Clean combustion of wood?

Clean combustion of wood?

Citation for published version (APA):

Khan, A. M. H. R., & Verhaart, P. (1989). Clean combustion of wood? Eindhoven University of Technology.

Document status and date: Published: 01/01/1989

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

Clean Combustion of Wood

Part II

by

Dr. A.M. Hasan R. Khan

*

The woodburning Stove Group Eindhoven University of Technology

Eindhoven, The Netherlands

*

Permanent address: Institute of Fuel Research and Development, Bangladesh Council of Scientific and Industrial Research.

Contents

page

nr

ACKNOWLEDGEMENTS ll

1. INTRODUCTION 1

2. THE STOVE DESIGN AND EXPERIMENTAL. PROCEDURE 3

3. RESULTS AND DISCUSSION 5

3.1 The General Behaviour of The Downdraft Stove 5 3.2 Effect of Wood Species on The Quality of

Combustion 10

3.3 Effect of Chimney Height on The Quality

of Combustion 14

3.4 Efficiency at Optimum Charging Rates ·for

Different Constant Chimney Heights 24

4. CONCLUSIONS 35

Acknowledgements

The work reported in this study was made possible by a fellowship granted by the Commission of the European Communities, Brussels to Dr. A.M. Hasan R. Khan and arranged by the Science and Technology Division, Ministry of Education, Government of the People's Republic of Bangladesh. The work was supervised by Dr. K. Krishna Prasad of the faculty of Physics, at the Eindhoven University of Technology. For this investigation, the laboratory facilities of the Woodburning Stove Group were used. Dr. Hasan was assisted by Ernst Schutte, Piet Verhaart and Niek Verhoeven in the laboratory and data processing.

1.

Introduction

The biomass stoves commonly used in developing countries are not properly designed. As a result, they consume unnecessarily large amounts of fuel during cooking and other heating purposes and in addition pollute the kitchen atmosphere due to incomplete combustion, which subsequently affects the respiratory system of the users. During the last decade, the first problem was given top priority and attempts were made to solve this by proper designing the biomass stoves. Some limited measurements indicate that improved designs save 50 to 803 fuel when compared with the traditional ones. But very little attention was given, till now to the second problem. Since 1985, prevention of kitchen atmosphere pollution by incomplete combustion of biomass has become a principal point of interest at the Eindhoven based Woodburning Stove Group.

Wood consists of non-poisonous chemical elements viz. Carbon, Hydrogen and Oxygen. If it is burnt completely there should be no air pollution, because the end products of these reactions are water and carbon dioxide. But practically it never happens. When wood is heated, at 250-3QQOC, about 803 of the original dry mass of wood is expelled in the form of gases, vapours and finely divided droplets, leaving behind charcoal. To ensure complete combustion, sufficient amounts of air should be supplied both to the volatiles and the charcoal.

In a classical stove, fuel and air travel in a counter-current flow. The primary air enters below the grate and moves upward through the fuelbed consisting of primarily charcoal. During that time oxygen in the air meets with the charcoal heterogeneously at its surface and forms CO (Krishna Prasad et al., 1984). This CO migrates into the spaces between the particles and reacts with oxygen homogeneously and forms C02 and heat. Again this C0 2 reacts heterogeneously with the carbon surface and forms CO, which is an endothermic reaction. As a result, there is reduction in temperature of the flue gases. Thus the conversion of C0 2 into CO, which is dependent on the temperature and the time, will be less intense in the higher layers of the charcoal bed. Still the temperatures are high enough to cause pyrolysis of the fresh wood and the liberation of volatiles. When the volatiles emerge

from the top of the fuelbed, they hold gaseous and tarry decomposition products that derive from wood in the upper layers of the fuelbed and they contain no oxygen (conventional wisdom suggests that under such a design approach, the amount of air supplied at the bottom of the grate is only sufficient for the combustion of charcoal). The volatiles can only be burnt by supplying a sufficient amount of secondary air into the system. However, even then the efforts to burn it do not always meet with success, because (a) the air will further cool down the gas mixture and (b) the way in which the air is supplied not necessarily guarantees its proper mixing with the fuelbed effiuent. Incomplete burning will result in the deposit of tar and soot on the bottom of the utensil and large emissions of irritants, toxins and carcinogens in the kitchen environment.

The present work is a continuation of earlier work (Khan & Verhaart, 1989) in which experimental results were presented on a so-called downdraft stove, to relate stove design with some operational variables viz., size of woodblocks, charging intervals, effect of secondary air by-pass, power output etc. In the present investigation, a design parameter, viz. chimney height was examined in detail. Besides this, more experiments were done to attempt to relate stove design with wood species. Although the present design can not be considered as a practical cooking device, efficiency measurements were carried out. Especially the influence of cold pans on the temperature behaviour and combustion quality were examined.

The principle of operation of the stove, in contrast to conventional designs, is that the flow of air is in the same direction as the volatiles and fuel. A chimney for the stove is essential to provide the draft, which induces the liberated volatiles and air to flow downward through the fuelbed where they burn vigorously resulting in much higher temperatures (950 - llOOOC) than during conventional burning (550 ~ 75QOC). It has been experimentally found that this mode of burning leads to a very good combustion for certain operating conditions. It is suggested that this is due to:

i) more homogeneous mixing of volatiles ·and air due to higher flow velocities than in conventional combustion;

ii) higher temperatures; and

iii) shorter time interval between reaction of air and volatiles which prevents condensation and polymerization reactions resulting in products, which are more difficult to convert to C02 and H20.

2. The Stove Design and Experimental Procedure

All the experiments of the present investigations were carried out with the downdraft stove-III, as shown in the figure 2.1.

The specifications of this stove are different from those which were used in the earlier experiments (Khan & Verhaart, 1989). The latter had a chimney height of 1000 mm and a diameter of 105 and 95 mm, respectively. The distance between the fire-bowl and the chimney entrance was 190 and 0 mm, respectively. Figure 2.1: I ~ I - I 'i-' ·' ¢100 0 8

The combustion zone of these stoves had a height of 40mm. Both the stoves got burnt away due to high power output combined with continuously high temperatures which caused corrosion of the metal. Stove III was made with stainless steel. The present stove is capable of power outputs from 3.0 -9.0 kW. (This depends on the chimney length chosen). The entire stove was covered with insulating material (A1203).

