AU635780B2 - Gas burner - Google Patents

Description

COMMONWEALTH OF AUSTRALIA 6 58 0 FORM PATENTS ACT 1952 COMPLETE SPECIFICATION FOR OFFICE USE: Class Int.Class Application Number: Lodged: Complete Specification Lodged: Accepted: Published: Priority: Related Art: Name of Applicant: BOWIN DESIGNS PTY LTD.

.Address of Applicant: 37-41 Chard Road, Brookvale, New South Wales 2100, Australia Actual Inventor: John Vincent Joyce Address for Service: SHELSTON WATERS, 55 Clarence Street, Sydney omplete Specification for the Invention entitled: "GAS BURNER" The following statement is a full description of this invention, ''including the best method of performing it known to us:- (Complete of PJ7000 dated 20th October, 1989) 1 S01 792 2 1il 09 GAS BURNER BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to burners and in particular to burners producing low emission levels of oxides of nitrogen.

The invention has been developed primarily for use in flueless convection gas-fired space heaters, and will be described with reference to this particular application. However, it will be appreciated from the discussion herein that the invention is not limited to this particular field of use.

Unvented gas-fired burners are widely used as space heaters in dwellings and other buildings. Their thermal efficiency comes from their ability to reduce air infiltration rates, but they can be a source of indoor pollution especially in the amounts of NO formed *particularly NO 2 NOx is a term used to describe the combined "Oxides of Nitrogen" in particular NO, N 2 0 and NO 2 NO and N 2 0 for example are a concern in the outdoor environment, in particular with relation to acid rain,

S.

ozone and photochemical smog. NO 2 however, is of more concern to medical authorities due to the effect it has on lung function.

Medical research during the 1980's has suggested that much lower levels of NO 2 will affect lung la-

O

function than was previously thought. Until recently in NSW for example, a 3ppm upper limit per 8 hours of NO 2 was considered safe and in the USA the figure was per 8 hours. However, the Public Health Committee of the National Health and Medical Research Council in Canberra after considering all the new available medical data has decided that a level above 0.3ppm gives reason for concern and the World Health Organisation has now set a goal of O.'21ppm (not to be exceeded for more than one hour per month).

Furthermore, in the outdoor environment general concern is increasing over the levels of NOx in both the lower and upper atmosphere and various authorities around the world are introducing legislation to control emissions in combustion products.

Gas burners in general are of two types the Blue Flame Burner and Surface Combustion (Radiant) Burners.

The type most commonly used in convection space heaters is the blue flame burner as they operate at a lower temperature than the surface combustion Burners, making them safer for use in schools or the home. However, it is well established that blue flame burners generally .@6o produce NO 2 in the levels in the order of 15 to ng/Joule and as such are not considered to have potential for the reduction of NOx. For this reason research into producing low NOx burners has centred 2 primarily around surface combustion burners of different forms.

In the last twenty years research into the production of burners having lower NO X emission levels has concentrated on the use of excess air, alone or in combination with the incorporation of second stage burning. As a result, a number of these burners have become very complex in both design and operation procedures.

For example, the most successful to date have centred on using pressurised premixed air/gas mixtures burnt in a variety of metallic surface configurations, at*: ceramic surfaces or after burners. All have relied on high excess air and high combustion load. These requirements of pressurising systems, after burners and high combustion loads result in burners that are often bulky, complicated and inflexible in their operation.

Furthermore, whilst reduction in NOX emission levels have been achieved relative to the older types of 20 burners, it still appears that it has hitherto not been possible to even approach the target levels considered desirable.

Summary of the Invention 9 Accordingly, it is an object of the present invention to provide a low NO X burner of simple construction and flexibility of operation that overcomes or substantially ameliorates the above discussed disadvantages of the prior art.

3 According to one aspect of the invention there is provided a naturally aspirated gas burner apparatus including a plenum chamber having .a combustion surface formed from a conductive heat resistant material of selected area and porosity, the porosity being substantially uniform across and through the combustion surface, a fuel supply, an air/fuel mixing and delivery device extending into said chamber, the delivery device being adapted to supply an air/fuel mixture with an air component at least equal to that required for theoretical complete combustion into said plenum chamber for egress o through said combustion surface and combustion at or adjacent the surface thereof, said porosity providinga., flow rate of air/fuel mixture therethrough that results in a combustion temperature of 600 0 C to 900 0 C at said combustion surface whereby the formation of oxides of nitrogen in the products of combustion is reduced to about 5ng/Joule or below.

Preferably the burner is naturally aspirated.

20 Preferably also the combustion temperature at said combustion surface is in the range of 600-900 0

C.

Desirably the combustion surface is formed from one or more layers of mesh material. In preference the surface comprises three tightly secured layers of 30 x 32 x 0.014" nickel based steel mesh of 32% porosity.

The present invention, it is thought, allows complete combustion to take place resulting in 4 low levels of CO emission ie .002 CO/CO 2 making the burner suitable for unvented indoor use, whilst maintaining temperature levels within a zone which inhibits the formation of NO. Constraining the production of NO which under certain conditions converts to NO 2 is believed to assist in the reduction of all types of oxides of nitrogen to levels previously thought unobtainable.

