Note: Descriptions are shown in the official language in which they were submitted.
256~54
-- 1
The present invention relates to a low pressure
mercury discharge lamp for DC operation and a method of
operating the lamp.
When a low pressure mercury discharge lamp
exemplified by a fluorescent lamp is DC-operated,
electrode loss is reduced as compared to the case of
AC operation and luminous efficacy with respect to lamp
input is increased. Flickering of the lamp is also
decreased.
However, DC operation of such lamps is also liable
to the problem of the cataholesis phenomenon as de-
scribed in S. Yabashi et al, ~DC operation of vertically
placed fluorescent lamp~, preprint of Japan Illuminating
Engineering Soc., No. 18, 1984, p.18. When a lamp is
DC-operated, mercury atoms in the tube are converted
into positive ions so that most of the mercury vapor
filled within the tube is attracted to the cathode side.
As a result, the mercury pressure at the anode side is
greatly decreased, and the discharge tube becomes dark
at the anode side as lamp ON time elapses.
It is an object of the present invention to provide
a low pressure mercury discharge lamp for DC operation
which is restrained from non-uniform brightness due to
the cataholesis phenomenon.
~ 25 In order to achieve the above object, a low
; pressure mercury discharge lamp for DC operation
~ comprises: a discharge tube filled with an ionizable
' ~
. .
L256~5~
medium containing a predetermined amount of mercury;
a cathode, arranged at one end of the discharge tube,
for emitting electrons; an anode, arranged at the
other end of the discharge tube, for collecting the
electrons; and an amalgam, arranged at the anode side
of the discharge tube, for adsorbing mercury in the
discharge tube when the lamp is turned off and for
liberating mercury into the discharge tube when the
lamp is turned on.
In the lamp of the above construction according
to the present invention, when the mercury vapor
pressure at the anode side begins to decrease during
the operating time of the lamp, a corresponding
amount of mercury is liberated from the amalgam.
Since mercury is replenished near the anode, apparently,
mercury is uniformly distributed in the interior of
the tube. For this reason, the cataholesis phenomenon
in which the anode side is darkened according to
;~ ~ the operating time of the lamp is suppressed, and
light of uniform brightness can be emitted from
the overall discharge tube over a long period of
time.
Other objects and advantages will be apparent from
/ :
the following description taken in conjunction with the
accompanying drawings, in which:
Fig. 1 is a partial sectional view of a low
pressure mercury discharge lamp for DC operation
12S6~54
according to a first embodiment of the present
invention;
Fig. 2 is a circuit diagram of a circuit for
operating the discharge lamp shown in Fig. l;
Fig. 3 is a partial sectional view of a low
pressure mercury discharge lamp for DC operation
according to a second embodiment of the present
invention;
Fig. 4 shows a partial sectional view of a low
pressure mercury discharge lamp for DC operation and a
circuit diagram thereof according to a third embodiment
of the present invention;
Fig. 5 is a partial sectional view of a low
pressure mercury discharge lamp for DC operation with
an amalgam arranged at a different position;
Fig. 6 is a partial sectional view of a low pres-
sure mercury discharge lamp for DC operation according
to a fourth embodiment of the present invention;
Fig. 7 is a front view of a display screen in which
a low pressure mercury discharge lamp for DC operation
according to the present invention is arranged;
Fig. 8 is a sectional view of a low pressure
mercury discharge lamp for DC operation according to a
fifth embodiment of the present invention:
Fig. 9 is a sectional view of an amalgam assembly
used in the embodiment shown in Fig. 8;
Fig. 10 is a graph showing light output as
~Z5~1S4
a function of mercury vapor pressure;
Fig. 11 is a graph showing mercury vapor pressure
distribution within a discharge tube;
Fig. 12 is a sectional view of a low pressure
mercury discharge lamp for DC operation according to a
sixth embodiment of the present invention;
Fig. 13 is a graph showing light output from a
central portion of the tube as a function of temperature
of coldest point on the cathode side;
Fig. 14 is a circuit diagram of a circuit for
explaining a first embodiment of an operating method of
a low pressure mercury discharge lamp for DC operation
according to the present invention;
Fig. 15 is a sectional view of a discharge lamp
for explaining a modification of the operating method
according to the first embodiment of the present
invention;
Fig. 16 is a circuit diagram of a circuit for
explaining a second embodiment of an operating method of
a low pressure mercury discharge lamp for DC operation
according to the present invention;
Fig. 17 is a graph showing time T required to
develop the cataholesis phenomenon as 8 function of a
turn off period t2;
Fig. 18 is a graph showing time T required to
develop the cataholesis phenomenon as a function of an
inverse of lamp current IL;
~S6~5~
-- 5
Fig. 19 is a graph showing time T required to
develop the cataholesis phenomenon as a function of
an inverse of a distance ~ between a cathode and an
anode;
Fig. 20 is a graph showing time T required to
develop the cataholesis phenomenon as a function of an
inverse of a cross-sectional area S of the discharge
tube; and
Fig. 21 is a graph showing coefficient A as a
function of mercury vapor pressure Po at the coldest
point near the cathode in the lamp OFF state.