The experimental procedure, data collection and its processing are given in Khan & Verhaart (1989). To be able to compare the results of the present investigations to the earlier ones, volume/surface areas of the woodblocks are maintained approximately the same.

3.

Results and Discussion

3.1 The General Behaviour of the Downdraft Stove

The general behaviour of the downdraft stove is illustrated by analysing one typical experiment in detail. The experiment chosen is SI. No.2 from tabl~IV (charging rate = 2 blocks/30 s; power output = 5.9 'kW; chimney height = 62 cm; wood species = White Fir; and volume/surface area =

0.360 cm). The variables studied are stated below. Concentration of CO and C02 in the flue gases

The concentration of CO is used as an indication of combustion quality. Figure 3.1 represents the concentration of carbon monoxide and carbon dioxide in the flue gases as a function of time.

0.20 20 0.16

\~I~

r-;

15 N 0 _{0.12 0}

u

_u

'<l- 10 '<l-_J _J 0 0.08 0

>

1l

5

>

0.04 0.00

vJ

0 0 10 20 30 40 50 60 TIME (MIN.)

Figure 3.1: Concentration of CO and C02 in the flue gases as a function of time.

The figure shows that the concentration of carbon monoxide. in the flue gases is very small and remains almost constant during the experiment. The

average concentration of carbon monoxide in the flue gases is 0.009

3

by volume ( u

=

0.007). The figure shows some large peaks. This is probably due to poking of the fuelbed during the experiment. The concentration of carbon dioxide varies from 10 ~ 13

3

by volume. The average concentration is 11.5

3

by volume

(u

=

1.48).

Concentration of oxygen and CO/ C02 ratios in the :flue gases

The marker for measurement of clean combustion in the combustion gases is the CO/C02 ratio, whereas CO and C02 are considered as incomplete and complete combustion products respectively. In reasonably good combustion, this ratio must be kept as low as possible (preferably smaller than 13).

Figure 3.2 _{represents the concentration of oxygen and CO /C02 ratio in the} flue gases. N 0 ~ 0 _J 0

>

Figure 3.2: 15 ~---, 2.00

~

1.50 10 ~ ~ N 1.00 0

u

_-... 0 5

u

0 ~--~--~--~---~--~--~ 0.00 0 10 20 30 40 50 60 TIME (MIN.)

Concentration of 02 and CO/C02 ratios in the :flue gases as a function of time

The CO/C0 2 ratio was almost constant during the experiment and its average value is 0.078

3

(u = 0.059). Even the peaks of CO/C02 that occur frequently are under 0.5

3

for most of the time during the experiment. This is excellent compared to the maximum allowable ratio. The concentration of oxygen varies from 8 - 11

3

by volume and the average oxygen content in the flue gases is 9

3

by volume ( u = 1.32).

Excess air factor

Figure 3.3 represents the excess air factor in the flue gases. The horizontal line through the experimental data indicates the average value, which is 1. 75 ( q = 0.23). Thus clean combustion is obtained without too much excess air.

a:

0

1-u

<( LL

a:

<( 4 3 (j) 2 (j) w

u

x w Figure 3.3: 0 Power output 10 20 30 40 50 60 TIME (MIN.)

Excess air factor in the :flue gases as a function of time

The average power output can be computed by using the following equation:

Where:

p

-average power output (kW)

sum of the charge weights (kg) ·

the as-fired combustion value of the fuel (kJ /kg) total duration of the experiment (s)

(3.1)

The fuel consumed as a function of time is shown in figure 3.4. The figure shows that the fuel consumption during the experiment is linear. The average power output of this experiment, computed by means of figure 3.4

and equation 3.1 is 5.9 kW (u

=

0.72).

Power density

The average power density is calculated by using the following expression.

(3.2)

Where:

Pp average power per unit grate area (W /cm2) A grate area (cm 2)

With a gr~te diameter of 12 cm (area is 113.1 cm2) and an average power output of 5.9 kW, the power density of this experiment is 52 W /cm2.

800 Oi 0 w 2 600 :J (f) z 0

u

₄₀₀ _J w :J LL 200 0 10 20 30 40 50 60 TIME (MIN.)

Figure 3.4: Fuel consumed as a function of time.

Chimney draft

The pressure difference bet ween the surrounding air of the stove and the flue gases in the chimney during operation of the stove is called the chimney draft. The equation generally used for chimney draft is (Verhaart, 1981).

(3.3) where:

P ch chimney draft (Pa)

Pa density of ambient air (293 K - 1.21 kg/m3)

pH density of the flue gases taking the average temperature over the chimney (kg/m3)

H height of the chimney (m)

g gravitational acceleration (9.8 m/s)

During the experiments, the temperatures of the flue gases were measured at the chimney entrance and exit. These temperatures were used in determining the densities of the flue gases. The values of the density can be derived from tables (Eckert and Drake, 1972) taking the physical properties of air as those of the flue gases. Figure 3.5 presents the chimney draft during the experiment. It is to be noted that the chimney draft is almost constant in

8 -ro Q, f---u_ ₆ -<: a: 0 _~

>-w

z

4 -2 I () 2 0 10 20 30 40 50 60 TIME (MIN)

.

Figure 3.5: Chimney draft as

a

function of time.

the experiment. The average chimney draft of the experiment is 5.18 Pa ( u = 0.13). It should however be very clear that the chimney draft calculated and presented in figure 3.5 is different from the net draft through the system. To compute the net draft, the pressure losses due to flow resistances

in the stove, are to be deducted from this chimney draft. The author has however not been able to make an estimate of these pressure losses.

Qualitative tests of the flue gases

During the course of the experiment some qualitative tests of the flue gases were made (Khan & Verhaart, 1989).

(i) Smell

(ii) Inflammability

(iii) Presence of dust particles (iv) Accumulation of charcoal

The smell was determined by sniffing the flue gases from time to time. The flue gases issuing from this specific experiment have no smell, no dust particles and did not catch fire when a flame was held just above the chimney. During burning of the woodblocks in the fire bowl, there was also no accumulation of charcoal on the grate.