Ordinary surface combustion burners havy usually been designed to operate at stoichiometric (100%) air/fuel ratio as this generally gives the most o 0 efficient conversion of heat and provides the highest operating temperatures. For these same reasons, this has also been considered the worst condition in which to *0 operate a burner if it was necessary to try and reduce the levels of NOX emission.

Accordingly, it is surprising to note that although the burner hereinafter described makes use of excess air amongst other methods to suppress the combustion temperature levels, experiments have shown that the S burner may be operated at stoichiometric conditions and still produce extremely low levels of NOX. However, the burner in this form is not as compact per MJ/m 2 .hr

S

as when operated with levels of excess air.

It is also interesting to note that low pressure burners using high excess air while not using an air pump of some kind had not previously been considered acceptable, due to problems experienced with flashback.

The results have shown that it is possible to produce a surface combustion burner that has emission levels from the flue products low enough to meet an indoor air quality of 0.lppm.

Brief Description of the Drawings Figure 1 is a schematic exploded view of a first embodiment of a gas burner according to the invention suitable for use in a convection space heater.

Figure 2 is a longitudinal sectional side view of the assembled gas burner shown in Figure 1.

5* 0

O

Figure 3 is a transverse sectional end view of the burner taken on line 3-3 of Figure 2.

,Figure 4 is a transverse sectional end view taken on line 4-4 of Figure 2.

W S.igure 5 is a graph showing the relationship between temperature and nitrogen dioxide emission levels for the first %nd second embodiment of the invention .O operated under a variety of conditions and with various- o modifications.

Figure 6 is a graph showing the relationship between burner loading and nitrogen dioxide emission levels for various configurations of the first embodiment burner.

Figure 7 is a graph showing the effect of using excess air on the emission levels of nitrogen dioxide for various configurations and operating conditions of 6 the first embodiment burner.

Figure 8 is a graph illustrating the relationship between the CO/CO 2 ratio and burner loading for all the configurations tested.

Figure 9 is a graph of temperature against nitrogen dioxide emission levels for various configurations of the first embodiment burner.

Figure 10 is a graph showing the burner loading against nitrogen dioxide emission levels for the first embodiment burner operated in an overloaded condition.

Figure 11 is a graph depicting the averaged general relationship between burner loading and nitrogen dioxide emission levels obtained by pooling the data from the tests conducted.

Figure 12 is a graph showing the averaged general relationship between CO/CO 2 ratio and burner loading.

Figure 13 is a graph showing the averaged general relationship between temperature and nitrogen dioxide.

Figure 14 is a graph showing the averaged general 20 relationship between the percentage air in fuel/air mixture and the emission levels of nitrogen dioxide.

a Description of the Preferred Embodiment Two preferred embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings.

Referring to the drawings, the burner 1 comprises a substantially tubular plenum chamber shown generally at 2, having at one end an air mixing and delivery device 7 shown generally at 3. The plenum chamber 2 is formed from a substantially cylindrical extruded aluminium body 4 having a plurality of longitudinally extending cooling fins 5 extending radially outwards from one longitudinal half of its surface. Two gutters 6 also extend longitudinally on diametrically opposite sides of the tube, each having a deformable lip 7 which is serated on its innermost surface. The portion of the body 4, not having fins 5, is cut away bar two short lengths 8 one *n o* at each end of the tube, which serve as a framework to which the other components are secured.

000 The other half of chamber 2 is .formed from three superimposed layers of heat resistant radiant mesh material 9. The mesh layers 9 are firmly compressed, formed into shape to correspond with body 4 and secured in gutters 6 by crimping lips 7 inwardly. The serations gtip the mesh 9 to provide a high strength connection with body 4. Sealing of this connection is unnecessary as any leakage would be consumed as it passed the flame front.

The air mixing device 3 comprises a gas injector nozzle 10 attached by means bracket 11 to a venturi 12.

At the end of venturi 12 distal to the injector there is provided a substantially semi-circular baffle 13 secured to the wall of the aluminium body 4.

A tapered spreader baffle 14 extends from immediately behind the semi-circular baffle 13 up to the end of the plenum chamber 2. This baffle serves t.' 8 evenly distribute the air/gas mixture along the burner at a substantially constant pressure level so that the mixture burns evenly along the length of the burner.

In use the gas is injected into the mouth of the venturi, drawing and mixing with the ambient air provide a variable air/gas mixture. Combustion of the mixture takes place through the layers of mesh material.

In order to prevent "hot spots" and to keep the combustion temperatures low and even, it is necessary to ensure that the layers of mesh remain tightly secured.

It has been found that warping of the mesh can be minimised bT cutting the mesh on the cross to ensure all mesh filaments are of an approximately even length thereby preventing deformation through differential expansion.

The layers of mesh material are preferably positioned one relative to another such that the openings in each layer do not align and are not in registry with openings in an adjacent layer. In other words, there is no direct path through the openings between the outer combustion surface of mesh layers 9 and plenum chamber In this respect, subsequent layers of mesh act as a barrier to reflected waves of radiant energy (from the surface of the object to be heated) to prevent the reflected energy from entering the plenum chamber and overheating the burner. Importantly, the outer combustion surface of bur. ,r 1 may also be formed of a single layer of mesh, or 9 nraterial, having openings therethrough dimensioned so as to create a labyrinth to prevent reflected infrared energy from being returned to the burner plenum from an adjacent object.

The dimensions and ratings of this first embodiment will now be described.