A low pressure mercury discharge lamp for DC
operation according to a first embodiment of the present
invention will be described with reference to Figs. 1
and 2. A phosphor film 12 is coated on the inner
surface of a straight discharge tube 10 made of soft
glass. Mounts 14a and 14b are sealed to the two ends
of the tube 10. Internal lead-in wires 18a and 18b are
sealed in a stem tube 16a of the mount 14a. A coil
filament type anode 20 is arranged between one end of
internal lead-in wire 18a and one end of internal lead-
in wire 18b. Internal lead-in wires 18c and 18d are
sealed in a stem tube 16b of the mount 14b. A coil
filament type preheat cathode 22 is arranged between one
end of the internal lead-in wire 18c and one end of the
internal lead-in wire 18d. The cathode 22 emits
electrons, and the anode 20 collects electrons. After
' ~
~2S6154
the interior of the tube 10 is evacuated through one of
exhaust tubes 24a and 24b extending externally from the
stem tubes 16a and 16b, an inert gas as an ionizable
medium is introduced therein through one of the exhaust
tubes 24a and 24b. The exhaust tubes 24a and 24b are
hermetically sealed. Amalgam 26 for controlling the
mercury vapor pressure at the anode side of the tube 10
during normal operating time is contained in the tube
24a. The amalgam 26 is, e.g., an alloy obtained by
adding mercury to an alloy of bismuth (Bi) and indium
(In). In the lamp ON time, the amalgam 26 liberates
mercury, corresponding to the lack of the mercury vapor
pressure at the anode side, into the tube 10 due to heat
from the anode 20 and, in the lamp OFF time, it adsorbs
free mercury in the tube 10. Instead of placing the
amalgam, an amalgam-forming metal and mercury can be
respectively introduced in the tube 24a and in the tube
10, and an amalgam can be formed within the tube 24a.
In this embodiment, small amounts of auxiliary
amalgàms 28a and 28b are arranged near the anode 20 and
the cathode 22. The amalgams 28a and 28b are obtained
by forming thin films of an amalgam forming metal, e.g.,
indium, on the surfaces of metal bases (e.g., stainless
;~ steel bases), and are respectively supported by the
internal lead-in wires 18b and 18c. When the lamp is
ignited, the amalgams 28a and 28b temporarily liberate
compensational mercury into the tube 10 before
.,
~256~54
substantial mercury is liberated within the tube 10 from
the amalgam 26. In the lamp OFF state, the amalgams 28a
and 28b adsorb free mercury in the tube 10, which is
necessary for liberating into the tube 10 in reignition.
The low pressure mercury discharge lamp of the
above construction is assembled in a circuit 30 having
the two ends connected to an AC power supply 32, as
shown in Fig. 2. One end of the anode 20 is connected
to one end of the AC power supply 32 through a resist-
ance ballast 34 as a current limiting element and a
rectifying bridge circuit 36. One end of the cathode 22
is connected to the other end of the AC power supply 32
through the rectifying bridge circuit 36. A capacitor
38 is connected to the output terminal of the circuit
36. The other end of the anode 20 is connected to the
other end of the cathode 22 through a glow switch
starter 40 and a primary winding 44 of a pulse
transformer 42. A secondary winding 46 of the pulse
transformer 42 is connected to an outside auxiliary
conductor 48 arranged near the wall of the tube 10.
An output voltage from the AC power supply 32 is
converted into a DC voltage by the rectifying bridge
circuit 36. The DC voltage is applied to the anode 20
and the cathode 22 through the capacitor 38 and the
resistance ballast 34. A high voltage pulse is gen-
erated by the pulse transformer 42 in response to an
action of the starter 40 and the high voltage pulse is
.~
-` ~256~54
applied to the wall of the tube 10. As a result, a glow
discharge is generated in the tube 10 to cause a glow
arc transition, and the discharge lamp is DC-operated.
When the lamp operates in this manner, the
auxiliary amalgams 28a and 28b near the electrodes 20
and 22 are influenced by heat from these electrodes 20
and 22 and by ion bombardment due to discharge. The
amalgams 28a and 28b are quickly heated and actively
liberate mercury into the tube 10. Therefore, a
sufficient amount of mercury is supplied into the
tube 10 within a short period of time, and rising
characteristics of light output are excellent.
When the wall temperature of the tube 10 begins to
increase by discharge, the temperature of the amalgam 26
in the exhaust tube 24a is gradually increased. Then,
mercury is liberated into the tube 10 through an opening
of the tube 24a in an amount corresponding to the
temperature increase. The internal mercury vapor
pressure after
the light output reaches normal operation state is
controlled to be a vapor pressure which is determined
by the temperature at the position of the amalgam 26.
When the low pressure mercury discharge lamp is
; DC-operated in this manner, mercury liberated in the
tube 10 is converted into positive ions which are
gradually attracted toward the cathode 22 having a
negative potential. As a result, the amount of mercury
- ~256~54
at the anode 20 side is reduced considerably, and
the mercury vapor pressure at this side begins to
decrease. As noted earlier, the amalgam 26 is contained
in the exhaust tube 24a at the side of the anode 20.
Therefore, when the mercury vapor pressure begins to
decrease, mercury is liberated from the amalgam 26 in
an amount corresponding to the decreased amount, and is
replenished near the anode 20. Apparently, mercury
is uniformly distributed within the tube 10, and a
generation of the cataholesis phenomenon in which the
anode 20 side is darkened is suppressed. Decrease and
non-uniformity of brightness during a long time is
suppressed, and the overall tube 10 uniformly emits
light.
According to an operation test performed by the
present inventors, with a conventional low pressure
mercury discharge lamp, the cataholesis phenomenon
occurred within about 30 minutes after an ignition of
the lamp. However, in a low pressure mercury discharge
lamp having an amalgam 26 near an anode 20 according to
the present invention, the cataholesis phenomenon did
not occur even after 10 hours from the ignition of the
lamp. These results demonstrate that non-uniform
distribution of mercury in the tub~ 10 is prevented by
liberation of mercury from the amalgam 26.
When the discharge lamp operating in the state
`:
~:~ as described above is turned off, a larger amount of
~256~54
-- 10 -
mercury is distributed at the side of the cathode 22.
As the temperatures of the tube 10 and the portion near
the amalgam are decreased, free mercury in the tube 10
begins to be adsorbed by the auxiliary amalgams 28a and
28b. Since the adsorbing capacities of the auxiliary
amalgams 28 and 28b are small, most of the free mercury
in the tube 10 is adsorbed by the amalgam 26 at the
anode 20 side. Thus, the state before lamp operation is
restored, and the same operation is repeated when the
lamp is turned on again.