3.2 Effect of Wood Species on The Quality of Combustion

As mentioned earlier, the basic ingredients for forming wood are carbon dioxide and water. They are converted by photosynthesis into three main constituents of wood viz. cellulose, hemi-cellulose and lignin. The quantities of these constituents vary with wood species. Normally hard wood contains about 43% cellulose, 35% hemi cellulose and 22% lignin, while soft woods contain about 43% cellulose, 28% hemi cellulose and 29% lignin (Shafizadeh & De Groot, 1976). The combustion values of wood fuel largely depend on the chemical composition of the wood species. For example cellulose & hemi cellulose have a calorific value of about 17.5 MJ/kg, whereas lignin has 26.6 MJ/kg. Thus it may be concluded that wood having a higher lignin content has a higher combustion value (Tillman, 1978). The composition of wood also determines how it releases iU energy. The combustion of wood starts with pyrolysis. During pyrolysis, cellulose and hemi-cellulose particularly promote the release of volatiles, while lignin mainly promotes the formation of char.

Although the effect of wood species was studied earlier (Khan & Verhaart, 1989), it was at that time not possible to carry out the experiments with similar volume/surface areas of the woodblocks due to non-availability of proper block sizes of Oak and Beech. But the nearest possible sizes were

used. Secondly the aimed charging rate of Oak and Beech could not be maintained because of too high power output. While using them in the experiments, the entire apparatus became red hot. The power loading capacity of the earlier stove (downdraft stove-II) was 3 - 7 kW.

The present design of the Downdraft stove-III, has a power loading capacity of about 3 - 9 kW, which is higher than the earlier ones. Therefore this parameter was studied again for similar volume/surface areas of the different species. The wood species examined are White fir, Beech, Meranti and Oak. The results are listed in the table-I. Figure 3.6 represents the average concentrations of CO, C02 and 0 2 in the flue gases for different wood

species. The figure shows that the average CO and C02 concentrations in

the case of Beech, Meranti and Oak vary from 0.007 - 0.01

3

and 10.9 -12.54

3

respectively. Whereas in the case of White Fir, the concentrations of CO and C02 are 0.055

3

and 8

3

respectively. The differences can be

explained as follows. The density of white Fir is smaller than those of Beech, Meranti and Oak. Thus with similar volume/surface areas, the average weight of each woodblock of White fir is less than that of Beech,

0.04 0

u

'(12. _J 0

>

0.02 0.00 Figure 3.6: - Vol%

co

c:::J

Vol%

co

2 "' 0 ';:f?, ₀ 10 ₀ _J

>

"' 0

u

5 '(12. _J 0

>

White Fir Beech Meranti Oak

WOOD SPECIES

Average concentrations of CO, C02 and 02 in the flue gases for

TABLE !:EFFECT OF VODD SPECIES

Chi11ney height (c•): 100

Yol1111e/surface area (c•): 0.360

Charging rate (blocks/sec): 2/30

A.verage

_Dens

it~

Po11er

A.verage fl•e gas composition

SL.

food

11eight

{kg/1113 o.tpat

Ir. species

1 block

{kY}

co

CD2

12

CO/CD2

(g)

{Yol.7.} {Yol.'!.} {Yol. 7.)

fl.}

1

Vhite Fir

5.4 410 5.9 0.055 7.99 12.62 0.76 2

Beech

.

6.8 650 8.2 0.008 12.00 8.72 0.06 3

leranti

7.0 600 7.2 0.007 10.90 9.60 0.06

4

Oak

7.0 620 8.1 0.010 12.54 8.30 0.09

*

No smell in the flue gases

=

I

Presence of dust particles in the flue gases

=

D

Absence of dust particles in the flue gases

=

A

No accumulation of charcoal_bed on the grate

=

N

Excess Dra•jht

air

{Pa

le•arks

*

/actor

2.50 7.51

I, D, N

1. 74 8.51

I, D, N

- 8.34

I,

A,

N

Meranti and Oak (see also table-I). Therefore at similar charging rates (blocks/s) for the different wood species, the power output for White Fir will be less (5.4 kW against 7 kW). These experiments were done with a

100 cm chimney. In section 3.3, it will be shown that each chimney length has an optimum power output at which the lowest CO/C0 2 ratio is measured. In case of a 100 cm chimney, this optimum is found around 7.5 kW. It _{is therefore understandable that higher CO and lower C02 values are}

recorded in the case of White Fir. Figure 3.7, which shows the average CO /C02 ratios an.d excess air factors for the different wood species, shows the same behaviour and thus confirms this (The excess air factor for Meranti and Oak could not be calculated due to non-availability of their compositions. The computation method that was used here is described by Sielcken, 1983). 0.60 ~ ~ N 0 _0.40

u

_... 0

u

0.20 Figure 3.7:

- COIC02

C::.=J

Excess air

factor

White Fir Beech Mer anti Oak

WOOD SPECIES ([ 0

f---u

<( LL 2 ([ <( ({) ({) w

u

x w

Average CO/C02 ratios and excess air factors in the flue gases for different wood species.

Due to the higher power outputs and smaller quantities of 02, the temperatures of the flue gases for these wood species were higher which resulted in higher chimney drafts (see table-I).

3.3 Effect of Chimney Height on The Quality of Combustion

The chimney is an essential component of the downdraft stove to induce air flow in a direction opposite to . the buoyancy forces produced by the burning fuel on the grate of the stove. However for the classical burning mode, a chimney is not necessary for the maintenance of burning since the buoyancy forces will s11pply the necessary air to the burning zone. The chimney is mainly used to lead the burnt/unburnt combustion products away from the kitchen environment. In this section we will present the effect of chimney height on the performance of the downdraft stove. The average results are presented in tables.

Effect of varying chimney height with constant charging rate

To maintain the same nominal power output at different chimney heights, the same charging rate was applied. The chimney height was varied from 43 cm to 100 cm. Figure 3.8 shows the effect of chimney height on the carbon monoxide and carbon dioxide contents in the flue gases. It clearly shows that with the increase in chimney height, the CO content of the flue gases decreases to 0.009

3

(vol.) at 62 cm chimney height and then again increases to 0.055

3

(vol.) at 100 cm chimney height.

T!BLE II: EFFECT Of CHIINEY HEIGHT !T CONST!NT CH!lGING l!TE

Yood species: Yhite fir

Volume/surface area (ca): 0.360

Charging rate (blocks/sec): 2/30

Chi1mey

Po11er

Average flae gas composition

Sl.

height

oatpst

/fr.

(c•}

{kY}

CD

CD2

02

Cl/Cl2

{Yol. X) {Yol.X} {Yol.

1.)

(1.)

