JPH0211717Y2 - - Google Patents

Description

[Detailed explanation of the idea]

INDUSTRIAL APPLICATION FIELD OF THE INVENTION The present invention relates to high efficiency high pressure sodium (HPS) arc discharge lamps. This type of lamp has a non-vitreous arc tube, such as alumina, containing a fill material containing sodium and mercury and a starting gas. In particular, this invention reduces the tube wall temperature to approximately 1100
The present invention relates to improving the efficiency of such lamps by design changes that reduce the tube wall loading and reduce the average arc current while maintaining temperatures above 0.degree. [Prior Art, Problems to be Solved by the Invention] The useful visible radiation resulting from an arc discharge in a mixed sodium and mercury vapor is only one of the numerical modes of energy dissipation by this type of arc. is well known in this field. To optimize the efficiency of high-pressure sodium lamps containing this type of arc,
It is necessary to minimize the energy dissipation of all non-useful modes as a result of the combined effects of variables such as arc temperature, sodium and mercury pressure, power input per unit length, tube diameter and tube wall temperature. As a result of such confirmation, the inventors believe that the current design of HPS lamps, optimized for diameter, tube wall loading, sodium and mercury vapor pressure by empirical techniques well known in the art, is the best in the art. We have found that knowledgeable researchers are subject to a number of unique compromises that they had not previously suspected. For example, it has been found that for a given size tube and given constant power input and sodium pressure, efficiency increases by 6-10% per 100 mm with increasing tube wall temperature. The reason for this increase is that the self-absorption of the sodium D-line radiation in its self-reversing part is caused by the temperature of the tube wall at constant sodium pressure.
This is because the density of colder neutral sodium atoms near the tube wall decreases as T _W increases, as P _Na /kTw
This is because it decreases as T _W increases. Here, P _Na is sodium vapor pressure, and k is Boltzmann's constant. Figure 1 shows that the additional radiation emitted at a tube wall temperature of 1500〓 at constant arc temperature and sodium vapor pressure is shown as a shaded area in the spectral power distribution compared to the case of 1300〓. It is shown. At the same time, the loss of energy per unit area from the arc due to heat conduction to the tube wall decreases as T _W increases. This is because the temperature gradient between the arc and the tube wall is reduced. FIG. 2 illustrates the dependence of efficiency measured as a function of arc tube wall temperature from an experiment in which the tube wall temperature of a lightly loaded arc tube was varied by manipulating it inside an independently controllable furnace. It is something. Therefore, if all other variables are held constant, this factor increases as the pipe wall load (power per unit area of the external wall surface) increases.
This will increase efficiency. This is because the tube wall temperature increases as the tube wall load increases. High tube wall loads are best achieved by operating at high input power per unit arc length in small diameter tubes. This tends to require that HPS lamps of experimentally developed design operate at tube wall loads of about 14 W/cm ² or more, and this in turn requires about 30 W/cm or more per unit arc length. The above inputs and the inner diameter of the pipe, which is usually less than 1 cm, are required. The arc temperature resulting from such operating conditions is typically about 4000° and increases with increasing input power per unit length. As a result of the inventor's research,
The dependence of two of the main useless radiant energy losses of the arc (infrared spectrum and infrared continuum spectrum) on arc temperature is considerably greater than that of the useful visible radiation in the sodium D line. I found out. Therefore, as the arc temperature increases, these two kinds of useless energy losses increase more rapidly than the desired D-line radiation, reducing the ratio of useful visible radiation to non-useful infrared radiation, and concomitantly decreasing the efficiency. decrease. Therefore, at constant tube wall temperature, constant sodium vapor pressure and constant tube diameter, efficiency will decrease with increasing input per unit length and thus tube wall loading. Correspondingly, from this factor, efficiency will increase as power per unit length and arc temperature decrease. Therefore, one can immediately appreciate the inherent compromises in prior art lamps. That is, one factor increases efficiency as the input per unit length and therefore the tube wall load increases, and the other factor decreases the efficiency as the input per unit length and therefore the tube wall load increases. It has not heretofore been possible to take advantage of two separate effects: increased efficiency at reduced power input per unit length and increased efficiency at higher tube wall temperatures. This is because, in the prior art, the lamp power per unit length and the tube wall temperature are permanently coupled. In practice, since the effect of tube wall temperature is slightly larger than the effect of power input per unit length, the net effect in an actual prior art lamp is that the efficiency is
When measurements are made at optimal sodium pressure,
The efficiency was such that it increased slowly with respect to power input per unit length up to the maximum value allowed by the temperature capacity of the arc tube material. The dependence of efficiency on sodium pressure for a fixed tube diameter and input per unit length is well known experimentally in the art, and the spacing between the red and blue peaks of the self-reversing sodium D line is between 80 and 100.