A low pressure mercury discharge lamp for DC
operation according to a second embodiment of the
present invention will be described with reference to
Fig. 3. A discharge tube 110 bent in a substantially
double-U shape is housed in an envelope 152 with a base
150 for an incandescent lamp. A phosphor 112 is coated
on the inner surface of the tube 110. The envelope 152
houses therein a resistance ballast 34, a rectifying
bridge circuit 36, a glow switch starter 40, a pulse
transformer 42, and a capacitor (not shown) which
together constitute a circuit 30 as shown in Fig. 2.
A small incandescent lamp is used as the resistance
ballast 34. An amalgam 126 obtained by adding mercury
to an alloy of bismuth and indium is housed in an
exhaust tube 124a at the anode side. This discharge
lamp is a compact fluorescent lamp which can replace a
normal incandescent lamp.
5615~
According to this embodiment, as in the case of the
first embodiment, the cataholesis phenomenon upon DC
operation can be prevented. When the compact fluores-
cent lamp is DC-operated, the circuit 3û is rendered
small and the overall device can be rendered light in
weight. Since flickering is eliminated, the discharge
lamp can be conveniently used in place of a normal
incandescent lamp.
A low pressure mercury discharge lamp for DC
operation according to a third embodiment of the present
invention will be described with reference to Fig. 4.
In this embodiment, a cylindrical nickel anode 120 is
mounted between internal lead-in wires 18a and 18b.
When the anode 120 of this shape is used, sputtering of
the anode material by electrons is eliminated, leading
to a long life of the discharge lamp. Since other parts
remain the same as in the first embodiment, the same
reference numerals denote the same parts and a detailed
description thereof will be omitted.
In order to operate the discharge lamp, the inter-
nal lead-in wires 18a and 18b connected to the anode 120
are coupled to one end of an AC power supply 32 through
a rectifying bridge circuit 36, and an internal lead-in
wire 18d connected to a cathode 22 is coupled to the
other end of the AC power supply 32 through the circuit
36 and an inductor 60. A glow switch starter 40 is
inserted between an internal lead-in wire 18c and the
1256154
lead-in wires 18a and 18b.
In the above embodiments, an amalgam 26 is put in
an exhaust tube 24a or 124a. However, a mesh member 126
to which an amalgam is adhered can be winded on a wall
of a stem tube 16a, as shown in Fig. 5.
The present invention is similarly applicable to a
low pressure mercury discharge lamp of separated inner
tube type disclosed in U.S.P. No. 4,199,7û8. In this
lamp, an inner tube which regulates a discharge path is
lû arranged in an outer tube filled with mercury and an
inert gas. The present invention is also applicable
to a DC discharge lamp disclosed in Japanese Patent
Laid-open Application No. 58-204468. In this lamp, an
inner tube which regulates a discharge path to form a
plurality of discharge paths is housed in an outer tube,
a single cathode is arranged in the inner tube, and a
plurality of anodes are arranged in the outer tube but
outside the inner tube. In other words, the structure
of a low pressure mercury discharge lamp of the present
2û invention is not particularly limited, as long as it is
DC-operated.
A low pressure mercury discharge lamp for DC
operation according to a fourth embodiment of the
present invention will be described with reference to
Fig. 6. In this embodiment, an amalgam 26a is put in an
exhaust tube 24a at an anode 20 side, and an amalgam 26b
is also put in an exhaust tube 24b at a cathode 22 side.
':
~256~54
- 13 -
The amalgams 26a and 26b are obtained by adding mercuryto an alloy of bismuth and indium. Since the remaining
parts are the same as in the first embodiment, the same
reference numerals denote the same parts and a detailed
description thereof will be omitted. A circuit for
operating the discharge lamp of this construction can be
similar to that as shown in Fig. 2.
When a low pressure mercury discharge lamp is
DC-operated, mercury emitted in the tube 10 is turned
into positive ions which are gradually attracted toward
the cathode 22 having a negative potential. As a
result, the mercury at the anode 20 side is decreased
in amount, and the mercury vapor pressure at the anode
20 side begins to decrease.
However, with the lamp of the above-described
construction, the amalgam 26a is put inside the exhaust
tube 24a at the anode 20 side. When the mercury vapor
pressure begins to decrease, mercury is liberated from
the amalgam 26a in the amount corresponding to the
pressure decrease, and mercury is replenished near
the anode 20 side. Apparently, mercury is uniformly
distributed inside the tube lO, and the generation of
the cataholesis phenomenon in which the anode 20 side
is darkened is suppressed. Non-uniform brightness is
suppressed for a long period of time, and the overall
tube 10 can uniformly emit light.
In the ON state of the lamp, mercury in the tube 10
12S6~5~
_ 14 -
is attracted to the cathode 22 side. When mercury
continues to be supplied to the anode 20 side, the
amount of mercury at the cathode 22 side increases.
Especially when the lamp is turned on for a long period
of timej an excessive amount of mercury is supplied to
the cathode side, and the mercury vapor pressure at the
cathode 22 side exceeds the optimal pressure.
In this case, in the lamp of the above-mentioned
construction, the amalgam 26b is also arranged in the
exhaust tube 24b at the cathode 22 side. The amalgam
26b adsorbs excess mercury at the cathode 22 side.
Thus, an excessively high pressure of mercury vapor
pressure at the cathode 22 side is suppressed, and a
decrease in brightness at the cathode 22 side is
suppressed.
As a result, the mercury amount is kept optimum at
both the anode 20 and cathode 22 sides, so that the
overall tube 10 can emit uniform light over a long
period of time.
In the lamp OFF time, the mercury vapor pressure at
the cathode 22 side is slightly higher than that at the
anode 20 side. Therefore, the amalgam 26a at the anode
20 side slowly adsorbs mercury, and mercury is shifted
from the cathode 22 side to the anode 2û side. The
supply of mercury at the anode 20 side will not be
deficient.
According to the above-mentioned construction,
::
125~;154
since the amalgams 26a and 26b are arranged at both the
anode 20 and cathode 22 sides of the tube 10, polarity
designation is not necessary when the lamp is operated
in DC voltage in a luminaire, and handling is easy.