1 43 5.10 0.049 13.61 7.05 0.38 2 50 5.·30 0.037 13.53 7.15 0.23 3 62 5.88 0.009 11.51 8.99 0.078

.

4 85 5.85 0.026 9.02 11.59 0.33 5 100 5.84 0.055 7.99 12.62 0.76

*

Smell in the flue gases

No smell in the flue fases

Inflamability of the lue gases

Presence of dust particles in the flue gases

!bsence of dust particles in the flue gases

!ccumulation of charcoal-bed on the grate

No accumulation of charcoal-bed on the grate

Excess

air

factor

1.40 1.47 1. 75 2.20 2.50

=

s

=

x

=

I

=

D

=

A

=

c

= N

Draajht

{Pa

le•arks

*

3.45

S, I, !, C

4.12

X, I, !, C

5.20

X, !, N

6.76

X, !, N

7.51

X, D, N

The C02 content of the flue gases decreases with an increase in chimney height. It can be seen from table-II that the chimney draft increases with increasing chimney height and allowing more air to pass through the system, while the power output remains relatively constant. Thus the concentration of C02 decreases due to dilution with more air. The optimum CO value,

found in this experiment at a 62 cm chimney height is only valid for this particular charging rate applied. For other charging rates, the optimum value would probably be shifted to the right or to the left. However the fact remains that for all the chimney heights examined, the average CO values at this power output (5.5 kW) are very low.

Figure 3.9 shows the effect of chimney height on the CO/C02 ratios and

oxygen contents of the flue gases. The figure clearly shows that the CO/C02

ratio decreases with an increase in chimney height and attained the lowest

14 0.60 12

D

_!:::.

..

0 N 0 _{0.40 10} 0 ... 0

'*

()

d

-.

l

>

0.20 8 0.00 ...__ _ _ __._ _ _ _ _._ _ _ _ ..._ _ _ _ .__ _ _ ___. 6 20 40 60 80 100 120 CHIM-EV I-EIGHT (CM)

Figure 3.9: Average concentrations of oxygen and CO/C02 ratios in the :flue gases for different chimney heights

value of 0.078

3

at a 62 cm chimney height and thereafter increases to O. 76

3

at 100 cm chimney height. However it is striking that all the CO/C02 ratios obtained at different chimney heights remain below 0.83.

(for this specific power output). It was found that in case of a metal stove with chimney the CO/C02 ratio varies from 5.4 - 9.6

3

(Vermeer and

Sielcken, 1983). The concentration of oxygen increases with the increase in chimney height, which is obvious, as the draft increases.

Figure 3.10 shows the excess air factor of the flue gases and chimney draft for different chimney heights. The dashed line represents the total amount of air that flows through the system (VT in m3/kg of wood). Although the rate of increase of the excess air factor is smaller than that of the draft, the total amount of air runs parallel to the increase in chimney draft.

D

5

t-~

LL

a:

<i

~

w l)

x

w 12 9 9 6 6 3 3 ( 0 '--~~---'-~~~-'-~~~..._~~----'----~~-' 0 20 40 60 80 100 120 CHlf'vf'..EY I-EIGHT (CM)

Figure 3.10: Chimney draft, excess air factor and total volume of gases (in m3/kg of wood) for different chimney heights

Effect of varying charging rate with constant chimney height

It has been observed that a good. quality of combustion in a stove, at a particular chimney height depends on the charging rate of fuel (applied power output). In other words, there appears to exist a correlation between chimney height and charging rate of ·fuel in order to obtain clean combustion. Clean combustion for a range of power outputs can only be obtained in case the chimney is adapted to the applied power output. In order to ascertain this effect, a series of experiments have been carried out at three different chimney heights viz. 43, 62, 100 cm. with different power outputs. The results are listed in the table III, IV and V.

... 00

TABLE Ill: EFFECT OF PDYEI OUTPUT VITI CONSTANT CHIINEY HEIGHT

Chi11ney height (c•): 43

Vood species: Vhite fir

Voluae/surface area (c•): 0.360

Charging

.

Po•er

'verage fl•e gas co#lposition

•

Sl. rate

oat pat

Kr. {blocks/s) {kY}

Cfl

Cfl2

112

1 2 3 4

{YoL.

7.)

(Yol.

7.)

(Yol.

7.)

2/30 5.07 0-.049 13.61 7.05

2/35 5.09 0.021 10.27 10.10

.

2/45 4.0 0.055 6.80 13.72

2/50 3.40 0.115 5.30 15.00

Smell in the flue gases

No smell in the flue gases

Inflamability of the flue gases

Absence of dust particles in the flue gases

Accumulation of charcoal-bed on the grate

No accumulation of charcoal-bed on the grate

Cfl/Cfl2

fl.) 0.38 0.21 0.87 2.34

Excess

air

factor

1.4 2.1 3.2 4.1

=

s

=

I

=

I

=

A

= c

=

N

Chimney temperatare (OC}

Sensible

lra•1ht

heat

le•arks

:1

Entrance

Exit

{Pa

Losses

fl.)

708 555 3.54 27.6

S, I, A, C

709 634 3.53 46.3

I, A, N

580 561 3.35 59.4

I, A, N

TABLE IV: EFFECT OF POYEl OUTPUT VITH CONSTANT CHIINEY HEIGHT

Chilllley height (ca): 62 Vood species: Vhite fir

Voluae/surface area (ca): 0.360

Charging

lor1er

Average flae gas composition

Excess

Sl. rate

oat pat

air

Ir. {blocks/tJ) (kY}

Cl

C02

02

CO/C02 factor

(Yol. 7.) (Yol.7.} (Yol.'!.}

fl.)

1 2/25 6.3 0.016 13.04 7.65 0.124 1.65 2 2/30 5.9 0.009 11.51 8.99 0.078 1. 75 ' 3 2/45 4.1 0.092 6.63 13.85 1.46 3.22 4 2/60 3.1 0.146 4.41 16.04 3.58 4.92

*

No smell in the flue gases

=

I

- Presence of dust particles in the flue gases

=

D Absence of dust particles in the flue gases

=

A No accumulation of charcoal-bed on the grate

=

N

Chimney temperature (OC}

Sensible

Draajht

heat

le11arks

-Entrance

Exit

{la

losses

(1.)

