Centre of Glass Science and Technology

Bangkok University Thailand

Glass melting furnaces –reflections about

efficiency and design features

Dr.Matthias Lindig, Germany

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 1

 Centre of Glass Science and Technology
Bangkok University Thailand

Content
Glass melting furnaces –reflections about
efficiency and design features
General ideas
Design boundaries (melting rates, temperatures ....)
Design elements - barrier wall, bubbling, boosting
Design elements - burner port, regenerator, throat
Energy consumption - limits
Efficiency comparison
What is the optimum design?
Special furnace design features
Furnace benchmark
Furnace efficiency – final remark

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 2

 Centre of Glass Science and Technology
Bangkok University Thailand

General ideas

Glass furnace heat balance

Heat recovery Glass tank

Wärmetauscher Schmelzwanne
heat input

total energy flow

energy transferred to furnace
Regenerator energy transferred to tank/glass
wall loss heat to glass

heat to stack
heat loss - bottom /
sidewall
heat loss - crown

Acc. to R.Conradt

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 3

 Centre of Glass Science and Technology
Bangkok University Thailand

General ideas
Regenerative furnace; mass and energy flow

in

sf
re
fire

wc
ht Hwc Hwo
off ex

exch
stack wu Hoff Hre
wo

Hin
Hex
Hwu

Hstack

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 4

 Centre of Glass Science and Technology
Bangkok University Thailand

General ideas

Identification of the core figures concerning glass

furnace efficiency improvements

in

ηre regenerator
efficiency
sf
re
fire
ht Hex enthalpy of
wc off glass leaving
ex
furnace
exch
stack wu

wo
Hw total tank
Acc. to R.Conradt wall losses

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 5

 Centre of Glass Science and Technology
Bangkok University Thailand

General ideas

Enthalpy of glass leaving furnace (energy content)

in

re
sf
fire
H ex = ∆H chem ⋅ (1 − λ ) + ∆H T ( gl )
ht
wc off ex H T ( gl ) = c p ( gl ) ⋅ (Tex − 250 C )
exch
stack wu ∆H chem Data can be calculated
according to thermodynamic
wo database

c p (gl ) thermodynamic database

available and by calculation

Acc. to R.Conradt

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 6

 Centre of Glass Science and Technology
Bangkok University Thailand

General ideas

Enthalpy of glass leaving furnace (energy content)

in H ex = ∆H chem ⋅ (1 − λ ) + ∆H T ( gl )
H T ( gl ) = c p ( gl ) ⋅ (Tex − 250 C )

re
sf λ Cullet ratio
fire
ht
wc off ex
Flint container glass:
∆H chem = 138 kWh/t
exch
stack wu ∆H T = 514 kWh/t
mit Tex 1380 °C und c p 0,38 kWh/t.K
wo
∆H ex = 569 kWh/t λ = 0,6

∆H ex Is a physically fixed number, strongly dependent

on the glass quality

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 7

 Centre of Glass Science and Technology
Bangkok University Thailand

General ideas

Energy balance 90 m² end-fired furnace

in

glass exit energy Hex 560 kWh/t

furnace wall loss Hw 250 kWh/t
sf Hre
re heat revocery regenerator 500 kWh/t
fire
ht heat to stack and Hstack, wc 250 kWh/t
wc off wall loss regenerator
ex

exch Net calorific value gas Hu 12 kWh/kg

3
stack waste gas volume / kg fuel Voff 16,7 m /kg
wu 3
batch gases /t of glass Vb 0,08 m /t
3
wo combust air volume Vair 14,7 m /kg
o
combust air temp Tre 1200 C
o
waste gas temp Toff 1400 C

energy input Hin calculated 1060 kWh/t

energy input Hin real tank 1064 kwh/t

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 8

 Centre of Glass Science and Technology
Bangkok University Thailand

General ideas

Energy balance – estimation regarding influence of design features

in
reduction of chemical heat
regenerator design, by using de-carbonized
packing, insulation sf
re materials (CO2 emission
fire
ht only displaced); reduction
wc off ex of process temperature
with special design
exch
features; heat transfer into
stack wu glass; burner design
wo

Insulation, design of combustion chamber, heat transfer

efficiency into glass (heat transfer into glass vs.heat circulation
between combustion chamber and regenerator)

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 9

 Centre of Glass Science and Technology
Bangkok University Thailand

General ideas

Design boundaries
Melting rate
Melting temperature
Raw materials
Cullet addition Design features
Fuel quality regenerator
Burner port, burner, combustion
Wall, boosting, bubbling
Length/width, depth….and more

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 10

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 11

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – theoretical heat demand

Theoretical energy requirement (kcal/kg) relative to cullet content and

water content of batch

Water content of batch

Cullet
content 0% 1% 2% 3%

0% 605 kcal/kg 621 kcal/kg 636 kcal/kg 652 kcal/kg

25 % 553 kcal/kg 567 kcal/kg 582 kcal/kg 598 kcal/kg
50 % 499 kcal/kg 514 kcal/kg 528 kcal/kg 543 kcal/kg
75 % 448 kcal/kg 460kcal/kg 475 kcal/kg 489 kcal/kg

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 12

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – theoretical heat demand

Theoretical energy requirement (kcal/kg) relative to cullet content and

water content of batch

Cullet Water content of batch

content
0% 1% 2% 3%

0% 605 kcal/kg 621 kcal/kg 636 kcal/kg 652 kcal/kg

25 % 553 kcal/kg 567 kcal/kg 582 kcal/kg 598 kcal/kg
50 % 499 kcal/kg 514 kcal/kg 528 kcal/kg 543 kcal/kg
75 % 448 kcal/kg 460kcal/kg 475 kcal/kg 489 kcal/kg