A low pressure mercury discharge lamp according to
a fifth e~bodiment of the present invention will be
described with reference to Figs. 7 to 9. The discharge
lamp of this embodiment is a discharge lamp used as a
display element of, e.g., a color video display
apparatus.
Fig. 7 shows a large display screen which is used
in a scoreboard at a ballpark or a sports stadium or on
~; ~ a wall of a building. Several thousands or ten times
that of discharge lamps 212-1 to 212-n are densely
arranged on a rectangular screen board 210. Any picture
can be displayed on the screen board 210 by selectively
switching or light controlling a plurality of discharge
lamps 212-1 to 212-n arranged in the screen board 210.
Fig. 8 illustrates details of the discharge lamp 212-1.
One open end of a straight glass tube 214 is sealed by a
convex Iens 216, and the outer circumferential surface
; of the lens 216 serves as a display surface 216a. The
other open end of the glass tube 214 is sealed by a
::
glass stem 218. Lead-in wires 220a and 220b are sealed
in the glass stem 218. A preheat cathode 222 comprising
a filament coil is connected between the distal ends of
the lead-in wires 220a and 220b which are guided into
~Z5615
-- 16 -
the glass tube 214. A support wire 224 serving also as
a power supply line is sealed in the glass stem 218.
The support wire 224 extends along the axis of the tube
214 and the distal end of the wire 224 is located near
the lens 216. In order not to cause short-circuit
discharge with the cathode 222, the support wire 224 is
inserted into a glass insulating tube 226. An annular
anode 228 is coaxially supported at the distal end of
the wire 224 and is electrically connected thereto. A
light reflecting film 230 of an alumina coating film or
the like is coated on the inner circumferential wall
surface of the glass tube 214. A phosphor film 232 is
formed on the light reflecting film 230. The phosphor
film 232 extends to one end of the glass tube 214 in
which the lens 216 exists, and covers the anode 228.
A base member 234 is mounted at the other end of the
glass tube 214. The lead-in wires 220a and 220b and
the support wire 224 are connected to base pins 236a,
236b and 236c projecting from the base member 234. The
glass tube 214 is filled with predetermined amounts of
mercury and an inert gas.
An amalgam assembly 238 for controlling the
internal mercury vapor pressure during steady lamp ON
time is arranged at the anode side of the insulating
tube 226. The amalgam assembly 238 of this embodiment
consists of a plate-like metal base 240 of stainless
steel or the like, and a thin film of an amalgam
$256~54
forming metal 242 such as indium (In) formed on the
circumferential surface of the base 240, as shown in
Fig. 9. The amalgam assembly 238 bonds with mercury in
the glass tube 214 and forms an amalgam. During the
lamp ON time, the amalgam assembly 238 liberates mercury
into the glass tube 214 by heat from the anode 228.
However, when the temperature within the glass tube 214
is lowered as in lamp OFF time, the amalgam assembly 238
adsorbs free mercury in the tube 214. The discharge
lamps 212-1 to 212-n of this construction are densely
arrayed on the screen board 210 with its display surface
216a facing the display direction. The base pins 236a,
236b and 236c are connected to a controller unit 246
connected to a DC power supply 244.
In order to operate the discharge lamp 212-1, a
; preheating current is supplied to the preheat cathode
222 so as to allow it to emit electrons into the glass
tube 214. The emitted electrons are attracted toward
the anode 228 by a voltage applied to the electrodes
222 and 228, and migrate within the glass tube 214,
resulting in starting of discharge between the two
electrodes. When discharge is started in this manner,
the temperature of the amalgam assembly 238 is gradually
increased, and mercury is liberated into the tube 214 in
an amount corresponding to the temperature increase. A
positive column is formed between the preheat cathode
222 and the anode 228 in the tube 214. The phosphor
'~
:
~256~54
8 -
film 232 is excited by ultraviolet rays emitted from the
positive column and emits visible light. The visible
light is radiated outward through the lens 216. When
the light output reaches the steady state, the subse-
quent mercury vapor pressure is controlled to be a vaporpressure determined by the temperature of the amalgam
assembly 238.
When the discharge lamp 212-1 is DC-operated in
this manner, as described above, mercury liberated in
the glass tube 214 is turned into positive ions which
are attracted toward the preheat cathode 222 having a
negative potential. As a result, the mercury amount at
the anode side 228 is decreased considerably, and the
mercury vapor pressure at the anode side 228 begins to
decrease.
In the above configuration, the amalgam assembly
238 is arranged at the anode 228 side in the glass tube
214. When the mercury vapor pressure begins to decrease
at the anode 228 side as described above, mercury is
liberated from the amalgam assembly 238 in an amount
corresponding to the pressure decrease, and mercury is
replenished near the anode 228. Therefore, apparently,
mercury ions are uniformly distributed within the tube
214, and the generation of the cataholesis phenomenon in
which the anode 228 side is darkened is suppressed. As
a result, the phosphor near the display surface 216a
which is most closely associated with the clearness of
~25615~
-- 19 --
the image is actively excited. Therefore, the overall
display sur~ace 216a can emit bright light over a long
period of time, and a clear image of excellent quality
can be displayed.
When the discharge lamp 212-1 in this state is
turned off, a larger amount of mercury is present at the
preheat cathode 222 side. However, as the temperature
of the glass tube 214 and the amalgam assembly 238
decreases, free mercury in the glass tube 214 is
gradually adsorbed by the amalgam assembly 238. Thus,
the state before starting the operation is restored, and
the same operation is repeated when the lamp is turned
on again.
A low pressure mercury discharge lamp for DC
operation according to a sixth embodiment of the present
invention will be described with reference to Figs. 10
to 13. The discharge lamp of this embodiment is a
discharge lamp used as a display element as in the case
of the fifth embodiment.