722 692 5.16 15.9

X,

A, N

772 658 5.18 16.1 I, A, N

556 525 4.71 22.8 I, D, N

TABLE V: EFFECT OF POVEI OUTPUT VITH CONSTANT CHIINEY HEIGHT

Chi1111ey height (ca):

100

Vood species: Vhite fir

Vol1111e/surface area (ca):

0.360

Charging

!011er

Average fl•e gas composition

Excess

SL. rate

ostpst

air

Ir. {blocks/s) {kf}

CD

CD2

02

CD/CD2 /actor

{Yol.7.} (Yol.'/.} (Yol.

i.) (i.)

1 3/30 8.3 0.048· 14.67 6.10 0.267 1.35

2 2/25 6.7 0.016 9.67 11.04 0.18 2.22

3 2/30 5.9 ' 0.055 7.99 12.62 0.76 2.50

4 2/45 4.6 0.126 5.62 14.97 2.53 4.0

*

- No smell in the

.

flue gases

=

I

Presence of dust particles in the flue gases

=

D

- Absence of dust particles in the flue gases

=

A

No accumulation of charcoal

-

bed on the grate

=

N

Chimney temperatsre

(O

C}

Sensible

Dra•1ht

heat

lemarks

~

Entrance

Exit

(

Pa

losses

(7.)

860 720 8.60 14.1

I,

A,

N

685 587 8.05 17.9

I, D,

N

659 389 7.51 15.3

I, D,

N

As an example, figure 3.11 shows the effect of power output on the CO /C02 ratios of the flue gases as a function of time for a 62 cm chimney.

It _{clearly shows that at a power output of 5.9 kW, the minimum CO/C0 2}

ratio is recorded {0.0783). By reducing the power output it increases sharply. It is also evident from the figure that if the power output is increased beyond 5.9 kW the CO/C0 2 ratio tends to increase slowly. A similar behaviour was recorded in case of the 43 cm and 100 cm chimney height. 6

"'

0 4 0 ... 0 0 2

.

,._. 6.3 kW

-9-

5.9

kW

_.._ 4.1 kW

-II-

3.1

kW

0

!!111m1._...._ ....

~.,__..

....

.._IMlllij . . ._..._. . . .

i1W1-.

0 Figure 3.11: 10 20 30 40 50 60 Tl~ (MIN}

Effect of power output on the CO /C02 ratios in the flue gases as a function of time with a 62 cm chimney.

The average CO/C02 ratios for varying power outputs at different constant chimney heights are plotted in figure 3.12. The figure shows that a reduction in power output at 43, 62 and 100 cm chimney heights, results in a sharp increase of the CO/C02 ratios.· The highest values of· CO/C02 ratios at 43, 62 and 100 cm chimney heights are 2.34, 3.58 and 2.533 respectively. On the contrary with the increase of power output at the different constant chimney heights, the CO /C02 ratios decrease until they reach their minimum values and afterwards increase slowly. The minimum CO/C02 values recorded for the 3 chimney heights correspond to different power outputs {4.7, 5.9 and 7.3 kW respectively). It may be concluded from

the above findings that it is possible to design a domestic downdraft stove with a chimney height of 43 to 50cm, that can be operated at 4 - 5.5 kW with very low CO/C02 contents in the flue gases.

4 3 ~ ~ N 0 2

u

..___ 0

u

3 Figure 3.12: .6. 43 cm

•

62 cm 0 100 cm 6 9 POWER OUTPUT (kW)

Average CO/C02 ratios in the flue gases for 3 different chimney heights as a function of the power output.

A reduction in chimney height results in a reduction of the chimney draft and vice versa. It has also been shown that for each of the chimney heights examined, there is a power range in which the CO /C02 ratios remain below 1

%.

Thus the above results suggest that by introducing a damper in the chimney to vary the draft, it might be possible to obtain a power output range from 3 - 9 kW, yet ha".ing acceptable CO/C02 ratios (

<

13). However new experiments and a mathematical model recently developed show that the power range at a constant chimney height that can be controlled b¥ a damper will be smaller. The results of these experiments and the model will be published later (Kuiper, 1990; Moerman, to be published). Figure 3.13 shows the chimney draft as a function of the power output for different constant chimney heights. It clearly shows that the draft at constant chimney height only marginally increases with increasing power output.

ro Q; f-LL <{ [[ 0

>-w z 2 I () Figure 3.13: Figure 3.14: 8 6 4 2 3 3 -t> 6. 43 cm

•

62 cm 0 100 cm 6 9 POWER OUTPUT (kW)

Chimney draft as a function of power output for different constant chimney heights.

6 9

POWER OUTPUT (kW)

Average excess air factors as a function of the power output for different constant chimney heights

Figure 3.14 shows the excess air factor as a function of the power output for different constant chimney heights. It is obvious from the figure that the excess air factor decreases with the increase of power output at constant chimney height. According to tables III, IV and V, for a constant chimney height, the 02 concentration increases and the C02 concentration decreases

sharply with a reduction of the power output. This results in an increase of excess air at smaller power outputs. This excess air must result in a reduction of the flue gas temperatures. However, the effect of this reduction in temperature on the chimney draft, which partially depends on these temperatures (equation 3.3), is marginal as was shown in figure 3.13. The cooling effect has a much stronger influence on the quality of combustion, which was shown in figure 3.11 and tables III, IV and V.

3.4 Efficiency at Optimum Charging Rates for Different Constant Chimney Heights

The present design is a laboratory model and as such not yet suitable as a cooking device. However, encouraged by the promising results of the present laboratory model, it was felt worthwhile to do some efficiency measurements. For the design as shown in figure 2.1, there are mainly two locations where a pan can be placed. The first one is at the chimney exit and the second one is in between the fire bowl and the chimney entrance. A third possibility would be to place a pan above the fire bowl. However the radiative heat from the fuelbed towards the pan bottom is relatively small. Besides, placing a pan here would make charging of the wood much more difficult. Three types of efficiency measurements were carried out.

(i) A pan was placed at the chimney exit (A).

(ii) A pan was placed in between the fire bowl and the chimney entrance (B). (iii) Two pans were placed on the stove, one at the chimney exit and one in

between the fire bowl and the chimney entrance (C).

These experiments were carried out for three different chimney heights ( 43 cm; 62 cm and 100 cm), applying the optimum charging rate for each chimney height as they were determined in the experiments described in section 3.3 (see tables III, IV and V for the charging rates that resulted in the lowest CO /C02 ratios). The distance between the chimney exit and the

pan bottom was always kept at 25 mm. In that case the chimney outlet area is the same as the flow area left open by this gap. During the experiments, the flue gas compositions and temperatures at the chimney

entrance and exit were monitored.

