1

Operational Characteristics of Existing Dry Cooling Systems in Eskom

EPRI Workshop on Advanced

Cooling Technologies, 8 July 2008

Francois du Preez

2

Content

• Eskom overview

• Dry Cooling History

• Kendal – Operating experience with Indirect System

• Matimba – Operating experience with Direct Cooling System

• Conclusion

• Acknowledgement

• References

3

Eskom Power Stations

Pretoria

Johannesburg

Bloemfontein

Cape TownEastLondon

Port Elizabeth

Durban

Maputo

MOZAMBIQUE

LESOTHO

NAMIBIA

CCC

CCC

CC

CCCC

P

H

P

G

GN

H

SWAZILAND

BOTSWANA

ESKOM POWER STATIONS 2008C

G

G

App. 42 000 MW installed capacity> 85% coal fired

TypeCoal (C)Gas Turbine(G)Hydro Electric (H)Pump Storage (P)Nuclear (N)

4

Eskom’s Dry Cooling History (Commissioning dates)

* Still to be finalized

• Decision to extend Grootvlei in 1966 with dry cooling system-Lack of adequate water resources close to coal fields

•1971 Grootvlei 5, 200 MW, Indirect

•1977 Grootvlei 6, 200 MW, Indirect

•1987-1992, Matimba 1-6, 6 x 665 MW, Direct

•1988-1993, Kendal 1-6, 6 x 686 MW, Indirect

•1996-1998, Majuba 1-3, 3 x 657 MW, Direct

5

Average specific water consumption

0

2000

4000

6000

8000

10000

12000

1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

Year

MW

0

0.5

1

1.5

2

2.5

l/kW

h

Total installed dry cooled capacitySpecific water consumption, l/kWh

Average specific water consumption for dry cooled station is 0.2 l/kWh

28% of coal fired capacity

6

Kendal Cooling system

• 6 x 686 MW Electrical Output

• Surface condenser with SS tubes

• Circulating water flow: 266 700 gpm

• Galvanised heat exchanger tubes– 11 sectors which can be individually

isolated

– Total of 1 980 km (1 230 miles) of finned tube/tower

– Horizontal, radial arrangement

• Tower dimensions

– Diameter at tower base 144 m (462 ft)

– Total height 165 m (541 ft)

• Thermal design

– Known turbine characteristics, energy output was maximized over given ambient temperature range

• 3.4 MW auxiliary power consumption/unit

– 3 x 50% pumps (units 1-3)

7

View of cooling towers and circulating pumps

8

Kendal cooling system thermal design

Kendal cooling system design

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

180.0

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0

Ambient Temperature, deg F

deg

F

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

"Hg,

deg

F

Saturation Temperature

Cold water temperature

Hot Water Temperature

"Hg

Condenser TTD

Annual average ambient temperature

(Design) Circulating water flow rate is 243 000 gpm(Actual) flow is 266 700 gpm

LP turbine protection is set at 24.8 in Hga

Design heat rejection rate = 895 MW

9

Wind effect on cooling tower

Full load duty = 895 MW

10

Meteorological effects on tower performance cont.

• Show 2 graphs of wind and ambient T

•Air inlet temperature different to ground level temperature

•Average ground level temperature may differ from average air inlet temperature (4°C difference at Kendal)

Kendal 2005-2006 data, Ambient air temperature distribution comparison Based on average temperatures measured at Kendal Cooling towers and K2 weather station

0

200

400

600

800

1000

1200

(-3)

-(-2

)(-

2)-(

-1)

(-1)

-0 0-1

1-2

2-3

3-4

4-5

5-6

6-7

7-8

8-9

9-10

10-1

111

-12

12-1

313

-14

14-1

515

-16

16-1

717

-18

18-1

919

-20

20-2

121

-22

22-2

323

-24

24-2

525

-26

26-2

727

-28

28-2

929

-30

30-3

131

-32

32-3

333

-34

34-3

535

-36

36-3

737

-38

38-3

939

-40

40-4

141

-42

42-4

3

Ambient air temperature [°C]

Hou

rs p

er a

nnum

Cooling tower dataK2 data

Air Temperature at weather station measured 1.2 m above ground level.

