~f~
semiconductor re~istor connected between the conductive
contact region with the N type island and the first
P-type island, providing a resistive element connected
between the base and emitter reqions of the NPN
S transistor.
The integrated circuit semiconductor switching
device based on a P-type body may include a semi-
conductor resistor connected between the conductive
contact region with the N -type island and the P-type
body, providing a re~istive element between the base
and emitter reqions of the NPN transistor.
An alternative arrangement to that of providing a
resistive element between the base and emitter regions
of the NPN transistor is that of providing an NPN
transistor with low emitter efficiency, ~his being
achieved by partly short-circuiting the base-emitter
junction of the PNP transistor, either by laying out
the conductive contact re~ion with the N~-type island
to contact the first P-type island, when the integrated
circuit semiconductor switching device is based on an
N-type body, or laying out the conductive contact
region with the N -type island to contact the P-type
body, when the inte~rated circuit semiconductor
switching device is based on the P-type body.
2~ The integrated circuit semiconductor switching
device ba~ed on an N-type ~ody may in~lude a further
N -type island that extends across the junction beeween
the firs~ P-type island and the M type body, and the
:
i'
-6- ~3~ S
first P type island may include a P-type re~ion
immediately adjacent to the further N -type island.
The alternative integrated circuit semiconductor
switchinq device based on a P-type body may include a
S further N -type island that extends across the junction
between the N-type island and the P-type body, and the
P-type body may include a P -type region immediately
adjacent to the further N -type island. In either
arrangement, the further N -type island provides a part
of the reverse breakdown diode and the P -type region
allows adjustment of the breakdown voltage.
An impurity concentration of between 102 and 10
atoms per cubic centimetre i5 suitable for the further
N -type lsland, and, an impurity concentration of
between 1015 and 1017 atoms per cu~ic centimetre is
sui~able for the P -type region.
The semiconductor switching device requires
connection in a specific sense, relative to the supply
voltage in a telephone network, say, in order to
function as a remote isclation device or a voltage
sensitive switch, and for such an application it is
desirable to provide a compound semiconductor switching
device comprising first and second semiconductor
switching devices connected in parallel with each o~her
and in opposite senses one relative to the other, which
compound switching device behaves as a voltage f
sensitiYe switch to an applied voltage of either
polarity. ¦.
~ '
~;
-7- ~3~3z~S
The first and second semiconductor switchin~
devices may be formed in a common semi~onductor body
and be interconnected by means of conductive contact
regions laid on a surface of the se~iconductor body.
The semiconductor body may be either ~-type or P-type
material.
A first and a second form of the semiconductor
switching device, and integrated circuit arrangments of
the first and second forms of the semiconductor
switching device, in accordance with the present
invention, will now be described by way of example only
and with reference to the accompanying drawings, in
which:-
Fig~ 1 is a circuit diagram representing the first
lS form of the semiconductor switching device;
Fig. 2 is a circuit diagram representing the
second orm of the semiconductor switching device;
Fi~. 3 is a circuit diagram representing dual
semiconductor switching devices of the second form:
Fig. 4 is a plan view of an integrated circuit
arrangement of the second form of the semiconductor
switching device:
Fig. 5 is a sectional view through the integrated
circuit arrangement of Pig. 4 taken along the line
X-X;
Fig. 6 is a sectional view through the integrated
circuit arrangement of Fig. 4 taken along the line Y-Y;
:~ Fig, 7 is a part plan view of an inte~rated
circuit arrangement of the first form of the
semiconductor switching device;
~ig. 8 is a sectional view throu~h the integrated
circuit arrangement of Fig. 7 taken along the line Y-Y;
S Fig. 9 is a sectional view through an integrated
circuit arrangement, with a modification, of the second
form of the inteqrated circuit arrangement.
Fig. 10 is a sectional view through an integrated
circuit arrangement comprisins two devices of either
the first or the second form, and;
~ig. 11 represents semiconductor switching devices
of the first form installed in a telep~one system.