The efficiency of a stove is defined as the ratio of the net amount of heat absorbed by the water in the pan and the amount of sensible heat supplied by the fuel. The equation for this is:

MwCp(Tb - Ti) + MeL

TJ

=

Mr Br

100

(3.4)

where

TJ = efficiency (3)

M = initial amount of water in the pan (kg) w

Me = mass of water evaporated from the pan during an experiment (kg) Mf = total mass of fuel consumed during an experiment (kg)

Cp = specific heat of water (4.19 kJ/kgK) Tb = boiling temperature of water (OC) T. = initial temperature of the water (OC)

l

L = latent heat of evaporation of water at atmospheric pressure and at

1oooc

(2257 kJ/kg)

Bf = net calorific value of the fuel (for White Fir with 103 moisture on dry basis, it is 16841 kJ /kg)

Before carrying out the efficiency experiments with only a pan at the chimney exit, it is possible to estimate the efficiency that can be expected.

This is done by computing the sensible heat that escapes from the chimney using the following expression.

Rs

-where:

(Vst ..X Pg+ l)(Te - Ta)Cpg

r 100 (3.5)

Hs sensible heat escaped through the chimney (3)

Vst = stoichiometric amount of air to btirn 1 kg of wood (m3/kg)

..X

=

excess air factor

Pg

=

density of the flue gases at ambient temperature (kg/m3) Cpg = specific heat at constant pressure of the flue gases (kJ/kg.K) Te = temperature of the flue gases at the chimney exit (OC). Ta = ambient temperature (OC)

The sensible heat that escapes from the chimney exit, for different constant chimney heights with varying power outputs, was already presented in tables III, IV and V. If the charging rates that result in the lowest CO/C0₂ ratios are considered to be the optimum, and the heat transfer efficiency from the flue gases to the pan is assumed to be 503, the maximum efficiencies that can be expected for a 43 cm, 62 cm and 100 cm chimney with only a pan at the chimney exit, are 233, 203 and 223, respectively. The results also show that a reduction in power output at constant chimney height, generally results in an increase of the sensible heat that escapes from the chimney exit. This increase is primarily due to the increase of the excess air factor.

The results of the efficiency experiments are listed in table VI. The efficiencies for different chimney heights are plotted in figure 3.15.

50 .---~~~~~~--.~~~~~~~--.~~~~~~~~---. - 43 cm

CJ

62 cm 40

l

>-

30

u

z

w

Q 20 LL LL

w

10 0 Figure 3.15:

Im

100 cm A B

c

PAN POSITION

Efficiencies for different chimney heights and pan positions

A: pan at the chimney exit; B: pan in between fire bowl and chimney entrance; C: pans at the chimney exit and in between the fire bowl and the chimney entrance.

The figure clearly shows that with an increase in chimney height, the overall efficiency ( C) increases. However the efficiencies measured with only one pan on the stove (either at the chimney exit or in between the fire bowl and

the chimney entrance), do not show any systematic change with the increase of chimney height. If we study the efficiencies for the individual chimney heights, it is difficult to extract any logic. For the 43 cm and 62 cm chimney, the sum of the efficiencies measured during circumstances A and B is larger than the efficiencies measured during circumstance C. In case of the 100 cm chimney, the sum of the efficiencies measured during A and B is considerably smaller than what is measured during C. Because of these unclear results, further tests on the efficiency of this stove under various circumstances are required.

IilLE YI: llllClllCY !T lnDUI CUIGIJG W'IS n1 II1HUIT CllST.lJT CllDIY BI!ilTS

Vood species: Vll.ite Jir Yolue/suface area (ca): 0.360

Cbrging rate {blocks/sec}: 2/35 for 43Cll ~iaey Charging rate blocks/sec : 2/30 for 62ca ckimle7 Charging rate blocks/sec : 2/25 for lOOCll ckimle7

!ot !oftr 'verage flse f43 eOllJloaitio• Sl. loeatiot1 oat 1st

Ir. (kf)

" I

"2

I

•2

I

"l"2

(roL. 7.) (roL. 7.) (rol.7.) ('!.)

C1i .. e' 1eig1t = 41 e11

1 N 5.1 0.021 10.27 10.1 0.211 2 A 4.5 0.077 14.35 6.45 0.505 3 B 4.3 0.053 12.0 8.6 0.42 4 c 4.0 0.077 10.4 10.0 0.688

Cbi .. er 1eig&t

=

it e•

5 H 5.9 0.009 11.51 8.99 0.078 6 A 5.6 0.021 13.43 7.35 0.139 7 B 5.2 0.033 12.41 8.35 0.265 8 c 5.1 0.035 13.34 7.43 0.25

C1i .. e' &eig1t = 100 e11

9 H 6.7 0.016 9.67 11.04 0.18 10 A 6.45 0.028 14.17 6.7 0.175 11 B 5.8 0.019 10.57 10.1 0.20 12 c 5.03 0.023 12.54 8.26 0.21

•

N No efficiency •easurement A Pot placed at the chianey exit

1-zeeaa ClUnae, tea,eratsre

r•e)

air latruee I

•:it

factor 2.1 709 634 1.52 635 594 1. 7 589 571 2.1 578 554 1. 75 772 658 1.6 710 641 1. 74 639 593 1.61 626 582 2.22 685 587 1.5 740 569

.