11

Cooling tower performance

Cooling tower performance

-2

0

2

4

6

8

10

Oct-95 Mar-97 Jul-98 Dec-99 Apr-01 Sep-02 Jan-04 May-05 Oct-06 Feb-08 Jul-09

Date

Col

d W

ater

Tem

pera

ture

vs.

des

ign,

deg

C

UNIT3

UNIT6

Washing of heat exchangers

Natural fouling only

12

Cooling system operational experience

• Thermal performance is independent of wind direction

• Wind effect predictable and repeatable

• Very stable operation due to large circulating water volume

• No noise

• Tower meet design performance after 20 years of operation

• Minimal maintenance cost (circulating water pumps)

• No corrosion observed on heat exchangers

• Nitrogen blanket to prevent freezing & corrosion in drained sectors

• Limited surface area under vacuum

• Visible structure

13

Matimba design

• 6 x 665 MW Output• Design: Known turbine characteristics,

energy output was maximized over given ambient temperature range

• Ave. back pressure 18.6 kPa (5.5 in Hga)• LP turbine protection: 65 kPa (19.2 in Hga)• Average steam velocity 80 m/s at 18.6kPa• Station orientated with prevailing wind

direction towards boiler• 2 x 5 m (16.4 ft) exhaust ducts• ACC details per unit

– 48 fans, 30 ft diameter– 8 streets with 6 fans per street– Street length 70.8 m (232 ft)– 45 m (147.6 ft) air inlet opening– 12 MW auxiliary power consumption

• Total platform footprint 35 700 m2 (8.8 acre)

14

Matimba thermal design

Matimba ACC thermal design

0

20

40

60

80

100

120

140

160

180

200

20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 110.0Ambient temperature, deg F

Tem

pera

ture

, deg

F

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

in H

ga

Saturation temperature

ITD

"Hg

Average ambient temperature

15

Matimba air inlet temperature to ACC

0

5

10

15

20

25

30

35

40

45

06:45

07:35

08:25

09:15

10:05

10:55

11:45

12:35

13:25

14:15

15:05

15:55

16:45

17:35

18:25

19:15

20:05

20:55

21:45

22:35

23:25

00:15

01:05

01:55

02:45

03:35

04:25

05:15

Time

Tem

pera

ture

, C

ACC inlet temperatureWeather mast temperature, 1.2 m AGLWeather mast temperature, 40 m AGL

September 2005

Weathermast located 500 m from ACC

16

Annual temperature profile

0

100

200

300

400

500

600

700

800

900

-1 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41

Ambient air temperature, deg C

Hou

rs p

er a

nnum

1.2 m AGL, Average temperaure = 19 °C

40 m AGL, Average temperature = 23.7 °C

Temperature measured 40 m AGL:

24 hour daily average = 23.02 °C 06:00 - 21:00 average = 23.7 °C

Above Ground Level (AGL)

17

Air inlet temperature to ACC

• ACC draw air from higher elevations

• Ground temperature data, say 1.2 m AGL, is not representative and as an average is about 4°C conservative (Matimba site).

– Similar for indirect system

• The instantaneous difference between ground level, 1.2 m AGL, air temperature and ACC inlet temperature can be as high as 5-8 °C at Matimba

• The seasonal air temperature variation experienced by the ACC is less than that measured at ground level.

– ACC experience smaller temperature variation than the extreme maximum and minimum measured on the mast at ground level.

• The variation in air temperature as a function of height above ground level to be considered for new designs depending on the daily and seasonal power demand profile

18

Majuba unit 1 trip during unsteady wind period

0

50

100

150

200

250

2004/11/1314:45

2004/11/1314:52

2004/11/1315:00

2004/11/1315:07

2004/11/1315:14

2004/11/1315:21

2004/11/1315:28

2004/11/1315:36

2004/11/1315:43

2004/11/1315:50

Date & Time

deg

F, M

W (%

), C

urre

nt

0

5

10

15

20

25

LP e

xhau

st p

ress

ure,

"H

g

Generator MW, % Saturation temperature ACC Fan Motor Current ACC air inlet temperature LP exhaust pressure

Turbine protection set at 20.7 in HgaAir Cooled Condenser

Turbine

Boiler2

Boiler3

Boiler1

Wind direction during trip

19

Summary of wind effect at Matimba

• Energy loss relatively small

• Capacity loss during incidents is a concern

– High risk periods are normally of short duration during unsteady windy period preceding a thunder storm

– 12 units trips occurred at Matimba during first 7 years of operation

– Unreliable power production

– Output dependent on wind direction, wind speed and ambient temperature

• “Exposed” units, 1 and 6, are generally more susceptible to wind effects

– Deterioration in fan performance is the predominant factor, more than hot air plume recirculation.

• Large operating margin to be maintained between condenser pressure and turbine protection setting.