+ 7,7 %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 13

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – theoretical heat demand

Theoretical energy requirement (kcal/kg) relative to cullet content and

water content of batch

Cullet Water content of batch

content
0% 1% 2% 3%

0% 605 kcal/kg 621 kcal/kg 636 kcal/kg 652 kcal/kg

25 % 553 kcal/kg 567 kcal/kg 582 kcal/kg 598 kcal/kg
50 % 499 kcal/kg 514 kcal/kg 528 kcal/kg 543 kcal/kg
75 % 448 kcal/kg 460kcal/kg 475 kcal/kg 489 kcal/kg

- 26,0 %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 14

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – theoretical heat demand

Theoretical energy requirement (kcal/kg) relative to cullet content and

water content of batch

Cullet Water content of batch

content
0% 1% 2% 3%

0% 605 kcal/kg 621 kcal/kg 636 kcal/kg 652 kcal/kg

25 % 553 kcal/kg 567 kcal/kg 582 kcal/kg 598 kcal/kg
50 % 499 kcal/kg 514 kcal/kg 528 kcal/kg 543 kcal/kg
75 % 448 kcal/kg 460kcal/kg 475 kcal/kg 489 kcal/kg

- 31,3 %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 15

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – practical heat demand

Practical heat demand 125m2 furnace

Furnace type regenerative end-fired furnace

Melting area 125 m²
Glass type soda lime
Cullet content 50 %
Melting rate 360 t/24h
Glass temperature at riser 1360 °C
Crown temperature 1580 °C
Waste gas temperature - at the port 1550 °C
- after the regenerators 450 - 500 °C
Air preheat temperature 1280 - 1300 °C
Primary fuel natural gas
Calorific value 8600 kcal/Nm³

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 16

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – practical heat demand

Practical heat demand 125m2 furnace

Cullet Gas CO2 from Specific energy

content quantity combustion Total CO2 consumption

0% 1981 Nm³/h 3936 kg/h 6789 kg/h 1136 kcal/kh

25 % 1826 Nm³/h 3628 kg/h 5770 kg/h 1047 kcal/kh - 7,8 %
50 % 1672 Nm³/h 3323 kg/h 4750 kg/h 959 kcal/kh - 15,6 %
75 % 1521 Nm³/h 3022 kg/h 3735 kg/h 872 kcal/kh - 23,2 %

Batch Gas CO2 from Specific energy

moisture quantity combustion Total CO2 consumption % change

0% 1560 Nm³/h 3100 kg/h 4525 kg/h 895 kcal/kh

1% 1597 Nm³/h 3174 kg/h 4600 kg/h 915 kcal/kh + 2,2 %
2% 1635 Nm³/h 3249 kg/h 4677 kg/h 938 kcal/kh + 4,8 %
3% 1672 Nm³/h 3323 kg/h 4750 kg/h 959 kcal/kh + 7,2 %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 17

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – practical heat demand

Practical heat demand 33m2 furnace

Furnace type regenerative end-fired furnace

Melting area 33 m²
Glass type soda lime
Cullet content 50 %
Melting rate 90 t/24h
Glass temperature at riser 1360 °C
Crown temperature 1580 °C
Waste gas temperature - at the port 1550 °C
- after the regenerators 450 - 500 °C
Air preheat temperature 1280 - 1300 °C
Primary fuel natural gas
Calorific value 8600 kcal/Nm³

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 18

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – practical heat demand

Energy consumption 33m2 furnace

Cullet Gas CO2 from Specific energy

content quantity combustion Total CO2 consumption

0% 568 Nm³/h 1129 kg/h 1843 kg/h 1303 kcal/kh

25 % 528 Nm³/h 1050 kg/h 1586 kg/h 1212 kcal/kh - 6,9 %
50 % 489 Nm³/h 971 kg/h 1328 kg/h 1122 kcal/kh - 13,9 %
75 % 450 Nm³/h 894 kg/h 1072 kg/h 1032 kcal/kh - 20,8 %

Batch Gas CO2 from Specific energy

moisture quantity combustion Total CO2 consumption % change

0% 461 Nm³/h 916 kg/h 1273 kg/h 1057 kcal/kh

1% 470 Nm³/h 935 kg/h 1291 kg/h 1079 kcal/kh + 2,1 %
2% 479 Nm³/h 953 kg/h 1310 kg/h 1100 kcal/kh + 4,1 %
3% 489 Nm³/h 971 kg/h 1328 kg/h 1122 kcal/kh + 6,1 %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 19

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – practical heat demand

10 % cullet ≈ 2,5 - 3 % pull or energy

Crown temperature 1590 °C

Flint
Specific melting rate t/m²/24h

Amber
UVAG
Green

0 10 20 30 40 50 60 70 80 90 100
Cullet in %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 20

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – practical heat demand

Practical experience in accordance with theoretical approach

10 % cullet ≈ 2,5 - 3 % pull or energy

Moisture, also being trapped in by wet cullets do have a similar impact

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 21

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – Insulation

Due to steal work, bracing

cooling areas peep holes and
other no complete insulation
possible

Insulation

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 22

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – Insulation

Due to steal work, bracing

cooling areas peep holes and
other no complete insulation
possible

Insulation

Uninsulated
surfaces

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 23

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – Insulation

Furnace 125 m²

Losses Proportion Area

Total wall losses 2538 kW 100 678 m²

Radiation 759 kW 30
Insulated surfaces 890 kW 35 610 m²
Uninsulated surfaces 889 kW 35 68 m²

Furnace 33 m²

Losses Proportion Area

Total wall losses 1155 kW 100 290 m²

Radiation 260 kW 23
Insulated surfaces 419 kW 36 255 m²
Uninsulated surfaces 476 kW 41 35 m²

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 24

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – Insulation

Total wall heat loss improvements for container glass furnaces

crown
10862 superstructure
tank side wall
1600 bottom
10000
6952
heat flow in W/m²

3080
8000 6456
1860
1441
6000
1081
4412 2150
4000
2770
1621
2000
1770
1321 1164

0
1960 1980 2000

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 25

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – Insulation

Container furnace bottom design

AZS 32
1200 ZS
AZS 32
Ersol 06
1000 120
0
5 Ersol 05
80 75
5
75 ZS
800 50 zircon mullite
300 75
sillimanite 60
mm 600 75
5 200 fireclay 40
fireclay 30
400 300 300
200 insulation

200 200 200 200

0
1960 1980 2000
2 2 2
1770 W/m 1321 W/m 1164 W/m

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 26

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – Insulation

In a large furnace the not insulated surface accounts for about 10 % of the total
area, but produces about 35 % of the total losses.