Referring to Fig. 1 of the accompanying drawings, the
first form of the semiconductor switching device comprises
a PNP bipolar transistor 1, an NPN bipolar transistor 2,
.
and a reverse bre~kdown diode 3 that is generally known
as a zener diode. The collector electrode of the PNP
transistor 1 ls connected to the base electrode of the
NPN transistor 2, the base electrode of the PNP transis-
tor 1 is connected to the collector electrode of the NPN
transistor 2, and the reverse breakdown diode 3 has its
a~ode and cathode electrodes connected respectively to the
collector electrode and the base electrode of the NPN
transistor 2. The emitter electrodes of the PNP tran-
sistor 1 and the NPN transistor 2 provide respective
terminals 4 and 5 for the semiconductor switching device.
The breakdown diode 3 is effectively connected in
parallel with the collector-base junctions of the PNP
-9~ 3Z~l~
tranSistor 1 and the NPN transistor 2 and has a lower
reverse breakdown voltage than either of those junctions.
The semiconductor switching device is responsive to the
magnltude cf a voltaqe applied to the terminals ~ and 5
and is intended for operation with an applied voltage in
the sense that makes the terminal 4 positive with respect
to the ~erminal 5. On the application to the terminals 4
and 5 of a voltage less than the breakdown voltage of the
diode 3 in the sense indicated, the revese brea~down diode
3 and the collector-base junction of tAe NPN transistor 2
both act to block current flow that would occur through
the emittex electrodes of the PNP transistor 1 and the
NPN transistor 2. On the application to the t~-~inals 4
and S of a voltage, in the same sense, that exceeds the
lS reverse breakdown voltage of the diode 3 (and which need
not exceed the breakdown voltage for the NPN transistor
2), the diode 3 penmits current flow. If the a2plied
voltage is increased gradually, current flowing into the
emikter electrode of the PNP transistor 1 and out of the
emitter electrode of the NPN transistor 2 will first
increase gradually as the applied voltage increases and
there will be corresponding current flows in the
collector electrodes of the transistors 1 and 2. The
transistor cuxrents will continue to increase with
increasing applied voltage up to a current (the ~reakover
current) beyond which the two transistors act together as
a negative impedance, each of the transistors then
~ 1o~ 3~
driving the other into its fully sa.urated state and they
then present, together, a high negative dynamic impedance
for a range of currents slightly in excess of the
breakover current. The negative dynamic impedance region
extends from the breakover current to a current known as
the holding current. For a range of currents in excess of
the holding current, the saturated transistors present a
low positive dynamic impedance to current flow. When the
transistors are in thelr low positive dynamic impedance
state, alteration of the eonditions at the terminals ~
and 5 to cause the current flow through the transistors 1
and 2 to move into the negative impedance region of their
combined current/ voltage relationship will result in a
rapid decrease of the current drawn through the terminals
~ and 5 and the turning off of both the transistors 1 and
2. The applied voltaqe at which the breakover current is
reached is ~nown as the breakover voltage. Reversal of
the applied voltage, that is, the application of an
increasing voltage that makes the terminal S positive
with respect to the terminal 4 results in substantially
no current flow through the transistors 1 and 2 since
their respective base-emitter junctions are reverse
biassed by such an applied voltage. There will eventually
be breakdown of both base-emittex junctions ln response to
the applied voltage and there will be current flbw through
the transistor but there is no negative lmpedance region
in the current/voltage relationship for applied voltages
.
!
,
3~3Z~
that make the terminal S positive with respect to the
terminal ~.
Referring to Fig. 1, the PNP transistor 1 and the
NPN translstor 2 may, in practice, be provided by,
respectively, the P1N1P2 and N1P2N2 g
region (P1N1P2N2) device in which the N1 re~ion acts
both as the base region of the PNP transistor 1 and the
collector region of the NPN txansistor 2, and the P2
region acts both as the collector region of the PNP
transistor 1 and the ~ase re~ion of the NPN transistor 2.
Referrlng to Fig. 1, the breakover voltage for the
semiconductor switching device is determined by the
reverse breakdown voltage of the diode 3 and is, as a
result, set by the choice of the doping levels of the
semiconduc~or regions that make up the reverse breakdown
diode 3. The breakover current fo.r the semiconductor
switching device is determined by the ~mitter efficiencies
of the transistors 1 and 2 and is, as a result, set by the
design of the emitters of the transistors 1 and 2.