2.0 682 554 1. 7 688 610

B Pot placed in between the firebowl and the chi1111ey entrance

1r .. _(!a

1"

3.53 3.41 3.35 3.30 5.18 5.08 4.93 4.89 8.05 8.13 7.97 8.10

C Pots placed at the chi1111ey exit and in between the firebowl and the chi1111ey entrance

lffieieacr ('!.) . 16.0 22.6 32.7 . 19.4 17.4 35.2

-14.3 14.8 40.8

During these efficiency measurements, flue gas compositions, flue gas temperatures and weight losses were also recorded. In the rest of this section, the results of these measurements will be discussed in more detail. To illustrate the general behaviour of the downdraft stove during the efficiency measurements, the chimney entrance and exit temperatures and the CO/C02 ratios recorded with the 62 cm chimney are shown as a function of time. Figures 3.16 and 3.17 show the flue gas temperatures at the chimney entrance and exit as a function of time during efficiency measurements with the 62 cm chimney. w 900 0:: => f-~ 800 0:: w Q_ 2 w f- ₇₀₀ w ,..._ () () Z o ~ 0:: ₆₀₀ f-z w

>-w 500 z 2 I () 400 0 Figure 3.16: 10 20 30 40 50 60 TIME (MIN)

Flue gas temperatures at the chimney entrance as a function of time during efficiency measurements with a 62 cm chimney

a= no pans; V =pan placed at the chimney exit;•= pan placed between fire bowl and the chimney entrance;

+

= pans placed at the chimney exit and in between the fire bowl and the chimney

entrance. ·

Both figures clearly show that the introduciion of a pan reduces the flue gas temperatures, especially when a pan is placed in between the fire bowl and the chimney entrance. The temperatures recorded at the chimney exit show a less fluctuating behaviour than the temperatures at the chimney entrance. This is probably because the temperatures of the flue gases at the chimney entrance are being affected by the presence of flames. Figure 3.18 shows the CO/C0 2 ratios of the flue gases

as

a function of time during the efficiency

900 w 0:: :::J ₈₀₀ I-<{ 0:: w Q_ 2 700 w I- .-..

u

I-0

x

-

.

w 600

>-w z 2 ₅₀₀ I 0 400 0 Figure 3.17: 10 20 30 40 50 60 TIME (MIN)

Flue gas temperatures at the chimney exit as a function of time during efficiency measurements with a 62 cm chimney

a

=

no pans; V

=

pan placed at the chimney exit; • = pan placed between fire bowl and the chimney entrance;+= pans placed at the chimney exit and in between the fire bowl and the chimney entrance.

measurements with a 62 cm chimney. In circumstance N, where no pans were placed on the stove, the CO/C0 2 ratios remain very low (average CO/C0 2 ratio is 0.078 3). By placing a pan at the chimney exit (circumstance A), the combustion quality slightly deteriorates (average CO/C0 2 ratio is 0.139 3). Although there is a reduction in flue gas temperatures, the deterioration in combustion quality will primarily be the result of the introduction of a resistance to flow, which reduces the net draft through the system. Thus less combustion air enters the combustion zone, which has resulted in higher CO/C0 2 ratios. The deterioration of the combustion quality in circumstance B (pan placed in between the fire bowl and the chimney entrance) will be mainly' due to the cooling down of the flue gas temperatures (average CO/C02 ratio is 0.265 3). The average flue gas temperatures at the chimney entrance dropped by about 135oc (see figure 3.16). Circumstance C (both pans are placed on the stove) shows the joint effect of cooling and flow resistance on the combustion quality. The recorded average CO/C02 ratio (0.25 3) is not as large as the sum of the average CO/C0 2 ratios found during circumstances A and B.

4 0 4

l

(\I .0 0 ₄

u

_-... 0

u

0 4 0 0 Figure 3.18: Power output (kW) (C) 5. 1 (8) 5.2 (A) 5.6 (N) 5.9 10 20 30 40 50 60 TIME (MIN)

CO/C02 ratios as a function of time during efficiency

measurements with a 62 cm chimney

N

=

no pans; A

=

pan placed at the chimney exit; B

=

pan placed in between the fire bowl and the chimney entrance; C

=

pans placed at the chimney exit and in between the fire bowl and the chimney entrance

Further results on the efficiency measurements for the 43 cm and 100 cm chimney will be presented in figures showing only the average values. Figure 3.19 shows the average power outputs recorded during the efficiency measurements with the different chimney heights. Although for each chimney height, the corresponding optimum charging rate was applied during circumstances N, A, B and C, ea~h time a reduction in power output was recorded. The biggest reduction in power is found with the 100 cm chimney. Between the situation in which no pans are placed on the stove (N), and when 2 pans are placed on the stove (Cl, the power reduces by about 25

3.

The author is not able to give a solid explanation for this phenomena. Possibly the reduction in flue gas temperatures (will be shown later) and the reduction in chimney draft are in some way responsible for this behaviour.

Figure 3.20 shows the average flue gas temperatures at the chimney entrance and exit during the efficiency measurements for the different chimney heights

8

s

6 ~ f-:J o_ f-:J 4 0

a:

w

s

0 ₂ o_ 0 Figure 3.19:

-

43 cm

D

62 cm ~ 100 cm N A _B

c

PAN POSITION

Averda£ recorded power outputs for different chimney heights and · erent positions of the pan(s).

N: no pans; A: pan placed at the chimney exit; B: pan placed in

between the fire bowl and the chimney entrance; C: pans placed at the chimney exit and in between the fire bowl and the chimney entrance.

In case of the 43 cm and 62 cm chimney, the entrance and exit temperatures show a continuous decrease from situation N to C. It is clear that placing a pan in between the fire bowl and the chimney entrance causes the largest drop in flue gas temperature. In the case of these 2 chimney heights, the figure also shows that the difference between entrance and exit temperature reduces from .situation N to C. This can be explained ·

as follows. When a pan is placed at the chimney exit, the introduction of a resistance reduces the flow velocities of the flue gases. Because of this, the heat transfer coefficient from the flue gases to the chimney wall will be smaller. Thus less heat is transferred from the flue gases to the chimney walls and lost to the surroundings. This results in less reduced flue gas temperatures at the chimney exit. A similar explanation can be given in case a pan is placed in between the fire bowl and the chimney entrance. The extraction of heat by this pan results in a reduction of the flue gas temperatures at the chimney entrance. Thus the chimney temperatures will

800 ~ 700 ~

w

a:: :::J f-<( a:: w Q_

-2 ₆₀₀ w f-500 Figure 3.20:

CHIMNEY TEMPERA TURES

ENTRANCE EXIT

-

43 cm

-D

62 cm

m

100 cm

--

-

-N A B

c

N A B

c

PAN POSITION

Average flue gas temperatures at the chimney entrance and exit during the efficiency measurements with different chimney

heights. ·

N: no pans; A: pan placed at the chimney exit; B: pan placed in between the fire bowl and the chimney entrance; C: pans placed at the chimney exit and in between the fire bowl and the chimney entrance.

be smaller. Therefore the difference in temperature between the chimney walls ·and the surroundings becomes smaller. This in turn results in a reduction of the amount of heat transferred from the walls of the chimney to the surroundings. Because less· heat is lost to the surroundings, the reduction of the flue gas temperatures at the chimney exit will be less. In case of the efficiency measurements with the 100 cm chimney, the temperature behaviour is rather cumbersome and difficult to explain. It is possible that the applied charging rate has been too small, causing this strange behaviour. Still more experiments are required to elucidate this.