In a smaller furnace the not insulated area increases to 12 % of the total,

which produces over 40 % of the losses.

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 27

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – furnace size

Example 100 m² end fired regenerative furnace 300 t/d
Heat losses approx. 2200 kW (radiation, cooling, walls)
Part of spec. energy consumption: 151 kcal/kg

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 28

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – furnace size

Specific heat losses related to furnace size

300

Spec. wall losses [kcal/kg]

Spec. wall losses kcal/kg

250

200

150

100
25 50 75 100 125
Melting area m²

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 29

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – furnace size

Furnace size depends on commercial considerations

long term expectation of market development
flexibility requested regarding colour and item
Investment costs

In general even though it might be reasonable to build a larger furnace with

more lines it ends up also in restriction

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 30

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – operation

Normal Exceptional

Crown temperature 1570 °C 1600 °C

Waste gas temperature at the port 1540 °C 1570 °C
Glass temperature at the riser 1360 °C 1390 °C

Tcrown
Twaste gas

Triser

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 31

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – operation

Furnace 125 m²

“Normal“ “Exceptional“

Gas flow 1672 Nm³/h 1725 Nm³/h

Specific energy consumption 959 kcal/kg 989 kcal/kg
Percentage increase in specific
energy consumption 3,13 %

Furnace B - 33 m²

“Normal“ “Exceptional“

Gas flow 489 Nm³/h 505 Nm³/h

Specific energy consumption 1122 kcal/kg 1159 kcal/kg
Percentage increase in specific
energy consumption 3,30 %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 32

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – operation

The impact on operation pattern on efficiency will be discussed again later on.
In general one has to monitor carefully the relationship between process data
and quality data in order to figure out the necessary minimum energy input

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 33

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Glass colour

Expected fossil melting rates

Crown temperature 1590 °C
Cullet content 50 %
Normal container glass quality

Specific melting rate

Flint approx. 2,90 t/m²/24h

Amber approx. 2,65 t/m²/24h

UVAG approx. 2,55 t/m²/24h

Green approx. 2,45 t/m²/24h

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 34

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries – glass colour

Recommended glass compositions for container

SiO2 R2O CaO MgO Al2O3 Fe2O3 Cr2O3 SO3

flint 71.8% 14.0% 10.0% 2.1% 1.8% < 0.07% < 0.001% 0.20%

green 71.6% 14.1% 10.3% 2.2% 1.8% < 0.5% 0.15–0.3% 0.10%

amber 71.5% 13.8% 10.1% 2.4% 1.8% < 0.25% < 0.03% 0.05%

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 35

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Glass colour

Fe2+/Fetotal
90 SO3 80- solubility
70 as60function
50 of 40
redox30number
20 10
0,25
amber oliv green flint

0,2
wt% S as SO3

0,15

0,1

0,05

0
-32 -27 -22 -17 -12 -7 -2 3 8 13 18
Redox number

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 36

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Glass colour

Glass colour has a significant impact on heat transfer and furnace efficiency

Effective heat conductivity of a green and flint container glass

180

160 green
flint
140

120

100
W/mK

80

60

40

20

0
900 1000 1100 1200 1300 1400 1500 1600
T/°C

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 37

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Glass colour

comparison glass effective heat conductivity GS TNO

30,00
GS green
25,00 TNO amber Spessart
TNO green
TNO emerald
20,00
TNO brown
TNO olive
W/mK

15,00

10,00

5,00

0,00
800 1000 1200 1400 1600 1800
T/K

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 38

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Glass composition

Fining with sodium sulphate Na2SO4

Typical 1,0 to 1,3 kg Na2SO4/100 kg sand in combination

with approx. 1/16 of carbon (from sulphate weight)

Example 300 t/d glass Amount sodium sulphate per day 1kg/100kg

10 % cullet 1990 kg Na2SO4 => 900 m³ SO3 at 1000 °C 1)

1550 kg Na2SO4 => 690 m³ SO3 at 1000 °C 1)

30 % cullet

50 % cullet 1110 kg Na2SO4 => 495 m³ SO3 at 1000 °C 1)

70 % cullet 665 kg Na2SO4 => 300 m³ SO3 at 1000 °C 1)

90 % cullet 220 kg Na2SO4 => 110 m³ SO3 at 1000 °C 1)

1) 50 % release, SO3 from cullet not considered

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 39

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Glass composition- melting behavior

Possible sources of foam (no claim of completeness)

Fining Na2SO4 high, ratio sulphate/carbon low

Raw materials Variation in redox state (esp. Sand, cullet)

Raw materials - lime, dolomite High amount of fines

Raw materials - sand To fine sand

Cullet High amount of fines

Al2O3 content glass Al2O3 ⇑ => tendency of foaming ⇑

Bubbling Usually more foam

Furnace atmosphere High excess O2 => more foam

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 40

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Glass composition- melting behavior

Fining agent addition and cullet addition corresponds with each other. The
degasing is obviously easier with larger amount of cullets.