Referring to Fig. 2 of the accompanying drawings,~ ~.
the second form of the semiconductor switching device i
includes a PNP transistor 1, an NPN transistor 2, a
reverse breakdown diode 3, and a resistor 6. The PNP
transistor 1, the NPN transistor 2, and the reverse
25 breakdown diode 3 are connected to each other in the same,i
manner as are the transistors and diode in Fig. 1 and the I
emitter electrodes of the transistors 1 and 2, as in
I
-12- ~3~.~3~
Fig. 1, are connected to the respec~ive terminals 4 and 5.
The resistor 6 is connected between the base and the
emitter electrodes of the NPN transistor 2. The resistor
6 acts as a current snunt to the base electrode of the
NPN transistor 2 and, as a result of its shunting action,
the resistor 6 affec~s the breakover current for the
semiconductor switching device. The breakover current of
the semiconductor switching device is thexefore set by the
device designer by means of the resist3r 6.
~he operation of the semlconductor switching device
represented by Fi~. 2, is essentially the same as that of
the semiconductor switching device represented by Fi~. 1,
the effect of the resistor 6 that is included in Fig. 2
and not in Fig. 1 being that the breakover current of
lS the device represented by Fig. 2 is dependent on the
magnitude of the resistox 6.
The current/voltage relationship of the semiconductor
switching device that is represented by Fig. 2, as is that
for the device represented by F.ig. 1, is asymmetrical in
that it includes a range o~ currents over which the
impedance of the device is negative when the applied
voltage is in the sense that makes the terminal 4 positive
with respect to the terminal 5, and includes no negative
résistance portion for situations where the terminal 4 is
negative with respect to the termlnal 5.
A symmetrical current/~olta~e relationship is obtained
by the connection of two semiconductor switching devices
~ 3~3~,f~s,
between the te~inals 4 and 5, where the two d~vice~ are
connect~d in oppo~ite ~ense~ bet~een the termin~ls 4 and S
~nd opera~ion changes from one de~ice to the other ~hen
the sense oE ~he volt~ge applled ~o the terminals 4 and 5
S is ch~nged.
Fi~. 3 of the accompanying draw~ngs represents a
c~mpoun~ seml¢~ductor swl~ching device comprisin~ two of
the de~ioes repre~ented by ~iq. 2 ~Qhne~te~ in opposite
sense~ one ta the o~her b~twe~n ~wo terminals l~ and 19.
One device cQ~prise~ a PNP tran~istor 10, a~ NPW
translstor 11, a reverse b~eakdo~n ~iod~ 12, an~ a
resistor 13, conne~ed toge~her ~s ar~ th~ corresponding
componen~s of the device represe~t~d by Fiy. 2, ~nd wi~h
the emi~er ~le~trod~s of the PNP tran~istor 10 and ~he
NPN tran~istor ll ¢onnected r~pe~ively to th~ terminals
18 ~nd 1~. The othe~ devl~e compri~es a P~P transisto~
14, an NPN t~ansistor lS, a rever~ b~eakdown ~iode 16,
and a resi~t~r 17, agaitl~ ~onne~t~d ~o~ether as ar~ ~h~
correspon~ng c~mponent~ o~ the de~ic~ repre~en~d by
2Q Fig. ~, ~ut wl~h the ~mitter alectrodes of the PNP
tr~n~ or 14 ~nd ~he NPN transis~or 15 connec~ed
respectl~ly to ~h~ termin~I6 19 and 1~
When ~h~ comp~und semicondu~tor swl~chlng d~viee ~h~t
is ~epresented ~y ~ 3 i~ xub~ected ~o ~n increasing
~xt~rn~lly appli~d vol~ge t~at make~ the t~minals 1a
more positive ~han the ~ermln~ the ~ ter-base
junctlon Qf the NPN ~r~nsi~tnx 15 ~lo~ks cu~r~n~ flow.