In a fluorescent lamp, a maximum output is known
,
to be obtainable if the mercury vapor pressure during
ON time is kept at about 6 x 10-3 Torr. As shown in
Fig. 10, light output decreases if the mercury vapor
pressure deviates from 6 x 10-3 Torr in either direc-
tion. Therefore, the mercury vapor pressure during lamp
ON time is preferably kept at about 6 x 10 3 Torr.
A straight fluorescent lamp for AC operation has
lZ56~5
- 2~ --
a mercury vapor pressure distribution substantially
uniform along the axial direction of the tube, as
indicated by characteristic curve ~ shown in Fig. 11.
Therefore, the mercury vapor pressure can be kept at a
level of 6 x 10-3 Torr along the entire tube. However,
in a straight fluorescent lamp for DC operation subject
to the cataholesis phenomenon, since mercury vapor is
attracted toward the cathode side, the mercury vapor
pressure at the anode side is considerably decreased,
and the mercury vapor pressure at the cathode side is
slightly increased. As a result, the mercury vapor
pressure distribution within the tube is as indicated by
characteristic curve ~ in Fig. 11.
With the mercury vapor pressure as indicated by
characteristic curve ~, the maximum output portion is
shifted toward the cathode side (point P), resulting in
a large decrease in light output at the anode side.
In order to prevent the cataholesis phenomenon, an
amalgam-forming metal can be arranged at the anode side
in the light emitting tube so as to adsorb mercury in
the tube in the lamp OFF time and to emit mercury in the
lamp ON time, as in the first to fifth embodiments.
~; However, the amount of mercury which can be
liberated from an amalgam is limited. When the lamp is
turned on over an extended period of time, no more
mercury can be liberated from the amalgam, and the
cataholesis phenomenon later occurs. The sixth
::~;
~ ~ `f~'
~2~i6~54
embodiment is to resolve this problem.
A discharge lamp 310 shown in Fig. 12 is a dis-
charge lamp as a display element arranged in a screen
board as shown in Fig. 7. The inner surface of a
cylindrical reflecting cylinder 312 has a light
reflecting surface 316 coated with a visible light
reflecting substance 314 such as an alumina deposition
film. A base member 318 is connected to one open end
of the reflecting cylinder 312. A U-shaped fluorescent
lamp 320 as a light source of the display element is
arranged inside the reflecting cylinder 312. A
discharge tube 322 is obtained by bending a straight
soft glass tube having an inner diameter of 10 mm, a
length of 100 mm and a wall thickness of 1.0 mm into a
substantially U shape at the central portion between two
ends 324a and 324b. The tube 322 has a substantially
U-shaped portion 326 between the two ends 324a and 324b.
An anode 328 and a cathode 330 are sealed to the ends
324a and 324b of the tube 322. The tube 322 is filled
with predetermined amounts of mercury and an inert gas.
A phosphor 332 for emitting one color light of red,
green or blue is coated on the inner surface of the
discharge tube 322. The ends 324a and 324b of the lamp
320 are supported by the base member 318. The anode 328
and the cathode 330 are connected to four base pins
334a, 334b and so on (remaining two pins not shown) of
the base member 318. The portion 326 is located inside
'
:
` ` 125~154
- 22 -
the other open end of the reflecting cylinder 312. A
lens 336 for projecting light from the fluorescent lamp
320 is fitted in the other open end of the reflecting
cylinder 312.
An amaigam 338 is put an exhaust tube 334a which
is at the anode 328 side. The amalgam 338 is put in the
exhaust tube 334a in the form of an alloy obtained by
adding mercury (Hg) to at least one of bismuth (Bi),
indium (In) and tin (Sn).
The base member 318 covers the ends 324a and 324b
of the discharge tube 322. At least one air hole (two
holes 340a and 340b in this case) is formed in the
portion around the end 324a of the base member 318. The
air holes 340a and 340b communicate the interior and
exterior of the reflecting cylinder 312. An opening 342
; is formed in the reflecting cylinder 312 at a position
near the linear portion at the anode side. An air flow
along a direction indicated by arrow D or a direction
opposite thereto is generated by the air holes 340a and
340b and the opening 342. Thus, the anode side end 324a
is cooled with air, and the anode side end 324a is
effectively cooled. ~_¦
~;~ In contrast, no air hole or opening is formed inthe cathode side end 324b. Rather, the cathode side
end 324b is covered with the base member 318 and is
insulated thereby with respect to heat. A heat insu-
:;~
l~; lating film can be formed on the cathode side end 324b
:
~:256~54
- 23 -
so as to increase the temperature at this end.
The discharge lamp 310 described above is operated
by an operating circuit as shown in Fig. 2 or 4. In
the ON state of the discharge lamp 310, mercury
vaporized in the tube 322 is turned into positive ions
which are gradually attracted toward the cathode 330
side which is at a negative potential. For this reason,
the mercury vapor pressure at the anode 328 side begins
to decrease. However, since the amalgam 338 is arraged
in the exhaust tube 344a, the amalgam 338 liberates
mercury in an amount corresponding to the decrease
amount, thus replenishing the required amount of
mercury. As a result, the cataholesis phenomen.on in
an early ON period of the lamp is prevented, and the
time required to develop the cataholesis phenomenon
is delayed.
In this embodiment, since the cathode side end 324b
is covered with the base member 318 or a heat insulating
film, the temperature at the coldest point at the
cathode side is increased above that in a conventional
case. Thus, the mercury vapor pressure at the cathode
side end 324b is higher than that in the case wherein
the end 324b is not insulated with respect to heat. It
is difficult for the mercury vapor at the anode side end
324a to move to the cathode side end 324b which is at a
higher pressure. Mercury at the cathode side is return-
ed to the anode side end 324a due to Brown Motion. The
~;25~;154
- 24 -
generation of the cataholesis phenomenon is suppressed
more effectively.