The fines in cullets can also be a significant disadvantage since the fines trap in
large amount of air. Seeds problems and scum can be caused by cullet fines

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 41

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Fossil fuel

Differences in heat transfer between gas and oil

Radiation of flames
120
Oil burner
heat transfer KW/m²

100 Gas burner

80

60

40

20

0
0 0,5 1 1,5 2 2,5 3 3,5 4
distance from burner

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 42

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Fossil fuel

emission coefficient of different kind of fuel

1
Sasol gas
0,9
North sea gas
emissivity coffeficient

0,8 HF oil
0,7 Coal tar fuel
0,6
0,5
0,4
0,3
0,2
0,1
0
0 2 4 6 8 10 12 14
C/H ratio

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 43

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Fossil fuel

gas max. temperature in the refining end

problems with melting of the batch

oil high energy transfer in the melting area

high heat transfer to batch > good melting of batch

experience oil enables slightly higher specific pulls

with oil lower specific energy consumption

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 44

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Fossil fuel- gas and oil firing in math.model

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 45

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - NOx guidelines

Increased superstructure height - recirculation

Higher superstructure temperatures => more NOx

Lower air and gas velocities - mixing delayed, longer flames

Number of burners - asymmetric firing

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 46

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - NOx guidelines

NO development depending on superstructure temperature

λ = 1,1
λ = 0,95

Combustion gas temperature in °C

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 47

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - NOx guidelines

NOx emission vs. crown temperature (measurements)

10,00

9,00
Furnace A
8,00
Furnace B
7,00 Furnace C
Furnace D
6,00
NOx in kg/h

5,00

4,00

3,00

2,00

1,00

0,00
1450 1500 1550 1600 1650 1700
T crown in °C

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 48

 Centre of Glass Science and Technology
Bangkok University Thailand

Design boundaries - Batch charging

Big batch tiles

bad melting behaviour
„diving“ batch tiles seeds
unstable temperatures crown+glass
lower specific pull
foam

Small batch tiles

good melting behaviour

no batch after bubbler/barrier wall
constant temperatures crown+glass
higher specific pull
less foam

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 49

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 50

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – barrier wall and deep refiner

Definition of deep Refiner length a and depth

L
ex

b
ey
c

B
sy

sx V5
V4

a
l b
g
f V2
c
V-LZ h
V1
V3

sy

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 51

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – barrier wall and deep refiner

Barrier wall height controls back flow - energy transport
Barrier wall avoids short circuit on the bottom
Stronger back flow equates higher residence time and better glass quality
Higher Barrier wall results in stronger, hotter forward flow and less back
current

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 52

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – barrier wall and deep refiner

Flow distribution in x-direction

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 53

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – bubbling

Bubbling for more stable current conditions and back flow

80 to 400l/h

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 54

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting

1. Melter Booster side electrodes

2. Barrier booster bottom electrodes
3. Melter booster bottom electrodes

1 2 3

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 55

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting

Bottom electrodes- vertical

Sidewall electrodes - horizontal

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 56

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting

High current density close to electrode

tip

Higher corrosion rate (electrode and

soldier block)

Tendency of energy concentration closer

to the hot melter surface

Lower efficiency

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 57

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting

More even current density between electrodes

Lower electrode corrosion rate

Lower risk for electrode holder and refractory

Higher efficiency

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 58

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

In all cases the same total energy input was given. Regarding melter boster
the bottom installation gives more even energy input compared to side wall
installation

barrier boost melter boost I melter boost II

(energy density in kW/m3)

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 59

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Top view to the barrier booster case and the melter booster
Since the barrier booster has a much stronger impact on the natural
current pattern of the furnace the energy input by the electrodes is more
evenly distributed

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 60

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Base case Barrier boost

Back flow Back flow

Low back current, low forward current speed Strong influence of the back current. Forward
current on higher temperature level.

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 61

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

In all cases the boosting improves the temperature regime in the charge end.
There are differences in the residence time

barrier boost

melter boost side

melter boost bottom

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 62

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Without
boosting

Melter boost
bottom

Barrier boost

0
back flow - Forward flow +

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 63

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Furnace data

melter surface m2 106

melter length m 15,25
melter width m 7,45
glass depth melting area m 1,25
glass depth deep refiner m 2,4
length deep refiner m 1,5
no wall
two doghouses and charger
amber glass
cullet % 50

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 64

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Calculation of three different booster configurations

melter booster barrier booster

electrodes power/kVA electrodes power/kVA kVA total
case 1 6 950 12 2250 3200
case 2 9 2250 6 950 3200
case 3 12 2250 6 400 2650

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 65

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

All models Joulean heat (kW/m3), electric power supply in model with three
phase transformer

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 66

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Cross section all three models- strong barrier boosting in case 1 achieves
highest glass temperature

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 67

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Barrier boosting gives strong support to back current

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 68

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

All models flow speed in x-direction
Bottom max temperature 1390°C, throat glass temperature 1 360°C

0 flow

- +

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 69

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 70

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

The melter booster helps to raise the bottom glass and supports the back current.

6 to 9 melter booster electrodes will be sufficient. 12 electrodes melter

booster does not give any advantage

The barrier boosting will be essential for conveying hot glass from hot spot
back to the charge end.

Strong barrier boosting (12 electrodes, case 1) are not recommended due to
increased risk of severe corrosion. The residence time is less compared to
case 2 and 3. The strong boosting is also pushing the glass forward

Too low energy input for the barrier boosting results in bottom glass
kreeping forward – not being affected by the bosting.

Case 2 with moderate power input for the barrier boosting seems to be the
most convenient solution.