-14- ~3~2-f~S
The applied voltage is of the sense to forward bi,as the
base-collector junction of the ~PN transistor 15 and the
diode 16 by way of the resistor 17, but current flow
through these routes is blocked by the collector base and
base-emitter diodes of the PNP transistor 14. A continued
increase in the applied voltage without change in its
sense would eventually force the reverse breakdown of one
of the blocking junctions, but before that occurs, the
reverse breakdown diode 12 wi11 permit current flow
through the transistors 10 and 11, and the semiconductor
switching device that includes the reverse breakdown diode
12 will change to its low impedance conductive state. The
reversal oî the sense of the applied voltage will result
in the changeover of the devices as regards which one
remains in its non-conductive state and which one changes
to its conductive state.
In Fig. 3, the first form of the semiconductor
switching device, as represented by Fig. 1, may be
substituted for each os the second fo.rm of the device,
as represented by Fig~ 2.
Referring to Fig. 4 of the accompanying drawings,
the integrated circuit arrangement of the second form of
the semiconductor switching deYice comprises a body 45
of N-type silicon into a surface of which are diffused
P-type regions 41, 4 2, and 4 3 . The P-type regions 41,
42, and 43 are situated alongside each other with the
region 43 occupying a position between the regions 41
and 42. A first N -type region 44 is diffused into the
P-type region 43 and a further N+-type region 47 is
diffused into the same surface of the N-type ~ody 45 as
is diffused the P-type region 43, the further N -type
region 47 being positioned to overlie a part of the
junction between the N-type body 45 and the P-type region
; 43. A further P-type region 48, diffused into the
surface of the N-type body 45, connects to the P-type
region 43 and extends in a part loop into the N-type body
~5. The surface of the N-type body 45 into which the
regions are diffused is covered by a layer 46 of silicon
dioxide except for windows provided in the silicon
dioxide permitting electrical contact to be made with the
P-type regions 41, 42 and 48 and the N+-type region 44.
Electrical contact with the P-type regions 41 and 42 is
provided by a first metallic layer 40 and electrical
..
contact with the N+-type reqion 44 and the P-type region
48 is provided by a second metallic layer 50, both of
which layers overlie the silicon dioxide layer 46 which
separates the metallic layers 40 and 50 from the surface
of the N-type body 45 where there are no windows in the ,¦
silicon dioxide layer 46. ~1
Figs. 5 and 6 of the accompanying drawings, ¦
representing respectively a part view along a transverse '
cross-section through the P~type regions 41, 42, and 43,
: and a part view along a 1Ongitudinal cross-section through
the P-type region 43, assist in clarifyinq Fig~ 4. In
::: : i,
¦ !
.
i
'
..
-16- ~3V3z~
Figs. 4, 5, and 6, the regions 41, 42, 43, and 45 provide
a PNP transistor structure (the regions 41 and 42 are
connected together by the metallic layer 40) that is the
PNP transistor 1 of Fig. 2, the regions 43, 44, and 45
S provide an NPN translstor structure that is the NPN
transistor 2 of Fig. 2, the regions 47 and 43 provide a
reverse breakdown diode that is the diode 3 of Fig. 2, and
the region 48 provides a resistor that is the resistor 6
of Fig. 2. The regions 43 and 45 are 5hared by the PNP
and the NPN transistors eff~ctively making the base
electrode of the PNP transistor the same as the collector
electrode of the NPN transistor, and vice-versa. The
metallic layers 40 and S0 provide the terminals ~ and 5
of Fig. 2.
Referrin~ to Figs. 2, 4, 5, and 6, the range of
reverse breakdown voltages fox the diode 3 may lie in
the range ~ volts to 20 volts, corresponding to impurity
concentrations ~or the P~type region ~3 of between 1017
and 101~ atoms per cubic centimetre. Reverse breakdown
voltages exceeding 20 volts but less than the reverse
breakdown voltage for the junction between the P-type ,l
region 43 and the N-type region 45 are achieved by the 'I
modified a.rrangement represented by Fig. 9 which shows an i~
additional regio~ 51 that is a P-type region positioned in ¦~
the P-type region 43 immediately next to the N+-type
region 47. In the modified arrangement represented by
Flg. 9, the P-type region 43 aCts as a co~mon regioD of
1-
-17- ~3Q3~5
the PNP and NPN translstors while the N -t~pe region 47
and the P -type region 51 act as the reverse breakdown
diode. ~his arrangement avoids any conflicting doping
level requirements for the P -type and P-type regions 51
and 43, when a reverse breakdown diode Wl th a breakdown
vol~age in excess of 20 volts is required (limited, of
course, to the reverse breakdown voltage f~r the junction
between the N-type region 45 and the P-type region 43).