Maximum efficiency occurs at sodium pressures such as angstroms. This results from the competition of two effects. That is, as the sodium pressure decreases toward very low levels, the lumens per unit radiant power of sodium D radiation decreases to a constant 525
approaches m/W. However, the total sodium D-line emission decreases with decreasing sodium pressure and therefore the total efficiency decreases. On the other hand, at sodium pressures above the optimum value, due to the associated broadening of the sodium D line, this radiation increases in the far and near infrared regions, which are insensitive to the eye. Therefore, the average lumen/radiation power of sodium D-line radiation decreases towards 300 m/W. However, the total fraction of input energy radiated in the sodium D line tends to approach the saturation value with increasing sodium pressure, and therefore the total lamp efficiency should decrease with increasing sodium pressure in this region. be. Therefore, the maximum value of lamp efficiency is found at the optimum pressure between the low pressure region and the high pressure region. As a result of research conducted by the inventors of the present invention, it was found that the optimum sodium pressure for maximum efficiency varies depending on the pipe diameter d as shown below. Maximum efficiency is found at a D-line spacing of 80 to 100 angstroms, independent of tube diameter, but the sodium pressure P _oa required to produce this D-line spacing is proportional to P _Na = 1/√d. According to the formula, it decreases as the pipe diameter increases. The inventors further discovered that the various modes of energy loss from the arc vary depending on sodium pressure and tube diameter at constant arc and tube wall temperatures as follows. That is, the sodium D radiation per unit length of the arc is proportional to P ² _Na d ² . The infrared spectrum per unit length of the arc is proportional to P _Na d ^3/2 . The infrared continuous spectrum per unit length of the arc is proportional to P ² _Na d ² . Thermal conduction loss per unit length of the arc is largely independent of P _Na and d. If P _Na , which varies according to 1/√d, is limited to its optimum value, the diameter dependence of the various energy dissipation modes at constant arc and tube wall temperature is: The sodium D radiation per unit length of the arc is
It is proportional to d. The infrared spectrum per unit length of the arc is d
is proportional to. The infrared continuous spectrum per unit length of the arc is proportional to d. Thermal conduction losses per unit length of the arc are approximately independent of d. Therefore, the use of a larger diameter arc tube reduces the percentage of input energy dissipated by heat conduction to the arc tube tube wall, which is approximately one-third of the input power in a prior art 400 watt HPS lamp. It was found that this can be effectively reduced. While the total radiation losses increase with tube diameter, the heat conduction losses remain constant and thus represent a smaller proportion of the total losses. Since thermal conduction losses are the major non-luminous energy losses, luminous efficiency should increase when thermal conduction is reduced. That is, the luminous efficiency increases with increasing tube diameter (provided that the sodium pressure is adjusted to the optimum value at each tube diameter). It is also readily apparent that there are inherent compromises that have been imposed on lamp designers, but which have remained unacknowledged by industry specialists. That is, when the pipe diameter increases, the heat input to the pipe wall required to maintain a constant temperature should increase in proportion to the pipe diameter, but the main component of this heat input, the heat conduction from the arc, , remains constant. Therefore, without special measures to improve the thermal insulation of the tube wall, the tube wall temperature decreases as the tube diameter increases. Because of the strong dependence of efficiency on tube wall temperature, as already mentioned, the decrease in tube wall temperature with increasing tube diameter offsets and reverses the gain that would have been observed at a constant tube wall temperature. Put it away. It is further noted that it is not worth trying to maintain the tube wall temperature constant by increasing the input/unit length at the same time as the tube diameter increases.
This results in a greater increase in the useless infrared spectrum and infrared continuum energy than the visible sodium D line, due to the required increase in arc temperature and the higher temperature coefficient of the former. be. As a result, the effects of power input per unit length and tube diameter on the efficiency discovered by the present inventors have been denied in actual lamps due to the adverse effect of the tube wall, and are different from ordinary lamps. with empirical techniques
It was left undiscovered by many specialists around the world who were tackling the problem of HPS lamp design. The research results of the inventor can be summarized as follows. 1. As the tube wall temperature increases (all other factors held constant), the luminous efficiency increases because the self-absorption of radiation in the center of the sodium D line decreases. Each additional watt of radiation emitted in this spectral region contributes approximately 500 W to the total light output. 2 As the power input per unit length decreases (all other factors held constant), the luminous efficiency increases over prior art lamps. This is because the useless infrared radiation
This is because it decreases at a greater rate than line radiation. Note that this increase in efficiency with decreasing power input per unit length does not continue indefinitely until the power input per unit length disappears. The continued increase in efficiency is limited and ultimately reversed due to the fact that the heat conduction losses themselves have a lower coefficient of dependence on arc temperature than any radiation losses. At some low power per unit length, the energy losses due to thermal conduction become too large relative to the desired D-line radiation, limiting and reversing efficiency gains. Therefore, 20-25W/cm