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 71

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

melting area m2 115

throughput t/d 345
cullet ratio % 85
batch moisture % 3
glass olive

energy input m3/h 1478

burner nozzle 1 m3/h 518
burner nozzle 2 m3/h 518
burrner nozzle 3 (inner) m3/h 442

melter boosting (side) kW 3* 350

bubbling per nozzle l/h 280

Net heat value MJ/m3 35,5

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 72

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Furnace modeling with different melter boster (side and bottom

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 73

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

The furnace was calculated with bottom and with side melter boosting
The electric energy input was keep the same
The same combustion model was used for both cases

3D cut with batch layer distribution – no significant difference

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 74
 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

The chart shows the flow speed in x-direction (given in mm/s)

With bottom boosting the back and forward flow is increased

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 75

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

Temperature field and trajectories for booth cases
There is no visible difference in temperature field and flow pattern
The back current towards the charge end is dominating and governed by
the strong bubbling

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 76

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting examples

The bottom boosting and side boosting results in similar temperature field and flow
distribution.
The bottom boosting enforces the total flow pattern.
The residence time distribution for both glass models is comparable.
The bottom boosting might have an advantage for higher pull conditions.

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 77

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting/bubbling comparison

The current in case of bubbling is much stronger. The risk of getting high
corrosion rate near side wall and bottom is quite high.
Bubbling is energy consuming. Bubbling accelerates the forward flow towards the
throat.

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 78

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting/bubbling comparison

Boosting Bubbling

Increase pull Control of bottom temperature

Specific energy consumption ⇓ Min. residence time ⇓

Specific energy consumption ⇑

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 79

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting - assessment

End-fired furnace with Deep Refiner®, 125 m²

heavy fuel oil, 64 % cullet, flint glass

Max. melting Max. specific Specific energy Boosting Boosting

capacity melting capacity consumption kW %

340 t/24h 2,72 t/m²/24h 1,02 kWh/kg 0 kW 0%

360 t/24h 2,88 t/m²/24h 0,97 kWh/kg 165 kW 1,1 %
385 t/24h 3,08 t/m²/24h 0,97 kWh/kg 540 kW 3,4 %
410 t/24h 3,28 t/m²/24h 0,95 kWh/kg 970 kW 6%
(+20% pull increase)

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 80

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting - assessment

End-fired End-fired/boost
2
Melting area (m ) 125 125
Pull (t/d) 350 385
Specific pull (t/m².d) 2,8 3,08
Heating oil-air oil-air
Energy (kJ/kg) 3520 3400
Electric boosting (kW) 540

Energy - fossil kJ/kg 3520 3280

Energy electrcal power kJ/kg 120
Total energy (kJ/kg) 3520 3400

CO2 from tank (kg/t glass) 201 187

CO2 from electrical power (kg/t glass) 20
CO2 total (kg/t glass) 201 207

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 81

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting - assessment

1600

1400

1200
spec.energy in kWh/t

1000

800

600

400
other endport
200 Sorg endport

0
1,5 2,0 2,5 3,0 3,5 4,0 4,5
2
spec.pull in t/m d
Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 82
 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting - assessment

12,0

10,0

8,0
% el.boosting

6,0

low cullets
4,0
high cullets
Sorg low cullets
2,0 Sorg high cullets

0,0
0,0 1,0 2,0 3,0 4,0 5,0
2
spec.pull in t/m d

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 83

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – boosting - assessment

melting energy in
spec.pull t/dm2d
melting area m²

spez. Pull t/m2d

el. power in kW
spec.energy in
total energy in
kcal/kg glass
max pull. t/d

kcal/kg only
total energy

kWh/t glass
only fossil
Furnace

cullet %

heating
fossil

color
1 100 305 3,05 2,47 810 879 1470 62 25,2 Gas green
2 80,5 228 2,83 2,35 820 895 850 92 22 Gas green
3 100 340 3,40 2,76 830 912 1500 80 23,3 Gas green
4 80,5 210 2,61 2,21 860 932 700 92 22 Gas green
5 82 235 2,87 2,50 870 927 700 80 23,3 Gas green
6 108 295 2,73 2,30 880 952 1150 70 24,4 Gas green
7 115 365 3,17 2,61 912 992 1733 50 26,5 Gas green
8 82 220 2,68 2,31 932 1001 745 70 24,4 Gas green
9 91 247 2,71 2,47 960 1004 550 65 24,9 Gas UVAG
10 115 300 2,61 2,42 960 992 586 45 27 Öl green
11 97 284 2,93 2,76 961 989 382 73 24 Gas dark green
12 97 296 3,05 2,76 967 1016 713 65 24,9 Gas dark green
13 97 271 2,79 2,54 970 1017 625 65 24,9 Gas dark green
14 90 230 2,56 2,56 975 975 0 80 23,3 Gas green
15 108 246 2,28 2,28 981 981 0 50 26,5 Gas green
16 82 220 2,68 2,68 1000 1000 0 30 28,6 Öl green
17 82 215 2,62 2,62 1020 1020 0 45 27 Öl green
18 115 310 2,70 2,70 1046 1046 0 50 26,5 Gas green
19 90 313 3,48 2,83 1105 1226 1650 35 28,1 Gas UVAG

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 84

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – burner port

Burner angle a1, a2 between

15°to 20°depending on
furnace size
a1
B port width H4
Height H3 has strong influence
on gas/air mixing conditions
a4 a2
H3
H2
H1
a5 burner angle
glass level

L1

L2 a1

L3 length up to end of basin

L4 mength up to end of basin

a4
a3
a2 H2

a5 burner angle

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 85

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – burner port

Different port height- flow speed in x-direction

V= 8m/s

V= 13m/s

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 86

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – burner port

Different port height- temperature distribution

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 87

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – regenerator

Regenerator with one or two or three chambers

More than 1 chamber for

T1top K1
higher efficiency
L1
In case that the height is limited
L2
K2
T2top
With more than 1 chamber
h=ca.1m
Risk of plugging