An integrated circuit arrangement of '~.e first form
of the semiconductor switching device may ~e substantially
the same as that shown in Fig. ~ for the second form of
the device, except that the resistive regi_r. 48, shown in
Fig. 4, is omitted. In an integrated circ~ t arrangement
of the first form of the semiconductor swit-hing device,
the characteristics obtained by including the resistive
reqion 4a, shown in Fig. 4, may be provided by modifica-
tion of the structure of one of the active ~evices.
Figs. 7 and 8 of the accompanying drawlngs show one
modification that may be used in an integrated circuit
arrangement of the first form of the semiconductor
switching device to compensate for the absence of the
resistive region 48 that is shown in Fig. 4. Figs. 7 and
8 show that part of the integrated circuit arrangement of
the first form of the semiconductor switchin~ device
25 corresponding to the P-type region 43 of Fig. 4 and its
immediately adjacent regions.
Referring to Fiqs. 7 and 8 of the accompanying
i
I
-l8~ 3~
drawings, an integrated circuit arrangement of the ~irst
form of the semiconductor switching device includes a
window in the silicon dioxide layer 46. The window
extends through one end of the N -type region 44 and over
S the P-type region 43, permitting the metallic layer 50 to
lie in electrical contact with both the N type region 44
and the P-type region 43. This arrangement represents a
conductive bridge between the base and the emitter
regions of the ~PN transistor 2 of Fig. 1, resulting in
a transistor of low emitter injection efficiency and an
increase in the breakover current for the semiconductor
switching device of Fi~. 1. It will be understood,
therefore, t.~at the inclusion of the partly short-
circuited base-emitter junction in the NPN transistor ~
of Fig. 1 removes the need for the resistor 6 included in
the second form of the semiconductor switching device,
represented by Fig. 2.
Referring to Figs. ~1 to 9 of the accompanying
drawings, a doping level in the range lOl4 to lOl6 atoms
per cubic centimetre is suitable for the N-type region 45;
a doping level in the range 1017 to 1019 atoms per cubic
centimetre is suitable for each of the P-type regions 4l,
42, and 43; a doping level in the range 10~ to 1021 -
atoms per cubic centimetre is suitable for each of the
P-type regions 44 and 47; and a dopiny level in the range
1015 to 1017 atoms per cubic centimetre is suitable for
the P-type reqion Sl. A depth in the range 2 to 20
-1 9- 3~3~Z~S
microns is suitable for each of the P-type reqions 41, ~2,
and 43 and a depth of about half that of the P-type region
43 is suitable for each of the N-type regions 44 and 47.
Both limits are included in each of the ranges given
above.
Referring to Fig. 10 of the accompanying drawings,
an integrated circuit arrangement comprising two of the
semiconductor switching devices of either the first or the
second form includes an N-type silicon body 1~5 at one
surface of which are formed a first set of regions 1~1,
142, ;43, and 144 and a second set or similar regions 241,
242, 243, and 244. The regions 141, 142, 143, 241, 242,
and 243 are P-type and t.~e regions 144 and 2~4 are
N -type. The regions 141, 142, and 244 are connected
together by a metallised layer 140 and the regions 144,
241, and 242 are connected together by a metallised layer
150. The metallised layers perform the function of
connection terminals 18 and 19 to the dual device
structure. A silicon dioxide layer 146 is included.
The breakover voltage of the switching device shown
in Fig. 1 is determined by the breakdown characteristics
of the PN junction breakover diode that is connected in
parallel with the base-collector junctions of the
transistors, and it is possible to obtain relia~le
.
switching of the device in response to the supply voltages
normally used for telephone systems. Telephone system
supply voltages are normally of the order of 50 volts or
'
',:
-20- ~3~3Z~
below. The same situation of clean and reliable switching
is obtained for the switching devices shown in Figs. 2
and 3.