There is an optimum power per unit length in the vicinity of , which is considerably lower than the operating values of many prior art high pressure sodium lamps. 3. As the tube diameter increases (sodium pressure adjusted to optimum value, all other factors held constant), efficiency increases. because,
This is because useless heat conduction losses are reduced compared to useful radiation losses. FIG. 3 is a diagram plotting the relationship between efficiency and input per unit length using pipe diameter as a parameter (standardized to the efficiency of a conventional 400 watt lamp with an inner diameter of 0.7 cm). For each diameter, the tube wall temperature is assumed to be constant and the sodium pressure to be at an optimal value. In this simplified energy balance diagram,
Changes in the shape of the arc radiation temperature distribution with respect to the tube diameter are ignored. When this factor is included in more detailed calculations, the trend is the same, although the increase in efficiency with respect to tube diameter is not as large as without it. The existence of an increased value of efficiency at the optimum input per unit length can be clearly seen in these calculations. The optimum power per unit length appears to be in the vicinity of 25-25 watts/cm, which is considerably lower than values for many conventional lamps. SUMMARY OF THE INVENTION The concepts and principles described herein differ from the prior art understanding of how to optimize high pressure sodium lamps for maximum efficiency. For example, U.S. Patent No. 3,906,272
Although the figure discloses the optimum arc tube inner diameter and design center arc voltage drop for each wattage lamp, the patent states that this optimum diameter arises from two competing mechanisms discovered by the inventors. don't realize that. The inventors have determined that, with appropriate thermal insulation means to maintain the tube wall temperature sufficiently high, the efficiency increases with increasing diameter by at least twice the diameter disclosed as optimal in the above-mentioned patent. I found that I can continue to do so. The object of the invention is to provide a means for increasing the operating tube wall temperature of an HPS lamp in a range below the maximum allowed by the arc tube material (approximately 1500° for polycrystalline alumina), thereby increasing the tube wall temperature. of
It is an object of the present invention to provide an arc tube that increases efficiency by approximately 6-10% per 100㎓ increase. The operating tube wall temperature is
This can be increased by improved thermal insulation of the arc tube or by reducing primary thermal radiation or conduction of the arc tube material. For sodium-mercury amalgams, means must be provided to maintain the temperature at a value that yields optimum sodium vapor pressure. A specific object of the present invention is to combine a means for increasing the operating tube wall temperature with an arc tube having a considerably larger diameter than prior art arc tubes, and to reduce the tube wall temperature of the arc tube without regard to reducing the tube wall load. The aim is to achieve the efficiency gains associated with the large tube diameter by keeping it at or near the maximum allowed by the tube wall material (approximately 1500 mm for polycrystalline alumina).
A conventional arc tube has a tube outer diameter of about 0.6 to 1.0 cm,
It operated with a tube wall load of approximately 14-20 W/cm ² (when optimally designed). Traditional arc tubes also typically
It operated at approximately 25-50 W/cm per unit length of arc.
In the present invention, the power consumption per unit cm of arc length is generally smaller. To illustrate the changes in lamp design that result from the technology of the present invention, we use a 400W lamp that has been a commercial product since the late 1960s and has remained substantially unchanged with respect to mechanical dimensions, materials of manufacture, or performance standards since approximately 1973. Let's consider the HPS lamp. This kind of lamp is usually 50000m, 125
m/W, and does not exceed the performance standard on average. Arc tubes used by all manufacturers are generally similar in size. A lamp of this type can thus be considered well-optimized according to the teachings of the prior art. EXAMPLE Example 1 below uses transparent polycrystalline yttrium oxide (yttria) instead of alumina as the arc tube material, in accordance with the performance of prior art lamps and the teachings of the present invention. 3 illustrates a comparison with the performance of manufactured lamps. Both translucent ceramics have the property of being opaque in the infrared spectral region. Alumina becomes absorbent between about 4 microns and about 7 microns, and itria becomes absorbent between about 7 microns and about 9 microns. Therefore, ittria emits essentially less heat than alumina at temperatures of about 1200°C. U.S. Patent No. 4147744 and U.S. Patent No. 4115134
The thermal emittance of a translucent polycrystalline Yttoria arc tube has been measured to be about 0.11, while the thermal emittance of polycrystalline alumina is typically 0.020, as disclosed in the US Pat. This allows Ittria arc tubes to reach higher tube wall temperatures at a given power consumption per unit area and, more importantly for the purposes of the present invention, at lower power per unit area than alumina arc tubes. A temperature equal to the temperature can be reached. Thus, in accordance with the present invention, efficiency is increased by an Ittria arc tube that is maintained at or nearly the same temperature as arc tubes designed according to the prior art, has a larger tube diameter, and has a lower tube wall load. lamps with higher values can be provided. [Example 1]