SPECIFICATION

Burner Energy Rating 19,900 Btu Chamber size Diameter 1.97" (internal) Effective length 18.5" Mesh material 'Inconel' wire a.

S

Ge

S..

S. S

S

05.5 0e*0 S h

S

5*04 ce Effective mesh area Excess Air Baffle angle Baffle position Venturi Throat diameter Intake radius Length from throat to exit Average combustion temperature diameter 0.014 ins Woven mesh 39 x transverse strands per square inch.

(18,5 x 3.27) 60.5 in 2 28% 800 with bracket 1.06" from venturi exit 1.024" 6.142" (40 included angle) 850 0

C

9a EMISSION LEVELS

NO

2 1.8ng/J Ratio CO/CO 2 0.001 0.003 The design is substantially scaleable to produce burners of various energy.ratings.

After commencement of the tests it was ,.ei2ad to construct a second embodiment of the burner having the same energy rating and general specification with the same combustion surface area, only this time with a substantially planar or flat combustion surface to compare its operation with the convex embodiment.

The following experimental results were obtained which serve to illustrate, without limiting the aS invention.

These embodiments of the burner have shown to be capable of producing levels of nitrogen dioxide well

S

below those levels considered to be normal for standard Sburners. The standard blue flame burner currently used in commercial gas space heaters produce levels of nitrogen dioxide in the order of 15-20ng/J, whereas the invented Ilow-NO burner can produce levels as low as

X

lng/J.

The object of the testing was to produce a means of defining the operating conditions of the low-NOx burner to effect a predetermined emission of nitrogen dioxide.

The Australian Gas Association procedures were used to measure appliance emissions in a form relative to the 10 burner output, All NOx levels were measured using Monitor Labs nitrogen oxides analyzer model 8840 and are therefore subject to the accuracy and inherent limitations of such an instrument.

The nitrogen dioxide level can be expressed in units of nanograms per Joule (ng/J) which in turn will relate to room size. This will indirectly control the

NO

2 levels within a room where an unflued appliance is being operated. The levels measured within any given *10 room will therefore vary on the size of that room; the 00e ventilation; the content of the room; the absorption of nitrogen dioxide into walls; and the background level of

NO

2 Accordingly, because of this variability a fairly complex model was required to provide an accurate account for the level of NO 2 within a given room.

To evaluate the levels of emission the burner was mounted to a ;ig beneath a sampling hood. The

PP

background levels of nitrogen dioxide and carbon dioxide P0., were taken and later deducted from the burner sample levels. Below is a summary of the formulae used and b assumptions made in determining the results that follow.

caa* 11 UNITS FORMULAE AND ASSUMPTIONS Nitrogen Dioxide (NO 2 ng/J 195 x (Y2 Yl) x C (X2 Xl) x H Where Y1 concentration of NO 2 in the intake air in ppm (V/V) Y2 concentration of NO 2 in the exit gases in ppm (V/V) C volume of CO 2 produced per unit volume of gas when completely combusted and when both the gas and C02 are measured at MSC.

(Metric Standard Conditions) Xl concentration of CO 2 in the intake air in

(V/V)

X2 concentration of CO 2 in the exit gases in

(V/V)

.4 *4 4 4 0 *444 4O 4.

4e00 4* *o 4 944* *444 4 44..

4@.q c S S 4* 4 4S4

S

544* *44 4.* H gross heating value of the gas in MJ/m 3 at MSC (dry) X of 0 2 in the air/gas fuel mixture Excess Air (Ae) A.F.R. 1 x Stoichiometric air/gas ratio Where A.F.R. Air Fuel Ratio X 100% S20.93 X Stoichiometri, air/gas ratio for natural gas 9.44 (V/V) therefore Ae X (9.44 x 20.93) (9.44 x X) 1 x 100% 4 44 Ae X I- 1 x 100% 197.58 9.44XJ Temperature measurement was achieved by means of a surface probe of Ni-Al type. The probe tip was allowed to rest in contact with the surface of the mesh. The flame height above the mesh of the burner during normal operation is about 1.5-2.0 mm high and the Ni-Al surface probe is of 1/16" diameter (1.587mm) wire. With this criteria, the assumption has been made that the temperatures obtained in experiments are of a mean 12 mesh/flame temperature.

In some instances, the burners were overloaded intentionally. In such cases a flame breaks from the mesh surface and a secondary stage of combustion takes place.

The temperature of this flame was again measured with the surface probe and found to be in the order of 900 0 C. The burner loading was then determined as follows: Burner loading (MJ/m2hr) Determined gas rate x VPi

A

A

a 4,* 10 Where determined gas rate is measured in MJ/hr Pi pressure at the injector (kPa) A surface area of mesh (m As described, the burner mesh is of Inconel material consisting of approximately 60% nickel with a weave specification of 30 x 32 x .014". Three layers of mesh were used in the burner construction, these layers being held in compression to effect a minimal void between the layers.

Goo The low-NO burner was set in a number of operating conditions as described below and samples of the emissions for each condition were taken.

RESULTS

Tests commenced on the 30MJ standard cylindrical burner described having a 2.45mm injector nozzle. The aim of this first test was to determine the effect of 13 temperature with regards to emission levels of the various pollutants. The temperature was varied by allowing the burner loading to rise by increasing the pressure of the gas to the injector. The results are set out below in Table 1 from which it will be seen that the NO

X

emissions increased with increasing temperature but nonetheless were very low throughout the test. The limiting factor appeared to be the minimum loading at which good combustion could still be achieved.