The next question to be asked is about the influence of the reduction in power output and temperature behaviour recorded at the chimney entrance

and exit for different chimney heights, on the quality of combustion. Figure 3.21 shows the average CO/C02 ratios during the efficiency measurements with different chimney heights and positions of the pan(s). The CO/C0 2 ratios were chosen to eliminate the dilutive influence of excess air.

0.80

.

0.60

~

N 0 _0.40

u

... 0

u

0.20 0.00 Figure 3.21:

-

43 cm

D

62 cm ~ 100 cm N A

B

c

PAN POSITION

Average CO/C02 ratios during the efficiency measurements with different chimney heights and pan positions.

N: no pans; A: pan placed at the chimney exit; B: pan placed in

between the fire bowl and the chimney entrance; C: pans placed at the chimney exit and in between the fire bowl and the chimney entrance.

The most important aspect of this figure is that independent of the chimney height, the pan position or the number of pans, all CO/C02 ratios recorded are within the acceptable range (0 - 1 3) as per definition of clean combustion. The highest CO/C02 ratios are found in case of the 43 cm chimney. Especially the introduction of a resistance by placing a pan at the chimney exit results in a large increase of. the CO /C02 ratio. In figure 3.20 it was shown that this corresponds to a large reductfon of the flue gas temperatures. This is probably partially responsible. for the increase in CO/C02 ratio. Besides figure 3.19 did show that during the efficiency measurements· with a 43 cm chimney, the power output reduced from 5.1 kW to 4.0 kW. In figure 3.12 it was shown that in this power range, the CO/C02 ratio increases sharply. In case of the 62 cm, the deterioration of

the combustion quality is not as large as with the 43 cm chimney, which can be explained as follows. The power output during the efficiency measurements with the 62 . cm chimney reduced from 5.9 kW to 5.1 kW. According to figure 3.12, these power outputs are all very near to the optimum. Therefore not much change in CO/C02 is to be expected. The

fact that the _{CO/C0 2 ratios do increase with this chimney height will} probably be the result of the reduction in flue gas temperatures, as was shown in figure 3.20. The change in combustion quality during the efficiency measurements with the 100 cm chimney is only marginal. This is somewhat unexpected, since the power output during these experiments reduced from 6.7 kW to 5.03 kW, which according to figure 3.12 should have resulted in a considerable increase of the CO/C02 ratios. The fact that the flue gas

temperatures at the chimney entrance during the efficiency measurements with the 100 cm chimney do not change in the same order as for the 43 cm and 62 cm chimney, is probably the reason that the CO/C02 ratios remain very low.

4. Conclusions

(i) The present stove design not only is very sensitive to operational variables (Khan & Verhaart, 1989) but also to the design variable chimney height.

(ii) There appears to be a relation between chimney height and power output

to obtain clean combustion of wood.

(iii) It is possible to produce clean combustion of wood for extended periods at different optimum chimney heights (43-100 cm). The highest power output in these investigations was obtained with a lOOcm chimney, where the CO/C02 ratio is 0.063.

(iv) The power density of the stove can be as high as 72 W /cm2 (grate diameter = 12 cm), whereas in a classical single mouth stove from Bangladesh, it is 13.5 W /cm2 (grate diameter = 22.5 cm and having a power output of 5.3 kW) (Khan, 1989).

( v) It is demonstrated in this investigation that a domestic cook stove with downdraft principle can be constructed with a relatively short chimney (43 - 50 cm), which can be operated at 4 to 6 kW maintaining clean combustion.

(vi) The efficiency of the stove increases with the increase of chimney height. At the chimney heights of 43, 62,and 100 cm the efficiencies are 32. 73, 35.23 and 40.23 respectively.

(vii) Because the chimney exit temperature is reasonably high (500 - 600DC) and the gases are clean, it can be considered to use these gases for heating purposes like for example heating a bakery oven and other similar purposes.

(viii) It was found during the measurement of different efficiencies, that in general the introduction of a pan reduces the flue gas temperatures both at the chimney entrance and exit. This reduction in temperature is

probably partially responsible for the deterioration of the combustion quality. However all CO/C02 ratios recorded during the efficiency measurements remained well below the maximum limiting value of 1

%.

5. References

Eckert, E.R.G. & Drake, Robert M. Jr. 1972, Analysis of Heat and Mass Transfer, International Student Edition, McGraw-Hill, Kogakusha, LTD., Tokyo, Japan.

Khan, A.M. Hasan R. 1989, Cookstoves in Bangladesh: a case study, Woodburning Stove Group, Eindhoven University of Technology.

Khan, A.M. Hasan R. and Verhaart, P. 1989, Clean Combustion of Wood?, in Woodcombustion Studies, E. Schutte & K. Krishna Prasad (eds.), Woodburning Stove Group, Eindhoven University of Technology.

Krishna Prasad, K., Sangen, E. and Visser, P., 1984. Woodburning Cookstoves, in Advances in Heat Transfer, vol. 17, pp. 159 - 917. Academic Press, New York. Kuiper, S., 1990, Introducing a flow resistance to control the power output in a

downdraft stove. Report R-1042-S. Woodburning Stove Group, Eindhoven University of Technology.

Shafizadh, F. and De Groot, 1976, Thermal uses and properties of carbohydrates and lignins, Academic Press, New York.

Sielcken, M.0., 1983, Notes on gas analysis, in Technical Aspects of Woodburning Cookstoves, K. Krishna Prasad & E. Sangen (eds.), Woodburning Stove Group, Eindhoven University of Technology.

Tillman, D.A., 1978, Wood as an energy resource, Academic Press, New York Verhaart, P., 1981, On designing stoves, in· A Woodstove Compendium, G. De

Lepeleire, K. Krishna Prasad, P. Verhaart and P.Visser (eds.), Woodburning Stove Group, Eindhoven University of Technology

Vermeer, N.J. and Sielcken, M.O., 1983, An Experimental Metal stove, in Technical Aspects of Woodburning cookstoves, K. krishna Prasad & E. Sangen (eds.), Woodburning Stove Group, Eindhoven University of Technology.