The mercury vapor pressure distribution in the lamp
of this embodiment can be represented by characteristic
curve y in Fig. 11. The point corresponding to the
pressure of 6 x lû 3 Torr at which a maximum luminous
efficacy is obtained is at the central position of the
tube. With this arrangement, a decrease in the luminous
flux at the anode side can be decreased, and a light
output decrease of the overall lamp can be reduced to
a minimum. In the U-shaped discharge tube 322 shown in
Fig. 12, a position of maximum light output corresponds
to the U-shaped bent portion 326. In the case of the
light source 310 as the display element, light from the
lamp 320 is projected toward the front through the lens
336. Therefore, it is most preferable that a position
of maximum light output correspond to the U-shaped bent
portion 326 which faces the lens 336. With this
arrangement, the light output of the overall discharge
lamp 310 is improved.
As can be seen from the characteristic curve r in
Fig. 11, in order to obtain a mercury vapor pressure of
6 x 10 3 Torr at the central portion of the tube, the
mercury vapor pressure at the coldest point at the
cathode side must be doubled, i.e., 2 x 6 x 10 3 Torr.
The present inventors experimentally confirmed the
temperature of the coldest point at the cathode side to
25~5d,
- 25 -
obtain the mercury vapor pressure of 6 x 10-3 Torr at
the central portion of the tube.
The results obtained are shown in Fig. 13. It is
seen from these results that in order to obtain a
maximum light output at the central portion of the
discharge tube, the temperature of the coldest point
must be set at about 6ûC~ However, in order to obtain
a light output of 90 % or more in consideration of lamp
manufacturing allowance and pressure distribution
allowance, the coldest point temperature at the cathode
side may be kept within the range of 50 to 80C.
The application of a fluorescent lamp is not
limited to a light source of the display element 310
but can be extended to a U-shaped lamp for normal
illumination or a double U-shaped fluorescent lamp
obtained by bending the U-shaped lamp into a U shape
having a smaller radius of curvature. This embodiment
is also applicable to a straight fluorescent lamp. The
amalgam may be arranged on a wall of the stem tube or at
the other position at the anode side.
A method of operating a low pressure mercury
discharge lamp for DC operation according to a first
embodiment of the present invention will be described
with reference to Fig. 14.
A phosphor film 412 is formed on the inner surface
of a straight discharge tube 410 of soft glass. An
anode 414 and a cathode 416 are arranged at the two ends
~256~54
of the tube 410. The tube 410 is filled with predeter-
mined amounts of mercury and a starting rare gas. An
amalgam 418 is put in an exhaust tube 428a at the anode
414 side of the tube 410. The amalgam 418 consists of
an alloy obtained by adding mercury (Hg) to at least one
of bismuth (Bi), indium (In), and tin (Sn).
The discharge tube 410 is mounted in a luminaire
(not shown) such that the anode 414 is below the cathode
416. The tube 410 is used in a vertical orientation
with the anode 414 is located below the cathode 416.
One end of the anode 414 is series-connected to one
end of the cathode 416 through a glow switch starter
420. The other end of the anode 414 is connected to the
other end of the cathode 416 through a rectifying bridge
circuit 422. The rectifying bridge circuit 422 is
connected to an AC power supply 426 through a choke
coil ballast 424. An output voltage from the AC power
~; supply 426 is converted into a DC voltage by the circuit
422. The DC voltage is applied to the anode 414 and the
cathode 416. Thus, the tube 410 is DC-operated.
In this ON state, mercury vaporized in the tube
410 is turned into positive ions which are gradually
attracted toward the cathode side 416. Then, the
mercury vapor pressure at the anode 414 side begins to
decrease. However, since the amalgam 418 is arranged
in the exhaust tube 428a at the anode 414 side, it
liberates mercury in the amount corresponding to the
~256~54
- 27 -
decrease amount at the anode 414 side. The cataholesis
phenomenon will not occur during an early ON period of
the lamp.
In addition to the above-mentioned function of
the amalgam 418, mercury in the tube 410 and mercury
liberates from the amalgam 418 is gradually attracted
downward by gravity and tends to collect at the lower
end of the tube 410, since the tube is arranged such
that the anode 414 is located below the cathode 416.
The gravity acting on mercury ions balances with the
attractive force acting on mercury ions toward the
cathode 416 side, thereby balancing the mercury vapor
pressure inside the tube 410.
In this manner, the generation of the cataholesis
phenomenon in an early ON period of the lamp is delayed
; by a combination of the mercury liberating function from
the amalgam 418 and the gravity action of mercury.
In the OFF period of the lamp, extra mercury vapor
in the tube 410 is attracted toward the amalgam 418.
Mercury in the tube collects near the amalgam 418 by
gravity to stimulate adsorption of mercury by the
amalgam 418. Mercury which is not adsorbed by the
amalgam 418 collects near the anode 414 located below.
As a result, when the lamp is turned on again, it takes
a long period of time for mercury to collect at the
cathode 416 side, thereby suppressing the generation of
the cataholesis phenomenon.
.~ ~
:
2S~54
When a DC operation of the lamp of the above
embodiment was performed, the following results were
obtained. With a straight fluorescent lamp having an
inner diameter of 15 mm, a distance between the anode
and cathode of 300 mm and a lamp current of 0.215 A and
using no amalgam, when the lamp was oriented vertically
with the anode 414 located above, the total luminous
; flux after a 10 hour operation was 410 ~m. On the other
hand, when the identical lamp was operated in a vertical
orientation with the anode 414 located below, the total
luminous flux after a 10 hour operation was 820 lm.
When an amalgam 418 was put in an identical lamp
and the anode 414 was located above, the total luminous
flux after a 10 hour operation was 650 Qm. However,
when the amalgam 418 was used and the anode 414 was
located above, the total luminous flux after a 10 hour
operation was 860 ~m.
~; The tube 410 need not be operated in a vertical
orientation. The tube 410 can be operated in an
inclined orientation as long as the anode 414 is located
below the cathode 416. The amalgam 418 may be arranged
on a wall of the stem tube or at the other position at
the anode side. Thus, in the case of the U-shaped
; fluorescent lamp shown in Fig. 15, a similar effect
can be obtained if the anode 414 is located below the
cathode 416.