TABLE 1 Burner Temp NO 2 Pressure Loading °C <ng/J) kPa MJ/m 2 hr CO 2

CO/CO

2 650 1.99 0.2 260.3 0.9 .03 700 2.133 0.3 318.8 1.2 .0137 750 2.63 0.45 390.4 1.32 .0056 800 2.68 0.68 479.9 1.75 .0020 850 2.434 1.00 582.0 2.06 .0010 O* So OS S *r 0 0** 0:00%2 a l20 6 6 Determined gas rate at 1 kPa 28.72 MJ Ambient NO 0.105 p.p.m Ambient CO 2 0.055% Injector Size 2.45mm The test was then repeated on the same burner but using smaller increments of increased pressure in order to refine the data. The results are shown below.

14 TABLE 2 Measured Temp NO 2 Pressure OC (ng/J) kPa MJ/m 2 hr NO AE CO 2

CO/CO

2 700 2.144 0.45 390.4 0 10% 1.04 0.01 710 2.19u 0.50 411.5 0 10% 1.04 0.01 730 2.104 0.51 415.6 0 10% 1.11 0.006 760 2.107 0.67 476.4 0 17% 1.26 0.004 780 2.56 0.72 493.8 0 17% 1.28 0.003 790 2.626 0.75 504.0 0 25% 1.33 0.003 800 2.647 0.82 527.0 0 25% 1.37 0.0025 820 2.475 0.90 552.1 0 25% 1.41 0.002 825 2.536 0.95 567.2 0 25% 1.45 0.0018 835 2.537 1.00 582 0 25% 1.46 0.0017 840 2.560 1.10 610.4 0 35% 1.52 0.0015 eq .C0.

0 s o

C..

C..

S a..

C

C

Determined gas rate at 1 kPa Ambient NqO 0.080 p.p.m Ambient CO 0.02% 2 Injector size 2.45mm 28.72 MJ C C S. S See.

0 Still using the same basic burner, the injector was replaced with a larger nozzle of 3.00mm and again the pressure of the gas was varied to determine the effect on temperature and thereby monitor variations in pollutant emission levels. It can be seen that the burner output at lkPa gas rate was substantially higher at almost 48MJ. This resulted in overall increased temperatures and NO X emission although viewed with respect to existing burners the emission levels were still surprisingly low.

15 TABLE 3 Measured Measured TEMP NO 2 Pressure NO °C (ng/J) kPA MJ/m 2 h CO 2

CO/CO

2 850 4.547 0,40 1.1 613 1.16 .001 860 4.533 0.44 1.3 643 1.24 .001 870 4.516 0.50 1.25 685 1.26 .0007 880 4.607 0.52 1.4 699 1.33 .0007 890 4.780 0.58 1.55 738 1.39 .0006 900 4.602 0.68 1.65 799 1.50 .0006 910 4.636 0.74 1.8 833 1.57 .0005 920 4.683 0.75 2.0 839 1.60 .0005 930 4.820 0.78 2.0 856 1.60 .0005 OS *O

S

0 es .me.

S

S

20 *0055

S

0 0* Determined gas rate at 1 kPa 47.83 MJ Ambient NO 0.090 p.p.m Ambient CO 0.04% 2 Injector Size .00 mm Injector Size 3.00 mm The burner injector was then changed back to the standard 2.45mm nozzle. Tests were repeated varying the pressure in increments but this time the air mixture was adjusted at each stage such that the mixture remained at stoichiometric throughout the test whilst the temperatures varied. It is clear from the result below that the temperature overall was higher due to the lack of cooling effect from the inherent excess air but that overall again the emission levels were surprisingly low.

16 TABLE 4 TEMP NO 2 Pressure 0C(ng/J) kPa Mj/m 2 hr CO 2

CO/CO

2 720 2.747 0.44 386 1.01 .0097 740 3.077 0.5 411.5 1.06 .0074 760 3.474 0.55 431.6 1.11 .0057 780 3.432 0.6 450.8 1.17 .0045 795 3.45 0.65 469.2 1.21 .0037 820 3.235 0.75 504 1.30 .0025 835 4.353 0.8 520.5 1.38 .0020 850 4.374 0.85 536.5 1.14 .0018 860 4.694 0.9 552.1 1.44 .0017 875 4.803 1.0 582 1.53 .0015 880 4.827 1.1 610.4 1.60 .0012 Se 00 00 0 0000 0 6e00 0S

S

S..

CmOS

S

@000 0e 0 5044 Determined gas rate at 1 Ambient NO 2= .08 p.p.rn 2 Injector Size 2.45mm kPa =28.72 MJ 00 too* be.

S

Accordingly it was decided that the next test should determine the effect of the percentage air comporiz~nt whilst maintaining the gas pressure at a constant level. The test ;w.s conducted ntesadr burner with the 2.45mm. injector nozzle. The results are shown below.

17 TABLE 0* eo

A.