20

Matimba modifications to minimise wind effect• Modifications are limited due to position of ACC• Modifications simulated with Computational Fluid Dynamic models• Cladding removed between ACC and turbine house

–Improve air flow to ACC when wind is from boiler–Negligible effect in performance with wind towards boiler

• Cladding removed from turbine building above turbine floor and turbine house roof vents modified

• Weather mast erected with visual indication in the unit control room• Daily weather forecast received from the Weather Service• Automatic vacuum de-loader as well as vacuum rate of change run-

back added to unit control system

21

Reliability of ACC

• During commissioning a large percentage of fan blades cracked, most of which were repaired.

• Up to 40% of dephlegmator surface area was ineffective due to condensate build-up. Rectified by modification.

• Fan gearbox oil temperatures very high, typically above 90 °C

– Annual oil changes required.

• Natural and artificial fouling on air side removed by semi-automatic high pressure water cleaning system

• Significant corrosion in exhaust duct and tube inlet area

22

Corroded tube entry with trough wall perforation

23

Corrosion cont.

• Steam pH increased to 9.6 - 9.8 with great success in minimising steam side erosion

• Condensate extracted from ACC duct indicated that the liquid has a lower pH compared to the bulk condensate or steam

• This mechanism is subject to a separate EPRI investigation

8.5

9

9.5

10

10.5

2006/11/29 2007/01/18 2007/03/09 2007/04/28 2007/06/17 2007/08/06 2007/09/25 2007/11/14

Time

pH

ACC DuctCPP inlet

24

Conclusion

• Both direct and indirect dry cooling systems can be applied successfully in coal fired stations

• Specific water consumption if dry cooled station of both types is approximately 0.2 l/kWh

• Significant progress was made towards understanding environmental effects on the performance of dry cooling systems

• Average annual ambient temperature was underestimated in original specification

• Lessons learnt on existing installations should be incorporated in the specifications for new solutions

25

Acknowledgement

• Information and assistance provided by power station personnel

• Eskom Management for sponsoring the trip and permission to present the paper

26

References

• Ham, A.J., West, L.A., Eskom’s Advance into Dry Cooling, VGB Conference, South Africa, Nov 1987.• Goldshagg, H.B., Lessons Learnt from the Worlds Largest Forced Draft Direct Air Cooled Condenser, Paper presented at the

EPRI Int. Symp. On Improved Technology for Fossil power Plants – New and Retrofit Applications, Washington, Marc 1993.• Goldshagg, H.B., Winds of change at Eskom’s Matimba Plant, Modern Power Systems, January 1999.• Veck, G.A., Rubbers, P.J.E., Economic Evaluation of dry Cooling: The Options, VGB Conference, South Africa, Nov 1987• Trage, B., Hintzen, F.J., Design and Construction of indirect Dry Cooling Units, VGB Conference, South Africa, Nov 1987• Knirsch, H., Design and Construction of Direct Dry Cooling Units, VGB Conference, South Africa, Nov 1987• Duvenhage, K., Du Toit, C.G., Kröger, D.G., Methods to Combat the influence of cross winds on the fan performance in

forded draught air-cooled heat exchangers, Proc of the 1st South African Conference on Applied Mechanics (SACAM) 1996, Midrand, South Africa, July 1996.

• Goldshagg, H.B., Vogt, F., du Toit, C.G., Thiart, G.D. and Kröger, D.G., air-cooled Steam Condenser performance in the presence of Cross-winds, EPRI TR-108483 2113, Proceedings: Cooling Tower Technology Conference, pp. 1.61- 1.77, July 1997.

• Kröger, D.G., Air Cooled Heat Exchangers and Cooling Towers, Department of Mechanical Engineering, University of Stellenbosch, 1998.

• Phala, S., Aspden, D., du Preez, F., Goldschagg, H., and Northcott, K., “Corrosion in Air-cooled Condensers – Understanding and Mitigating the Mechanisms”. API Power Chemistry Conference. Sunshine Coast, Queensland, Australia. May 2008.

• Du Preez, A.F. and D.G. Kröger, Investigation into the Influence of Cross-Winds on the Performance of Dry-Cooling Towers, Proc. 8th Cooling Tower and Spraying Pond Symposium. Karlsruhe, Germany, 1992.

• Du Preez, A.F. and D.G. Kröger, Effect of Wind on Performance of a Dry-Cooling Tower, Heat recovery Systems & CHP, Vol. 13, No. 2, pp. 139-146, 1993

• Du Preez, A.F., Innovative Water Conservation Technology: Dry Cooled Power Stations, Paper presented at the WaterDome World Summit on Sustainable Development, Johannesburg, September 2002.