H1
Corrosion the area of the rider
V1
H2 arches
V2
Higher investment costs

G1 T1down 850-900 °C T2down

max. 950 °C G2

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 88

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – regenerator

Regenerator with one or two or three chambers

Endport 112m2 melting area

T1top K1
Height of first chamber 8m
L1 Cross section in m2
L2 Chamber 1 27m2 cruciform
K2
h=ca.1m T2top Chamber 2 17,5m2 chimney
Chamber 3 12,7m2 bricks
V1 H1 Constant flow speed
H2
V2
Careful control of
condensate deposit
required

G1 T1down 850-900 °C T2down

max. 950 °C G2

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 89

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – regenerator

Regenerator general design thump rules

Length / width 1,3 to 1,5

T1top
K1
Average flow speed 0,3m3/s
L1 Characteristics volume number:
L2 height
K2
T2top
h=ca.1m length ⋅ width

about 2,5
V1 H1
H2
V2 Checker volume :
~1m3/1t of glass per day
Checker heat load :
<100.000 kcal/m2d
G1 T1down 850-900 °C T2down
max. 950 °C G2

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 90

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – superstructure

20kcal/s m3

L B

T
h v x

l b

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 91

 Centre of Glass Science and Technology
Bangkok University Thailand

Design features – throat

Pull capacity

B ~ 12 t/dm2d

examples
H
length
cooling

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 92

 Centre of Glass Science and Technology
Bangkok University Thailand

Energy consumption limits

Sankey diagram example endport furnace 125m2

Heat 6,1 %
electrical energy
recovery Heat input
51,8 % 100 %

waste gas
75,4 % Chemical
10,0 % heat
evaporation
2,2 % enthalpy
Heat to 8,9 + 3,3 % 16,7 % furnace wall
stack 35,8 % induced air, losses
batch gases
& moisture 41,4 %
wall losses regenerator
heat to glass

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 93

 Centre of Glass Science and Technology
Bangkok University Thailand

Sankey diagram example endport furnace 125m2

Energy consumption limits
About 35% heat to stack
available for further heat 6,1 %
Heat
electrical energy
recovery Heat input
recovery
51,8 % 100 %

Batch preheat waste gas

75,4 % Chemical
10,0 % heat
El.power generation with heat
evaporation
exchanger 2,2 % enthalpy
Heat to 8,9 + 3,3 % 16,7 % furnace wall
stack 35,8 % induced air, losses
batch gases
& moisture 41,4 %
wall losses regenerator
heat to glass

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 94

 Centre of Glass Science and Technology
Bangkok University Thailand

Energy consumption - limits

Batch preheat
Batch preheat requires
constant waste gas
temperature. Batch and cullet
preheat applied for
recuperative furnaces or
gas/oxy fired furnaces
High investment costs
Only meaningful for larger
furnaces (>300t/d)
Cullet or batch and cullet
preheat up to 300°C

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 95

 Centre of Glass Science and Technology
Bangkok University Thailand

Energy consumption - limits

Batch and cullet preheat

Batch and cullet flow

affected by direct
contact with high water
containing waste gas

Soda ash reacts with

water between 35 -
106°C to form
Monohydrate. The
hydrate decomposes
above 106°C and reacts
with the sand grains
(physical bonding)

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 96

 Centre of Glass Science and Technology
Bangkok University Thailand

Energy consumption - limits

(shut down)
Budenheim W3 Wiegand W5 Wiegand W4
Batch and cullet preheat LoNOx LoNOx LoNOx
total furnace pull t/d 380,00 230,00 350,00
cullet ratio % 80 90 85
cullet bulk density kg / dm³ 1,40 1,40 1,40
quantity cullets t/d 304,00 207,00 297,50
quantity cullets t/h 12,67 8,63 12,40
fuel heavy oil natural gas natural gas
lower heat value kcal / unit 9600 8566 8566
fuel consumption unit / h 1500 900 1450
enregy consumption (fuel only) kcal / kg 909 804 852
waste gas volume per unit fuel (unit) 12,40 11,20 11,20
waste gas volume Nm³ / h 18600 10080 16240
excess oxygen waste gas % 1,00 1,00 1,00
oxygen at preheater entrance % 6,00 4,00 4,00
waste gas volume before preheater Nm³ / h 24821 11524 19114
number of cullet columes 4 2 4
colume width m 3,00 5,40 3,00
colume height m 5,70 6,00 6,20
colume depth m 0,42 0,36 0,42
cullet in preheater t 40 33 44
heat exchage area m² 68,40 64,80 74,40
waste gas/cullet Nm³ / kg 1,47 1,17 1,31
waste gas before preheater/ cullet Nm³ / kg 1,96 1,34 1,54
residence time in preheater h 3,18 3,79 3,53
waste gas/time and exchage area Nm³ / (h * m²) 272 156 218
including excess air Nm³ / (h * m²) 363 178 257
cullet temperature behind preheater °C 280
Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 97
 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark
Cullet preheat energy balance pull t/d 380
energy heavy oil l/h 1500
kW 16750
Heat loss due to preheater cullet % 80
radiation wall losses and cullet t/h 12,7
ambient air sucked into waste cp cullet kWh/T K 0,27
gas stream water content % 2
Q water evap kWh/kg 0,62
T IG cullet in °C 20
T AG cullet out °C 280

waste gas Nm3/h 18600

cp waste gas Wh/m3K 0,42
TIA waste gas inlet °C 530
TAE waste gas out °C 380

delta Q cullet kW 1046

delta Q waste gas kW 1172

efficiency % 89
savings % 6,2

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 98

 Centre of Glass Science and Technology
Bangkok University Thailand

Energy consumption - limits

Batch and cullet preheat no

preheat preheat
melting area m2 115 115
pull t/d 345 345
spec.melting rate t/m2d 3,00 3,00
fuel consumption nat.gas Nm3/h 1478 1453,5
el.power kW 1200 350
spec.energy consumption kW/t 1091 1015
air preheat °C 1200 1200
cullet % 90 90