~:
A method of operating a low pressure mercury
25615
- 29 -
discharge lamp according to a second embodiment of the
present invention will be described with reference to
Figs. 16 to 21.
Fig. 16 shows the construction of a U-shaped
fluorescent lamp and an operating circuit therefor. A
phosphor film 512 is coated on the inner surface of a
discharge tube 510 obtained by bending a soft glass tube
into a U shape. An anode 514 and a cathode 516 are
mounted at the two ends of the tube 510. The tube 510
is filled with predetermined amounts of mercury and a
starting rare gas. An amalgam 518 is put in an exhaust
tube 528a at the anode 514 side of the tube 510. The
amalgam 518 consists of an alloy obtained by adding
i~ mercury (Hg) to one of bismuth (Bi), indium (In) and tin
(Sn).
One end of the anode 514 is series-connected to one
end of the cathode 516 through a glow switch starter
520. The other end of the anode 514 is connected to the
other end of the cathode 516 through a rectifying bridge
circuit 522. The circuit 522 is connected to an AC
power supply 526 through a choke coil ballast 524.
~; Therefore, an output voltage from the AC power supply
526 is converted into a DC voltage by the circuit 522.
The DC voltage is applied between the anode 514 and the
cathode 516 to DC operate the tube 510.
In the ON state, mercury vaporized in the tube
~ 510 is turned into positive ions which are gradually
`:::
~2~ 5~
- 30 -
attracted toward the cathode 516 side. Therefore, themercury vapor pressure at the anode 514 side begins to
decrease. However, since the amalgam 518 is arranged in
the exhaust tube 528a at the anode 514 side, the amalgam
518 liberates mercury in an amount corresponding to the
decrease amount, so that apparently the mercury vapor
pressure distribution within the tube is maintained to
be uniform. As a result, the cataholesis phenomenon in
an early ON period of the lamp is prevented.
Now, the present inventors found that the time
required to develop the cataholesis phenomenon differed
in accordance with turn OFF period t2 even if the same
lamp was used in the various experiments. Based on
this, the present inventors surmised that a better
; 15 effect in suppressing the cataholesis phenomenon could
be obtained if the turn OFF period t2 were controlled
properly.
; Based on this assumption, the present inventors
;~ performed various experiments, and found that if the
following relation is satisfied:
tl/t2 < 780(S/l) (Po/Pg)-(l/IL) ... (1)
where
tl: turn ON period
t2: turn OFF period
S: discharge tube cross-sectional area (cm2)
l: distance between the anode and the cathode
telectrode distance)
~25615~
- 31 -
Po: mercury vapor pressure (Torr) near the cathode
during lamp turn ON period
Pg: filling gas pressure (Torr)
IL: lamp current (A)
the cataholesis phenomenon can be further suppressed.
The experiments performed will be described below.
Experiment 1
The present inventors knew prior to the experiments
that when the turn OFF period t2 is long, the time T
required to develop the cataholesis phenomenon is
different even if an identical lamp is used. In order
to confirm this experimentally, the present inventors
performed the following experiment.
A fluorescent lamp used had a constructian as
~-~ 15 shown in Fig. 16 and an inner diameter of 15 mm (cross-
;~ sectional area S = 1.77cm2?, a distance between the
anode and cathode (I) of 29 cm, a lamp current (IL) of
0.215 A, and a filling gas pressure tPg) of 5 Torr. The
relationship between the turn OFF period t2 and the time
I required to develop the cataholesis phenomenon was
examined. The obtained characteristic data is as shown
~:
in Fig. 17. The characteristics shown in Fig. 17 are
average values obtained with three lamps of identical
~ specifications (the same applies to Figs. 18 to 21). In
- ~ 25 these lamps, the coldest point at the cathode 516 side
was located at the tube end and was 40C.
The characteristics shown in Fig. 17 satisfy:
~ ~ .
~256~5~
- 32 -
T - . O. 28t2 ..(2)
Therefore, the time ~ to develop the cataholesis phenom-
enon upon lamp ON operation is proportional to the turn
OFF period t2 before an ignition of the lamp. That is,
T ~ t2 - (3)
This is attributable to the fact that the longer the
turn OFF period t2, the larger the amount of mercury
adsorbed in the amalgam 518 at the anode 514 side and
the more difficult for mercury to move toward the
lû cathode 516 side.
Experiment 2
In the course of investigation, the present
inventors realized that the larger the lamp current
IL, the shorter the time T required to develop the
cataholesis phenomenon. Ir, view of this, the
relationship between the lamp current IL and the
cataholesis phenomenon was examined in detail.
In this experiment, a discharge tube used had an
inner diameter of 15 mm (cross-sectional area S =
1.77 cm ), and a distance between the anode and cathode
(~) of 29 cm. The coldest point at the cathode side was
forcibly cooled at 40C. The filling gas pressure
was kept at 5 (Torr), and the turn OFF period t2 was
2 hours. Under these conditions, the relationship
between the lamp current IL and the time T was examined.
The obtained results are shown in Fig. 18. From
the results shown in Fig. 18, the relation:
--` 1256~54
-- 33 --
T (l/IL) ...(4)
is satisfied.
The above phenomenon is considered to be caused by
the following. The attraction force of the mercury
toward the cathode 516 side is influenced by the lamp
current IL. That is, when the lamp current is large,
the number of mercury ions is increased, and the amount
of mercury ions attracted to the cathode 516 side is
increased. As a result, the mercury vapor pressure
lû distribution within the tube has a gradient, resulting
in the cataholesis phenomenon.
Experiment 3
Meanwhile, the gradient of the mercury vapor
pressure due to electric potential gradient should
be cancelled in response to turning off the lamp.
Cancellation of the pressure gradient is attributed to
diffusion of mercury atoms. One of the factors influ-
encing the time required for mercury vapor attracted to
the cathode 516 side to be recovered to the anode 514
side by diffusion in the lamp OFF period is presumed to
be the distance ~ between the anode and cathode. In
other words, the longer the distance ~, the longer the
time required to recover the mercury.