[Table] Here, it is shown that both the current density and the tube wall load of the lamp of the present invention are considerably reduced compared to the conventional lamp, and that the efficiency is considerably increased despite the slight reduction in the tube wall temperature. Please note that there are. It is noted that US Pat. No. 3,906,272 does not disclose an optimal tube diameter for prior art 400 Watt lamps. However, from extrapolation of the curve to the 400 Watt level, following the prior art
It is confirmed that 0.732 cm can be considered very close to the optimum value. The tube wall temperatures listed in the preceding table and in this specification are based on the Comparison by deGroot, J.J.
Between the Calculated and the Measured
Radiance at the Center the D-line in a
High Presuure Sodium Vaper Discharge”
Proc.2nd IEE Conference on Gas Discharge，
It was measured by the radiometric method described in London, P. 124 (1972). This method is
It is believed to have an accuracy of 20-30°. Figure 2 shows a 150W, made in accordance with the present invention.
55V, HPS lamp to conventional technology 150W, 55V,
This shows the results compared to the HPS lamp.
The lamp according to the invention has an ittria arc tube with an internal diameter of 8 mm, whereas the lamp of the prior art has
It had an alumina arc tube with an inner diameter of 5.87 mm. This is very close to the 5.75 mm diameter disclosed in US Pat. No. 3,906,272 as being the optimal value for this type of lamp. [Example 2]