The switching device shown in Fig. 2 provides, in
addition to a clean and reliable switching characteristic,
a well defined swi~ching current that depends on the value
of the resistor connected in par~llel with the base-
emitter junction of the NPN transistor.
It will be understood by those skilled in the
electrical art that the device resulting from the
~tructure re~resented by Fi~s. ~ ~nd 6, say, may be
provided by a structure based on a P-type semiconductor
body rather than on an N-type semiconductor body as has
been described. In a structure based on a P-type body,
the P-type body would perform the function of the regions
43 of Figs. 5 and 6, iIl which case the N-type body 45 is
reduced to two N-type islands that accommodate the P-t~pe
islands 41 and 42, respectively, and the N -type region
44 becomes an island in the P-type body. The N+-type
island 47 would be positioned along a part of the
junction between the P-type body and one of the N-type
islands. In a structure based on a P-typ~ body, the
P-type loop 48 may be provided by the presence of an
additional contact between the P-type body and the meta1
contact S0 at a position removed from the N -type island
; by a~out the length of the P-type loop 48. In a
structure based on a P-type body, the arrangement
~.
-21- ~3~3Z~
represented by ~igs. 7 and a is provided by allowing the
metal contact to the N -type island to make contact with
the surface of the P-type body over a part of the
junction between the N -island and the P-type body.
S Similarly, in a structure based on a P-type ~ody, the
arrangement represented by Fig. 9 is pro~ided by means of
a P -type region adjacent to the N+-type island
corresponding to the N+-type island 47 of Fig. 6. A
compound device, corresponding to that represented by
Fig. 10, may also be provided in an arrangement based on a
P-type semiconductor body.
Fig. 11 of the accompanying drawings represents a part
of a tele?hone system including an incoming line 100, an
outgoing line 101, a subscriber's telephone apparatus 102,
and semiconductor switching devices 103 and 1a4 of the
first form as represented by Fig. 1 of the accompanying
.. ...
drawings. A network consisting of a diode 105 in series
with a resistor 106 is connected in parallel with the
telephone apparatus 102. The semiconductor switching
device 103 is connected in series with the incoming line
100 with such polarity as to permit current flow along
the incoming line 100 towards the telephone apparatus 102
and the semiconductor switching device is conected in
series with the outgoing line 101 with such polarity as
to permit current flow along the outgoing line 101 away
from the telephone apparatus 102. The semiconductor
switching devices 103 and 104 are located at the
-22- ~3~f~
telephone apparatus 102 effectively at the respective
ends of the incoming and outgoing lines 100 and 101.
The incoming and outgoing lines 100 and 101 connect the
telephone apparatus 102 to some form of central or
control office~
Referring to Fiy. 11, the inclusion of the semi-
conductor switching devices 103 and 104 in the respective
incoming and outgoing lines 100 and 101 permits the
perfonmance of tests on the telephone apparatus 102 from
the central office by way of the lines 100 and 101, and
permits the performance of tests on the lines 100 and
101, themselves. The testing of parts or the telephone
system is made possible because the semiconduct^r
switching devices 103 and 104 behave as voltage sensitive
switches that are controllable from the central office
by means of the voltages applied to the lines 100 and
101. The reverse breakdown voltage of the reverse break-
down diodes included in the semiconductor switching
devices is set at a value that causes the semlconductor
20 switching devices 103 and 104 to be conductive when the
telephone apparatus 102 is "off-hoo~", in which case the
semiconductor switching devices 1~3 and 104 are electri-
cally "transparent", that is, they have no effect on
normal system operation and are maintained in their
low-impedance conductive states by the current flow
through the lines 100 and 101 under the influence of the
~: normal supply voltage from the central office. ~oweverp
I
-23- ~3~Z~
the semiconductor switching devices 103 and 104 may be
switched off by reducing the central office supply
voltage and will then isolate the telephone apparatus 102
from the lines 100 and 101. With the semiconductor
S switching devices 103 and 104 in the switched off condi-
tion, the parameters of the lines 100 and 101 may be
checked with the knowledge that the telephone apparatus
102 is effectively disconnected form these lines. The
semiconductor switching devices 103 and 134 also permit
checks to be made to the telephone apparat~s 102 which
should appear as a high impedance, when "on-hook", ~o the
normal voltage supply from the central office. Also ~ith
the telephone apparatus 102 "on-hook", the diode 105 and
the resistor 106 permit checks to determine whether or not
there are semiconductor switching devices connected in the
lines 100 and 101 (conduction occurs only when the
voltage applied to the outgoing line 101 is positive
relative to that applied to the incoming line 100 and
exceeds the combined breakdown voltages of the breakdown
diodes in the semicondcutor switching devices 103 and
104) and to identify the incoming line relative to the
outgoing line.