A discharge tube used had an inner diameter of
15 mm (cross-sectional area S = 1.77 cm2). Forcible
cooling was performed to keep the coldest point at the
cathode side at 40C. The filling gas pressure (Pg) was
.
1256154
- 34 -
5 Torr, the lamp current (IL) was 0.215 A, and the turn
OFF period t2 was 2 hours. Under these conditions, the
relationship between the distance Q and the time T
required to develop the cataholesis phenomenon was
examined.
The obtained results are shown in Fig. 19. It is
seen from Fig. 19 that, the relation:
T ~ (1/Q) - (5)
is established.
Experiment 4
A factor preventing recovery of the mercury
vapor attracted to the cathode 516 side to the anode
514 side by diffusion during the lamp OFF period is the
cross-sectional area of the tube 510, i.e., the cross-
sectional area of the light emitting space. The smaller
the cross-sectional area, the longer the time required
for recovering the mercury due to diffusion resistance.
A discharge tube used had a distance (Q) of 29 cm.
The coldest point at the cathode 516 side was kept at
40C, the filling gas pressure (Pg) was kept at 5 Torr,
the lamp current ~IL) was set at 0.215 A, and the lamp
turn OFF period (t2) was 2 hours. Under these condi-
tions, the relationship between the cross-sectional
area S of the tube and the time T was examined.
The results obtained are shown in Fig. 20. It is
seen from the results shown in Fig. 20, that the
following relation:
::
~256~S4
-- 35 --
T S (6)
is established.
From experiments 1 to 4, the following relation is
established:
T (S/Q)-(l/IL)-t2 ....................... (7)
It is known that when the filling pressure of a
starting rare gas is high, the amount of mercury
evaporation is regulated and movement of the mercury
vapor is restricted. When the relationship between the
time T and the filling gas pressure Pg was examined, the
following relation was obtained:
T (l/Pg) ...(8)
Substitution of relation (8) in relation (7) yields:
T (S/Q) (l/Pg) (l/IL) t2 ... (9)
Relation (9) can be rewritten as:
T = A (S/Q)-(l/Pg)-(l/IL) t2 ... (10)
where A is a coefficient.
Experiment 5
The mercury vapor pressure in the discharge
tube 510 varies in accordance with the coldest point
temperature. When the mercury vapor pressure at the
cathode side changes, the amount of mercury shifted to
the cathode 516 side also changes, resulting in a change
in the time T.
In view of this, the present inventors examined the
relationship between the cataholesis phenomenon and the
mercury vapor pressure at the cathode 516 side affected
~2S6~5~
by the coldest point temperature at the same side.
A lamp used had an inner diameter of 15 mm (cross-
sectional area S = 1.77 cm2), a distance (~) of 29 cm,
a filling gas pressure (Pg) of 5 Torr, and a lamp
current (IL) of 0.215 A. When these conditions are
substituted in relation (10) and t2 = T/O. 28 obtained
from T -, 0.28t2 of relation (2) is substituted in
relation (10), A '_ 5 is obtained. The tube end defining
the coldest point at the cathode 516 side was at 40C
and the mercury vapor pressure Po at this point was
6.34 x 10 3 Torr.
A similar experiment was performed by variously
changing the temperature of the coldest point. The
mercury vapor pressure Po and the coefficient A corre-
sponding to each different temperature was examined.
The results as shown in Tabie 1 below were obtained:
Table 1
point Mercury vapor Coefficient
tempera- P0 (Torr) A
0 2.00 x 10-4 0.16
1.27 x lû-3 0.99
6.34 x 10-3 5.0
2.6 x lo-2 20
9.10 x 10-2 71
100 2.77 x 10~1 220
- -
~256~54
- 37 -
When the relationship between the mercury vapor
pressure PO and the coefficient A in Table 1 was
represented in a graph, the relationship was confirmed
to be linear as shown in Fig. 21. As a result, we have:
A Po ... (11)
Relation (11) is equivalent to
A = BPo ... (12)
and relation (10) can be rewritten as:
T = B-(S/Q)-(Po/Pg) (l/IL)-t2 ... (13)
Since A = 780Po in the characteristics shown in
Fig. 21, substitution of B = 780 in relation (13)
yields:
T = 780(5/Q)-(Po/Pg)-(l/IL)-t2 ............ (14)
Since T is time required to develop the cataholesis
phenomenon, the lamp can be turned on within the time T
after the turn OFF period t2 in order not to cause the
cataholesis phenomenon. That is, when tl < T, the
cataholesis phenomenon is not caused. Therefore, when
the following relation:
,20 tl < 780(S/Q) (Po/Pg)-(l/IL)~ ~ ............... (15)
i.e.,
tl/t2 < 78û(S/)-(Po/Pg)-(l/IL) _ ...(16)
is established, the cataholesis phenomenon is not
caused.
An experiment was performed under the above
conditions. A general fluorescent lamp is used which
has conditions of a filling gas pressure of 1 Torr to
12.~6:~5~
- 38 -
142Torr and a lamp current of 60 mA to 1,000 mA. The
fluorescent lamp has an inner diameter of 38 mm and a
tube length of 7 to 100 cm. It was confirmed that no
cataholesis phenomenon occurred.
In the lamp used in Experiment 5 and having the
coldest point at 40C, tl/t2 < 0.28. Therefore, if the
turn ûFF period t2 exceeds 18.75 hours, and the lamp is
turned on for a time interval not exceeding 5.25 hours,
no cataholesis phenomenon occurs.
The above embodiments are described with reference
to a U-shaped fluorescent lamp. However, the present
invention is similarly applicable to a straight
fluorescent lamp, an ultraviolet lamp or the other low
pressure mercury discharge lamp. The location of the
amalgam is not limitted to within the exhaust tube. The
amalgam may be arranged on a wall of the stem tube or at
the other portion at the anode side.