[Table] Here, the lamp according to the present invention has a considerably reduced current density and tube wall load compared to conventional lamps, and has a diameter smaller than that of the US Patent No. 3,906,272.
Note that even though it is 39% greater than what is disclosed as the optimal value in the issue, it has a higher efficiency as well. The efficiency gain for the lamp of Example 2 is greater than that of Example 1 because the tube wall temperature of the new lamp of Example 2 is close to that of the conventional lamp. Example 3 shows a comparison of the efficiency of a 50W lamp according to the present invention employing an Ittria arc tube to reduce thermal radiation loss and two types A and B of 50W lamps of the prior art. Conventional lamp A was manufactured for only about one year and is known to be not optimized according to the prior art, since the tube wall load is very low and the tube wall temperature is very low. Experimental lamps made in accordance with the present invention using identically sized Ittria arc tubes have significantly increased arc tube wall temperatures and a corresponding increase in efficiency. The recently announced prior art lamp B attempts to further optimize the 50W lamp by following well known prior art principles, i.e. by reducing the arc tube tube diameter, shortening the arc length and increasing the tube wall loading. It shows an attempt. [Example 3]

[Table] Arc tube Ittria Alumina Alumina
Arc voltage drop (V)
54. 52. 52.

[Table] The optimum tube diameter for this lamp according to US Pat. No. 3,906,272 is 0.335 cm. The lamp according to the invention is
Note that even with a deviation of more than 40% from this optimum tube diameter, the efficiency is equal. Furthermore, prior art lamp A was intentionally designed with a suboptimal wall load for alumina in order to improve its luminous flux maintenance and ease of manufacture. These advantages are retained in the lamp of the present invention, but are lost in lamp B according to recent technology. Thus far, the specific example used to illustrate the invention has used an Ittoria arc tube. However, the subject matter of the present invention, which has a large pipe diameter, low pipe wall load, and low arc current density, preferably exceeds 1100 °C even though the heat input per unit area to the pipe wall is reduced. Other means for reducing thermal radiation losses can also be used to provide arc tubes with arc tube surface tube wall temperatures in the vicinity of 1200°C. Example 4 below describes the use of an infrared reflective shield to reduce thermal radiation losses. [Example 4] A conventional 400W lamp with an inner diameter of 7.3mm and
An alumina arc tube with an outer diameter of 8.9 mm is
It was constructed by placing it inside the glass envelope of the Model 7720. However, a quartz sleeve with an inner diameter of 29 mm and an outer diameter of 33 mm was placed inside the envelope to surround the arc tube. An infrared reflective coating made of indium oxide and tin oxide was formed on the inner surface of the quartz sleeve. The operating characteristics of this lamp can be summarized as follows. Input 400W Arc tube wall temperature 1257℃ Spacing between D-line peaks 52 angstroms Efficiency 123.8m/W Efficiency calibrated for 10% tube wall reflection loss 136m/W At 400W, the tube wall temperature is the same as the conventional 7.3mm inner diameter. higher than the 1200°C typically associated with the design.
Thus, the use of quartz sleeves allows the use of larger diameters for the tubes. However, the use of such a sleeve would provide an additional glass barrier through which the light emitted by the arc tube must pass. Most of the radiation reflected from the two glass barriers is lost to absorption within the lamp. When this observed efficiency of about 124 m/W is corrected for this loss, the efficiency of the arc tube has been increased considerably over the efficiency of the same arc tube installed without thermal protection, and in fact It can be seen that this is considerably higher than the 125m/W obtained from a conventional 400W lamp.
This increase in efficiency is due to the reduced self-absorption of sodium D radiation caused by the low sodium atomic density near the tube wall as a result of the high tube wall temperature. In Example 5 below, an example will be described in which the principle of a thermal insulator as a radiation reflector is applied to an arc tube having a large diameter. [Example 5] The lamp (lamp C) has an inner diameter of 11.0 mm and an outer diameter of 12.5 mm.
A large diameter alumina arc tube with a diameter of 1.5 mm was constructed within a cylindrical type 7720 glass envelope. An infrared reflective coating similar to that of Example 4 was placed on the inner surface of the jacket. The performance of Lamp C was compared to a similar lamp without an infrared reflective coating (Lamp D), except for the addition of niobium heat at the arc tube end to increase the arc tube end temperature and therefore the pressure of the sodium-mercury alumina gas. The performance was compared with that of the previous model (with a shield). The performance of the lamp can be summarized as follows.

[Table] These results show that the infrared reflective film increases the arc tube temperature. A comparison of the luminous flux in a similar D line shows that there is a benefit from increasing the tube wall temperature. Traditionally designed lamps are
Operates at 125m/W at 400W, 135m/W at 1000W
Operates at m/W. A comparison with Lamp C in 700 shows that the present invention provides higher efficiency than using conventional HPS lamp design methods. Lamp C has higher efficiency at 700W than the 1000W conventional lamp. The present invention is embodied in the prior art
To illustrate how this differs from the HPS lamp design rules, data is presented in Table 1. Table-1
shows the dimensions, average current densities, tube wall loads, and arc loads for a number of high pressure sodium lamps, including all wattage lamps of 70 W and above, currently on the market, designed in accordance with the teachings of the prior art. . Here, current density = I/π (ID/2) ² , tube wall load = P/(π×OD×AL), arc load = P/
It was set as AL. Here, I=lamp current, P=lamp input, AL=distance between electrode tips, ID=inner diameter of arc tube, and OD=outer diameter of arc tube.