4OE# Measured Excess N02 Aeration (ng/J) NO CO 2

CO/CO

2 -16% 6.285 1.2 1.7 .0008 17% 3.46 0.1 1.56 .0016 2.249 0 1.49 .0017 Ambient NO 2 0.08ppm CO 2 0.02% Determined gas rate at 1 kPa 28.72 MJ Injector size 2.45mm The above test was then repeated this time keeping the temperature constant at 820°C and again varying the percentage air supply. The results are as shown in Table 6 below.

TABLE 6 Measured Excess N02 Aeration (ng/J) NO CO 2

CO/CO

2 7.07 1.0 1.61 0.0009 6.013 0.3 1.51 0.0014 17% 3.14 0 1.38 .0022 2.85 0 1.41 .0017 2.501 0 1.47 .0018 Ambient NO 2 0.08 Cog .02% Determined gas rate at 1 kPa 28.72MJ Injector size 2.45mm It was then decided to reduce the burner output by using a smaller 2.1mm jet such that at 1 kPa gas pressure the output was around 23 MJ, and the above aeration tests were repeated. The effects are illustrated in Table 7 below.

*eve 0049 tor

*A

00 0606e: 3 0 18 TABLE 7 Excess Aeration Measured N0 2 (ng/J)

CO

2

CO/CO

2 met' Se -38% 9.766 0 1.18 .0041 STOICHIOMETRIC 5.134 0 1.17 .0038 2.766 0 1.16 .005 37% 2.215 0 1.14 .0048 Ambient NO .44pprn CO 0.03% 2 1 2 Determined gas rate at 1 kPa 22.99MJ Injector size- 2.1mm The last test was repeated again with an even smaller 1.85mm nozzle such that the burner output at 1 kPa gas pressure was around 18 MJ. The results are shown below.

TABLE 8 6.6S 0 a, St .6 S. 4 20 61~ 6.

S.

4 640655

I

*Excess Aeration Measured

NO

2 (ng/J) C0 2

CO/CO

2 -37% 12% 6% 47.5% 6 .702 5 .129 2. 792 1.966 1.966 0.91 0.92 0 .92 0.86 0.86 .0116 .0125 .0134 .0177 0183 Ambient NO 2= 0.44ppm Determined gas rate at Injector size 1.85mm CO 0.03% 2 1 kPa 17.63MJ 19 9 As it appeared clear at this stage that the mesh was playing a significant role in reducing the combustion temperature it was decided to try altering the thickness or number of layers of mesh. Previous tests with only two layers of the mesh available were unsuccessful due to the "blow back" of the flame front that was experienced.

However, it was thought that use of a different mesh gauge and/or weave would overcome this problem although time constraints precluded such further tests at this stage.

F* Accordingly the next step^conducted used four layers

S

of the previously used mesh. The first test was on the o* standard burner using a 3mm nozzle and the pressure was e* raised in the same way as discussed in relation to Table 3. The results are shown below.

TABLE 9 a a o

S

TEMP NO 2 Pressure NO °C (ng/J) kpa (PPM) CO 2

CO/CO

2 780 5.433 0.3 0 1.46 .0011 805 5.266 0.4 1.8 1.63 .0008 830 5.168 0.5 2.15 1.78 .0007 850 4.935 0.6 2.5 1.96 .0006 870 4.524 0.7 2.7 2.10 .0005 Determined gas rate at 1 kpa 41.62 MJ Ambient NO 0.44ppm Ambient CO 0.03% 2 Injector size 20 0O* to.* :00.

0 00* 0 09 0:009 The nozzle was then changod back to the 2.45mm standard injection and the above test repeated. The results are shown in Table 10 below.

TABLE TEMP NO 2 Pressure NO ZC (ng/J) kpa (PPM) CO 2

CO/CO

2 710 4.230 0.32 0 0.92 0.128 750 4.737 0.45 0 1.05 ".0065 770 4.,526 0.52 0.05 1.18 .0038 790 4.249 0.66 0.1 1.28 .0024 810. 3.945 0.8 0.15 1.39 .0017 830 3.625 1.0 0.2 1.51 .0013 860 3.29 1.1 0.4 1.58 .0010 Determined gas rate at 1 kpa 28.76 MJ Ambient NO 2 0.44ppm Ambient CO 2 0.03% Injector size 2.45mm The test was repeated onice more using the larger nozzle and the results are recorded below.

.,TABLE TEMP NO 2 (ng/J) Pressure 0Ckpa

CO

2

CO/CO

2 740 5.145 0.58 1.85 .0005 780 5.49 0.5 2.15 .0004 800 5.423 0.4 2.27 .0005 825 5.145 0.3 2.42 .0006 Determined gas rate at 1 kna 60.91 I4J Ambient NO 2 =0.44ppm Ambient CO 2 0.03% Injector size 21 0 It was then decided to test the effect of five layers of mesh. Again the first test commenced using a 3mm injector and the results are shown below.

TABLE 11 4. 0 06 *so so 0 a* 064 a TEMP NO 2 Pres~sure NO 0C (rq/J) kpa (PPM) CO 2

CO/CO

2 750 5.006 0.3 0.8 1.4 .0012 800 4.447 0.4 1.0 1.62 .0008 820 4.387 0.5 1.7 1.80 .0006 840 4.006 0.6 1.8 1.98 .0005 855 4.219 0.7 2.05 2.06 .0005 875 4.146 0.75 2.15 2.16 .0005 Determined gati rate at 1 kpa =41.62 MJ Ambient NO 2='0.44ppm Ambient Co 0.03% 2 Injector size 3mm The injector was then converted back to the standard 2.45mm nozzle and the test repeated. The results are shown in Table 12 below.