cp cullet kWh/tK 0,24 0,24

cullet preheat temperature °C 20 350
energy savings cullet preheat kWh/t 0 84

total savings kW 0 1087

energy savings % 0 7,0

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 99

 Centre of Glass Science and Technology
Bangkok University Thailand

Energy consumption - limits

Batch and cullet preheat

no
preheat preheat
melting area m2 115 115
pull t/d 345 345 Instead of energy reduction
spec.melting rate t/m2d 3,00 3,00 Pull increase with no
fuel consumption nat.gas Nm3/h 1478 1453,5 temperature increase !
el.power kW 1200 350
spec.energy consumption kW/t 1091 1015
Means no increased wall
air preheat °C 1200 1200
heat loss and waste gas
heat loss
cullet % 90 90
Overall energy savings
cp cullet kWh/tK 0,24 0,24 ending up lower than 7%!
cullet preheat temperature °C 20 350
energy savings cullet preheat kWh/t 0 84 7% savings equates about
20t/d
total savings kW 0 1087
energy savings % 0 7,0

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 100

 Centre of Glass Science and Technology
Bangkok University Thailand

Energy consumption - limits

Batch and cullet preheat

energy input and efficicncy in %

a. Series connection batch preheat and air

preheat
b. parallel connection batch preheat and air

temperature in °C
preheat

1. Energy consumption in % 100% pull

2. Energy consumption in % 120% pull

3. Efficiency 100% pull

4. Efficiency 120% pull

5. Batch preheat in °C 100% pull

6. Batch preheat in °C 120% pull

preheated batch %

Barklage-Hilgefort, Trier GlastechnBer 56(1983)10, 269-279

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 101

 Centre of Glass Science and Technology
Bangkok University Thailand

Energy consumption - limits

BOC proposal for combination of
batch prehat and dust removal

Transverse Tubes

Gas Flow Channel

Batch Surface

Hopper Wall

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 102

 Centre of Glass Science and Technology
Bangkok University Thailand

Optimum design
Length to width 1,3 to 1,65
Higher length gives more residence time for glass and combustion gases
Higher efficiency in quality and energy transfer
Lower glass exit temperature

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 103

 Centre of Glass Science and Technology
Bangkok University Thailand

Optimum design
Larger combustion chamber
Efficient heat transfer near the batch layer
Reduction of glass exit temperature (submerged throat, avoid short circuit)

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 104

 Centre of Glass Science and Technology
Bangkok University Thailand

Special furnace design

1 or two throat design- an option for large pull and more than 3 lines
With two throat specific -reduction of throughput- less corrosion

340t/d
122m2 – 2,78t/m2d
950kW boosting
16 bubbling nozzles

100t/d

60t/d

80t/d

100t/d

100t/d

60t/d

80t/d
100t/d

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 105

 Centre of Glass Science and Technology
Bangkok University Thailand

Special furnace design

Furnace temperature profile in both cases comparable

1330°C

1325°C 1335°C

1314°C
One throat

1327°C

1326°C 1337°C

1306°C
Two throat

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 106

 Centre of Glass Science and Technology
Bangkok University Thailand

Special furnace design

1 or two throat

One throat- less

critical feeder
entrance temperature

Two throat- short

distance from throat
to feeder entrance-
about 50°C entrance
temperature

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 107

 Centre of Glass Science and Technology
Bangkok University Thailand

Special furnace design

One and two throat- fastest particle trace

Min.residence time about 5,5h Min.residence time about 4,9h

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 108

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark

Normalization of energy carriers

electric power 1 kWh equates 9 MJ efficiency 40%

Oxygen 1 m3 equates 4 MJ 1,1kWh for generation

Normalization to
100% energy input by fuel
50% cullet addition
0 years furnace campaign
oxy-fuel correction (-6% on total energy consumption )

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 109

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark

E E50 primary energy norm. to 50%c in MJ

Cullet normalization E50 = E energy at actual cullet fraction in MJ
− 0,0034 ⋅ S + 1,17
S cullet fraction in %
Normalization based on 100 furnaces
statistics; 10% cullet ratio increase
equates about 2,5-3% energy savings

E0,50 E50 normalized to 0 years in MJ

E50
Age correction E0,50 = E50 primary energy norm. to 50%c in MJ
1,0135n
Annual energy consumption increase 1,35%
n campaign in years

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 110

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark
Actual furnace data (end-fired furnace, 117 m²) wall losses increase
3,6% increase of total energy consumption per year (in that individual case)

4300

4100

3900

3700

Total heat 3500

losses / kW
3300

3100

2900

2700

2500
0 5 10 15 20 25
Campaign - months

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 111

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark

90 m² container furnace
energy consumption vs. specific pull
1500

spec.energy in kWh/t glass

1400

1300

1200

1100

1000

900
0,5 1 1,5 2 2,5 3

specific pull in t/m².d

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 112

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark

Specific energy consumption vs specific load for different

tank designs and melting areas from 20 - 180 m²

4000
Higher impact of regenerator Risk of poor melting
efficiency and wall losses performance
specific energy consumption in

3500

3000
kcal/kg glass

2500

2000

1500

1000

500

0
0,00 0,50 1,00 1,50 2,00 2,50 3,00 3,50 4,00

specific pull in t/m².d

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 113
 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark

70-80 m² container furnaces with various heating systems

End-fired
End-fired + boost
3,00
Specific energy in kWh/kg glass

Unit melter
Unit melter + booster
2,50 Cross fired
Cross fired+ boost
2,00

1,50 Most of the

furnaces with
1,00 the best
efficiency are
0,50 end-fired with
booster
0,00
0,00 1,00 2,00 3,00 4,00

Specific pull in t/m².d

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 114

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark

Furnace benchmark - adding el.power boosting direct to fuel and

normalized

1,5
246t/d
spec.energy consumption in kWh/kg

296t/d
296t/d norm
1,25 295t/d
295t/d norm
340t/d
glass.