Referring to Fig. 11, it will be noted that the ,~
semiconductor switching device 103 connected in the ¦~
. 25 incoming line 100 is connected in the opposite sense to the
semiconductor switch device 104 connected in the outgoing
line 101. The need for the installer to pay attention to .
-2~- ~3~3Z~5
the polar1ty of each semiconductor switching device when
installiny it may be avoided by providing two semiconduc-
tor switching devices, such as the devices 103 and 104,
iA a single four-terminal package in which the terminals
107 and 108 are identified as a first port, the terminals
108 and 109 are identified as a second port and the
package has clearly marked top and bottom surfaces. It
w1ll be evident that, as re~ards the pac~age with a pair
of semiconductor switching devices, it is immaterial
whether the first port is connected to the lines 100 and
101 of the telephone apparatus 102, and sim1larly for the
second port, provided that the top surface of the package
is always kept "above" the bottom surface. The need for
an installex to pay attentino to either the polarities of
the semiconductor switching devices or the attitude of
the package, when installing the devices, is avoided by
.. . .
providing compound semiconductor switching devices, as
representeA by Fig. 3, or as represented by Fig. 10, for
example, in which case each of the devices 103 and 104 is
replaced by a compound device.
For a telephone system in which the line supply
voltage is 48 volts, suitable brea~over voltages i-or the
semiconduetor switchin~ devices lie in the range 10 to 20
volts, both limits included, that is to say, the range of
breakover voltages extends slightly beyond the range
lying between one quarter and one third of the line supply
voltage.
-2~- ~3~32~S
When semiconductor switchlng devices as herein
described are included in telephone systems for the
purposes explained with reference to Fig. 11, they are
qenerally known as remote isolation devices (RIDS).
S The remote isolation function performed by
semiconductor switching devices as herein described may be
carried cut by electrical networks that include many more
circuit elements than the semiconductor switching devices,
but such electrical networks are costlier to manufacture~,
and are bulkier and less convenient to handle and install
than the sem1conductor switching devices descrlbed.
Semiconductor switching devices, as described herein,
may be employed, at minimal cost for installation, as
remote isolation devices or maintenance termination units
(MTUs) in a variety or ways. The installation of RIDs at
demarcation points in a telephone network will permit the
telephone operating company to divide up the network for
testing purposes by variation of the network supply
voltages. Thus a telephone maintenance centre or repair
service centre may determine the condition of virtually
any selected part of the network (as regards the existence
of resistive faults or open-circuit conditions) without
the need for personnel to leave the maintenance or repair
centre for the purpose of testing the network equipment.
The semiconductor switching devices, when specifically
intended for use as telephone network RIDs, may be
provided in enclosures appropriate to the intended use
-26- ~3~3Z~
ranging from single-line units to plug-in modules for
multi-line applications such as PABX trunk llnes.
In all cases where the semiconductor switching
devices are used as RIDs they will, at low cost, effect
S maintenance cost reduction, avoid revenue lcsses by
facilitating the quick repair and return to service of
high-revenue lines, prevent difficulties associated with
disagreements as to who is responsible for repair when a
line is faulty, permit a 2~-hour test facili-y without the
need to send repair personnel out to investigate customer
complaints of faults and generally provide an environment
leading to the maintenance of high quality transmissions.
i
~.