22 TABLE 12 TEMP NOoc (ng/J) Pressure kp a

NO

(PPM)

C0 2 675 715 735 755 770 785 795 800 810 3 .603 3.387 3 .38it 3.204 3 .07 3 .144 3 .027 3.084 2.964 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1 0 0 0 0 0 0 ).05 0.1 0.1 0.8! 1.2 1.2 1.3~ 1.4~ 1.52 1.5w

CO/CO

2 .0154 1 .0080 L .0057 1 .0034 6 .0028 8 .0021 5 .0019 1 0ie, 7 .0015 0g S S

S

*WSS

S

SSSS

'S

S

*5@

*SS*

S

55 5 4 5*5

S

*59*

S..

*5*e q*

S

S.

p

S.

p 5

S

Determnined gas rate at 1 kpa Ambient )qO, 0.44ppm Ambient CO 2= 0.03% Injector size 2.45mm 28.76 MJ In k.rder to dispel any thoughts that the reduction in NO xwas somehow related to the "nickel" component of the mesh, the test was repeated again using a fairly standard stainless steel mesh of similar weave and gauge. The iesults shown below do not vary significantly from those achieved using thc- "Inconel" mesh.

23 TABLE 13 TEMP NO 2 Pressure °C (ng/J) kpa

NO

PPM) CO 2

CO/CO

2 695 2.583 0.3 0 0.92 .0162 715 2.782 0.4 0 1.00 .0096 730 2.844 0.5 0 1.11 .0055 755 2.717 0.6 0 1.19 .0043 770 2.587 0.7 0 1.30 .0021 775 2.507 0,8 0 1.37 .0021 785 2.388 0.9 0 1.44 .0018 800 2.292 1.0 0 1.44 .0012 810 2.196 1.1 0 1.55 .0013 9 9 9.99 .9 *s 9 LU 94 s e .9.9 .9.9 0:099 0999 @99999 9 9 9 Determined gas rate at 1 kpa Ambient NO 0.44ppm 2 Injector size 0.03%45mm Injector size 2.45mm 28.76 MJ It was at this stage that it was decided to construct and test a prototype equivalent flat burner.

The tabulated results of the tests are shown below. In both tests it was only the gas pressure that was altered directly in order to effect a corresponding change in temperature. The results in table 14 relate to a flat burner and those in tables 15 and 16 relate to round burners. The results in tables 14 and 16 were obtained using natural gas and those in table 15 were obtained using L.P.G.

24 TABLE 14 FLAT BURNER *S .9 OS S

S

SObO 0* 9 55@

S

9* S. 550 Nj/rn 2 hr Temp at NO 2 Pressure at CO 2 Surface Mesh 0 C ng/J NO Injector %CO/CO 2 Loading 850 3.26 0 1.0 1.32 .0009 580 840 3.43 0 0.9 1.22 .001 551 835 3.06 0 0.8 1.17 .0011 519 g00 2.82 0 0.7 1.71 .0008 486 770 2.66 C) 0.6 1.60 .0009 458 750 2.71 0 0.5 1.45 .0010 410.5 730 2.66 0 0.4 1.33 .0014 367 690 2.473 0 0.3 1.15 .0027 318 640 1.89 0 0.24 1.00 .007 284.4 Determined gas rate at 1 kpa =28 NJ Ambient NO 2 0.086 p.p.m Ambient CO 2 =0.02%.

Natural Gas TABLE 15 ROUND BURN~ER NJ/n 2 .hr Temp at NO 2 Pressure at CO 2 Surface Mesh OC ng/J NO Injector %CO/CO 2 Lodding 740 3.14 0 1.1 1.0 .005 359 760 3.13 0 1.5 1.11 .0045 419 780 3.10 0 2.05 1.29 .002 490 790 3.06 0 2.26 1.23 .0016 514 810 3.00 0 2.75 1.35 .3013 567 820 2.6 0 2.95 1.51 .0009 587 830 ?.74 0 3.5 1.79 .0009 640 5.55 0 S. 55 5Se*,~ 9 0* 5 0 05 Determined gas rate Ambient NO 2 0.086 Ambient CO 2 0.02%

L.P.G

at 1 kpa 28 NJ p-p-m 25

I

TABLE 16 ROUND BURNER MJ/m2.hr Temp at NO 2 Pressure at CO 2 Surface Mesh *C ng/J NO Injector CO/CO 2 Loading 720 750 770 780 790 800 3.00 2.75 2.80 2.70 2.96 2.80 0.5 0.7 0.8 1.0 1.1 1.2 0.66 0.77 0.80 0.89 0.92 0.96 .0086 .0041 .0025 .0018 .0018 .0015 409 484 517 578 606 633

S

a

S.

.9

S..

0 Determined gas rate at 1 kpa 22 MJ Ambient NO 2 0.086 p.p.m Ambient CO 2 0.02% Natural Gas As the results obtained on the flat burner in Table 14 looked promising, a further set of four tests were conducted in the same manner. The results of the tests were averaged and are shown in the Table below.