1 340t/d norm

0,75

0,5
0,0 2,5 5,0 7,5 10,0 12,5 15,0
addition of el.power to total energy input in %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 115

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark Furnace benchmark - adding el.power boosting direct to fuel and
normalized

1,5
246t/d

spec.energy consumption in kWh/kg

296t/d
296t/d norm
1,25 295t/d
295t/d norm
340t/d

glass.
1 340t/d norm

0,75

0,5
0,0 2,5 5,0 7,5 10,0 12,5 15,0
addition of el.power to total energy input in %

In case of cullet normalization from 25% to 50% fuel and boosting power
reduced
In real furnace operation only boosting is reduced
Furnace performance in benchmark calculation ends up being
comparable poor

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 116

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace benchmark
Energy cost consideration and CO2 emission fee

Endport furnaces
heating with natural gas
cullet 50% green
glass

melting area pull spec.pull total energy Boosting fossile fuel total cost
m2 t/d t/m2d kWh/kg kW % kWh/kg Cent/kg

108 246 2,28 1,14 0 0 1,14 2,63

97 296 3,05 1,13 713 5,1 1,07 2,80
108 295 2,73 1,02 1150 9,1 0,93 2,70
100 340 3,40 0,97 1500 11,0 0,86 2,61

Natural gas 0,023 €/kWh

el.power 0,06 €/kWh
(ca. 0,006€/kWh
CO2 0,035 €/kg Gas)

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 117

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace efficiency

melting energy in
spec.pull t/dm2d
melting area m²

spez. Pull t/m2d

el. power in kW
spec.energy in
total energy in
kcal/kg glass
max pull. t/d

kcal/kg only
total energy

kWh/t glass
only fossil
Furnace

cullet %

heating
fossil

color
1 100 305 3,05 2,47 810 879 1470 62 25,2 Gas green
2 80,5 228 2,83 2,35 820 895 850 92 22 Gas green
3 100 340 3,40 2,76 830 912 1500 80 23,3 Gas green
4 80,5 210 2,61 2,21 860 932 700 92 22 Gas green
5 82 235 2,87 2,50 870 927 700 80 23,3 Gas green
6 108 295 2,73 2,30 880 952 1150 70 24,4 Gas green
7 115 365 3,17 2,61 912 992 1733 50 26,5 Gas green
8 82 220 2,68 2,31 932 1001 745 70 24,4 Gas green
9 91 247 2,71 2,47 960 1004 550 65 24,9 Gas UVAG
10 115 300 2,61 2,42 960 992 586 45 27 Öl green
11 97 284 2,93 2,76 961 989 382 73 24 Gas dark green
12 97 296 3,05 2,76 967 1016 713 65 24,9 Gas dark green
13 97 271 2,79 2,54 970 1017 625 65 24,9 Gas dark green
14 90 230 2,56 2,56 975 975 0 80 23,3 Gas green
15 108 246 2,28 2,28 981 981 0 50 26,5 Gas green
16 82 220 2,68 2,68 1000 1000 0 30 28,6 Öl green
17 82 215 2,62 2,62 1020 1020 0 45 27 Öl green
18 115 310 2,70 2,70 1046 1046 0 50 26,5 Gas green
19 90 313 3,48 2,83 1105 1226 1650 35 28,1 Gas UVAG

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 118

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace efficiency

Endport furnaces for container glass with different specific pull (pull referring to
fossil energy input)

4,00
spec.pull medium
3,75 spec.pull high
Cullet
spec.pull low addition, dam
3,50 wall, deep
total spec.pull in t/m2d

>2,6t/m2d
refiner,
3,25
~2,5t/m2d furnace
3,00 dimensions,
<2,4t/m2d quality?
2,75

2,50

2,25
~25kWh/t glass
2,00
0,00 2,00 4,00 6,00 8,00 10,00 12,00
% boosting relative to total energy consumption in %

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 119

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace efficiency

Endport furnace for container glass - boosting and fossil energy input vs. pull

1800 800

1600 nat.gas
700
boosting
1400
600
nat.gas in Nm3/h.

1200

boosting in kW.
500
1000
400
800
300
600
200
400

200 100

0 0
0 50 100 150 200 250 300 350 400
pull in t/d

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 120

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace efficiency

Endport fired furnaces- boosting in kW/m2 melting area vs. spec.pull

18
amber
16 feuille morte
14 dark green
2
boosting in kW/m

12

10

0
0 0,5 1 1,5 2 2,5 3 3,5 4 4,5
spec.pull in t/m2d

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 121

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace efficiency

t/c crown

Seeds
inspection

t/c glass t/c glass

furnace front riser

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 122

 Centre of Glass Science and Technology
Bangkok University Thailand

Furnace efficiency

Efficiency is a function-

- of different design parameter

- of customer related assortment (which equates not always maximum pull)
- of energy cost situation and energy sources variation according to market situation
- producers worry about process and quality data analysis
- use of intelligent data acquisition, analysis and control systems

Design is not everything- continuous process control, adjustments (manually or by

automatic control) is required.

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 123

 Centre of Glass Science and Technology
Bangkok University Thailand

Thank you for your attention

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 124

 Centre of Glass Science and Technology
Bangkok University Thailand

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 125

 Centre of Glass Science and Technology
Bangkok University Thailand

Seminar on Glass Science and Technology Sept 06- Dr.M.Lindig 126