【table】

【table】

[Table] An important point to note is the comparison between the 250 and 250S lamps. The latter has been optimized for higher efficiency than the former according to the teachings of the prior art. 250W lamp has a tube wall load of 14.6W/ ^cm2 ,
The 250S lamp has an ID of 0.732cm and provides a luminous flux of about 26500m, while the 250S lamp has a tube wall load of 19.44W/ ^cm2 ,
It has an ID of 0.587cm and provides a luminous flux of approximately 29000m. According to the teachings of US Pat. No. 3,906,272, the optimum tube diameter for this lamp is about 0.55 cm. Therefore, changing the dimensional parameters to obtain increased efficiency in accordance with the teachings of the prior art tends to cause the arc tube diameter to become smaller, resulting in increased tube wall loading. This teaching is diametrically opposed to the technical idea of the present invention. The lamps shown in Table 1 are generally designed for maximum efficiency according to the teachings of the prior art. None of the lamps are designed with a large enough diameter so that the current density is as low as 8.0 A/cm ² . Nor are any lamps designed with tube wall loads as low as 13 W/cm ² . Furthermore, the indicated efficiency generally appears to increase with increasing tube wall temperature, with all tube wall temperatures appearing to be above about 1100°C. Thus, the optimum tube diameter for each lamp cited in U.S. Pat. No. 3,906,272 is
It is concluded that it only represents the maximum possible diameter that could result in a minimum temperature of 1100°C. The lamp according to the invention, in a preferred embodiment, consists of a non-vitreous arc tube 1 with an electrode 2 sealed at the end. The arc tube 1 contains sodium, mercury and a starting gas, typically xenon. The metal frame 3 provides support for the arc tube and an electrical path for the upper electrode. A support wire 4 is embedded in the glass press 5 and constitutes an electrical connection to the lower electrode. The arc tube assembly is housed within a glass envelope 6. Tube 1 was made of Ittria and the results for 150W and 400W lamps made therewith are shown in Examples 1 and 2, respectively. To reiterate, according to the present invention, suitable means are provided to reduce heat loss from the arc tube surface,
Even higher efficiencies can be obtained with larger tube diameters if the tube wall temperature is maintained above 1100° C. even though the thermal energy input per unit surface area of the tube wall surface is reduced.

[Brief explanation of the drawing]

Figure 1 is a plot of sodium resonance radiation as a function of spectral radiation flux versus wavelength at tube wall temperatures of 1300〓 and 1500〓, and Figure 2 shows the relative efficiency as a function of arc tube wall temperature at the optimum sodium vapor pressure. The characteristic diagram shown in Figure 3 is the inner diameter of 2.0, 1.5,
For each arc tube of 1.1 and 0.7 cm, at the optimum sodium vapor pressure and constant tube wall temperature (approximately 1500〓)
FIG. 4 is a characteristic diagram in which the relative efficiency of the HPS lamp is plotted against the input power per unit arc length (cm). FIG. 4 is a perspective view of the HPS lamp according to the present invention. 1... Arc tube, 2... Electrode, 3... Metal frame,
4...Support wire, 5...Glass press section, 6...Envelope or jacket.

Claims (1)

[Claims for Utility Model Registration] 1. (a) A non-vitreous arc tube provided with electrodes at both ends and containing sodium, mercury and a starting gas; (b) The inner diameter of the arc tube satisfies the following inequality, i.e. I/π (ID/2) ² ≦8 Amps/cm ² where I is the design center operating current (Amps) of the lamp, with a design value ID (cm) large enough to satisfy (c) The arc length (distance between electrode tips (cm)) of the arc tube is determined by the following inequality: P/π×OD×AL≦13 Watts/ ^cm2 , where P is the designed central operating power of the lamp (watts) , OD is the outer diameter (cm) of the non-vitreous arc tube equal to the ID value plus twice the wall thickness, and (d) the arc tube of the lamp has a design value AL sufficient to satisfy: The temperature of the wall surface is 1100
a lamp comprising means for reducing heat loss from said non-vitreous arc tube surface, said heat loss reducing means consisting of the material of said arc tube itself or surrounding said arc tube to maintain a temperature above 0.degree. A high-pressure sodium lamp with improved efficiency, characterized in that it consists of thermal insulation means arranged inside the envelope or on the lamp envelope. (2) The sodium lamp according to claim 1, wherein the arc tube is made of yttoria.