TABLE 17 FLAT BURNER 9 TEMP NO 2 Pressure CO 2 °C (ng/J) kPa MJ/m 2 hr CO/CO 2 850 1.6 1.0 598 1.42 .0005 835 1.8 0.75 519 1.23 .001 750 1.8 0.5 423 1.38 .0011 Determined gas consumption at IkPa 29.55 MJ 26

I

Using the tabulated data disclosed, a series of graphs were generated to assist in interpretation of the results and enable the data to be used in the development of future burners.

In all the graphs the curves are identified by reference numerals corresponding to the table number from which the data was extracted such that a curve identified as Ti corresponds to the result illustrated in Table 1. The column from which the'data was taken \Q will be evident from the. variables designated to each of the axes of the graph. In all graphs the units correspond to those given in the tables.

Figure 5 illustrates the relationship between temperatures (on the x-axis) and NO 2 (on the y-axis) according to the data found in Tables 1 to 4 inclusive and Tables 15 and 16 for the first cylindrical embodiment and Table 14 and Table 17 for the second flat surface embodiment.

*Similarly, Figure 6 shows the relationship between *54e burner loading (on the x-axis) and NO 2 (on the y-axis) for the same configurations of the burner.

a *4 It is clear from these results that irrespective of the operating conditions, the burner can be considered to show inherently low emission levels of NO 2 It is also clear that the best results are achieved when the burner is run at its design loading. Overloading the 27 (W burner represents a step change to an increase in NO 2 emission levels. However, the curve T4 shows clearly that if the burner air/gas ratio is to be maintained at approximately stoichiometric, there is a clear optimum maximum burner loading for the cylindrical burner at least of about 500 MJ/M2hr, above which the rate of increase in NO 2 emis'-ons escalates.

Figure 7 illustrates the effect of excess air (on the x-axis) with respect to NO 2 levels (on the y-axis) in accordance with the results shown in Tables 5 to 8 S" inclusive). Whilst it appears that additional readings may have been beneficial, it shows clearly that NO 2 levels decrease with an increase in air component such that beyond an excess of 20% the addition of yet further *a primary air has no appreciable effect.

In summary, the above results indicate the burner can still be operated at stoichiometric with what is r a considered to be still low NO 2 emission levels.

Furthermore the excess air enables the burner to run in .o .20 an ultra-low NOX condition, where the air is providing an additional coolant to the combustion reaction. The burner, as previously mentioned, can also run in an o: overloaded condition such that the flame extends beyond the combustion surface. In this condition the nitrogen dioxide level is still very desirable in comparison to standard blue flame burners where the NO 2 levels are normally in the order of 15-20 ng/Joule.

28 yW Figure 8 has been configured to provide a means of determining a relationship between the combustion efficiency of the burner illustrated by the CO/CO 2 ratio and the port loading required to achieve those combustion levels.

Due to differing CO/CO 2 level requirements, depending upon local regulations and venting necessities, the burner can be operated over a broad spectrum. This graph provides a facility to determine the minimum port loading (thus' lower NOx) for the corresponding combustion level requirement.

Figure 9 shows the results of some preliminary investigations to determine whether different burner *o: lo, combustion surfaces would pertain to a variation in NOx products. Burners were assembled using stainless steel mesh; four layers of inconel; and five layers of 0a inconel mesh.

e, u The stainless steel mesh gave comparable results to the standard three layers of inconel. The four and five layer systems gave a contradiction in results and produced levels of nitrogen dioxide in excess of what Y was anticipated. An increased number of layers was expected to produce an increase ii, time for the combustion reaction to take place therefore the burner could run at cooler temperatures and still maintain efficient combustion, the cooler running temperature was expected to give lower NOx.

29 The four layer system produced higher NO than the three layer. The five layer burner, however, gave lower NO X results than the four.

By pooling the results depicted in Figures 5 to 9 discussed above, it was possible to generate a further set of graphs indicating the general relationships between the important variables for production of a low

NO

X burner. Accordingly Figures 11 to 14 inclusive can be used to determine burner loading, combustion

CO/CO

2 ratio, excess air required and the NO 2 level achieved. These graphs were not updated due to time constraints to show the results obtained on the second embodiment flat burner which reduced the emission levels obtained by a further 25% on average.

Whilst the tests were limited to use of rrsh of a specific size and'weave, it is understood that by varying the conductivity and porosity of the combustion surface, a variation in port loading would be required to achieve the same operating temperature. Similarly, r materials other than consecutive layers of mesh, such as for example a sintered metal material having similar pressure drop, porosity and conductivity characteristics, would probably achieve the same results.

It also has to be recognised that in cases where the low-nox burner was overloaded the flame lifts from the mesh surface to a height of up to 6" depending 30 on input. The most surprising development was that, in such conditions the nitrogen dioxide emission was still in the order of <5ng/J as shown in Figure 10. This obviously has advantages in ornamental log fire and gas stove burner design.

Whilst the majority of the tests centered on the first embodiment being the cylindrical burner, it is now evident that the shape was not contributing to the low levels achieved. The limited data obtained on the flat burner indicates that in fact a more even combustion can o S

S

be obtained enabling the burner to operate at even lower *o: NOx levcls. It appears upon analysis that the cylindrical burner is in fact a compromise as it is more *s e compact for a given output, but that due to the curvature of the mesh it is not possible to obtain an even temperature across the combustion surface.

Accordingly it is necessary to run at slightly higher temperatures in order to maintain good even combustion.

sees It is therefore believed that further tests and development of the flat surface burner will reduce the NOX emissions even further.

It will be appreciated by those skilled in the art that the foregoing describa only two embodiments of the invention, and that as discussed modifications could be made thereto to produce burners for other applications without departing from the scope of the invention.

31

Claims (17)

1. A naturally aspirated gas burner apparatus including a plenum chamber having a combustion surface formed from a conductive heat resistant material of selected area and porosity, the porosity being substantially uniform across and through the combustion surface, a fuel supply, an air/fuel mixing and delivery device extending into said chamber, the delivery device being adapted to supply an air/fuel mixture with an air component at least equal to that required for theoretical complete combustion into said plenum chamber for egress through said combustion surface and combustion at or adjacent the surface thereof, said porosity providing a flow rate of air/fuel mixture therethrough that results in a combustion temperature of 600°C to 900°C at said combustion surface whereby the formation of oxides of *4 nitrogen in the products of combustion is reduced to 4 about 5ng/Joule or below.

2. A gas burner apparatus according to claim 1 wherein the selected temperature is in the range of from 760°C to 850 0 C.

3. A gas burner according to claim 1 wherein the porosity of the combustion surface is in the range of to

4. A gas burner apparatus according to any one of the preceding claims wherein said combustion surface has openings therethrough dimensioned so as to create a labyrinthine path to prevent reflected infra-red energy .11 from being returned to the plenum chamber from an 32 adjacent object. A gas burner apparatus according to any one of the preceding claims wherein said delivery device is adapted to supply an air/fuel mixture with an air component of between 10% and 60% in excess of that required for theoretical complete combustion.

6. A gas burner apparatus according to any one of the preceding claims wherein the conductive porous heat resistant material is in the form of one or more layers of metallic mesh material.

7. A gas burner apparatus according to claim 6 wherein the combustion surface comprises three layers of 30 x 32 x 0.014" nickel-based steel mesh of 32% porosity.

8. A gas burner apparatus according to any one of the preceding claims wherein the combustion surface has a porosity and pressure drop equivalent to three layers of x 32 x 0.014" steel mesh of 20% to 60% porosity.

9. A gas burner according to claim 6 wherein the mesh material has a weave pattern disposed at approximately :I V. 450 to the longitudinal and lateral extent of the plenum chamber. 1. 0. A gas burner apparatus according to any one of the preceding claims wherein the plenum chamber is substantially cylindrical. e# S11. A gas burner apparatus according to claim wherein the combustion surface comprises approximately one longitudinal half of said cylindrical chamber. S"12. A gas burner according to claim 10 wherein the 33 plenum chamber is made from an extrusion having sealing end surfaces attached thereto.

13. A gas burner apparatus according to any one of claim 10 to claim 12 wherein the plenum chamber includes a plurality of radially extending cooling fins.

14. A gas burner apparatus according to any one of claim 6 to claim 13 wherein the plenum chamber includes two longitudinally extending serrated lugs deformable to secure said layers of mesh material to the sides of said chamber. A gas burner apparatus according to claim 1 wherein the combustion surface is substantially planar.

16. A gas burner apparatus according to any one of the preceding claims further comprising means for distributing the air/fuel mixture in said plenum chamber.

17. A method of operating a gas burner of the kind comprising a plenum chamber having a combustion surface formed from a conductive heat resistant material of selective porosity, the porosity being substantially uniform across and through the combustion surface a fuel supply, an air/fuel mixing and delivery device extending *into said chamber, said method comprising steps of: delivering to the chamber an air/fuel mixture with O.V an air component at least equal to that required for theoretical complete combustion for egress through said substantially uniform combustion surface and combustion at or adjacent the surface thereof; and, "s lecting the porosity of the combustion surface to 34 achieve a combustion temperature of between 600 0 C to 900°C at said combustion surface whereby the formation of oxides of nitrogen in the products of combustion are reduced to about 5ng/Joule or below.

18. A method of operating a gas burner accord. ,g to claim 17 including the step of selecting a combustion surface porosity of between 20% and

19. A method of operating a gas burner according to claim 17 or claim 18 comprising the step of selecting a combustion temperature at said combustion surface of between 760°C 850°C. A method of operating a gas burner according to any one of claims 17 to claim 19 wheiein the temperature is selected by adjusting the port loading for a given combustion surface.

21. A method of operating a gas burner according to claim 20 wherein the fuel delivery system is adjusted to achieve a combustion surface loading of 200-650 MJ/m 2 hr on said combustion surface.

22. A method of operating a gas burner according to any one of claims 17 to claim 21 further comprising the step of adjusting the air/fuel mixing and delivery device to provide an air supply of between 10% and 60% in excess of that required for theoretical complete combustion. :23. A method of operating a gas burner according to any one of claim 17 to claim 22 including the step of increasing the combustion surface loading to approximately 300% of the design loading to produce a 35 flame beyond the combustion surface.

24. A gas burner apparatus substantially as herein described with raference to the accompanying drawings. A method of operating a gas burner substantially as herein described with reference to the examples and accompanying drawings. DATED This 26th Day of November, 1992 BOWIN DESIGNS PTY LTD Alttorney: LEON~ K. ALL.EN Fellow Institute of Paitent Attorneys of Australi~p? of SHELSTON WATERS 36