Leslie Barker, Chief Engineering Advisor, Eskom GBE Fleet Technology, Gauteng, South Africa

Leslie Barker, Chief Engineering Advisor, Eskom GBE Fleet Technology, Gauteng, South Africa

Improving Gearbox Efficiency and Reliability using High Performance Gear Oils. LF Barker Generation Business Engineering. Fleet Technology Department.

Overview of presentation Background to Energy Savings Lubricant selection Lab Testing of oils. Laboratory trial results Options to optimise Energy from Lubricants In-Service trial Data Gathered Results Conclusions Acknowledgements

Background Research into the use of high performance gear oils indicates that these oils could significantly reduce energy usage. They also offer extended equipment life, reduced oil consumption and improved overall reliability of gearboxes. This presentation reports the preliminary results of recently conducted laboratory and in-service trials using synthetic oils that conclusively support these claims.

Background to Energy Savings Electrical Power is severely constrained due to time required to build new generating capacity. Eskom has embarked on program to investigate areas where energy can be saved. Most initiatives to date focus on electrical consumption eg. Electric motors, lighting, water heaters and demand patterns etc. Very little focus to date on mechanical efficiencies and process efficiencies.

High performance Gear Oils. Historically, use of synthetic lubricants to improve gearbox life was considered expensive. In 2007 a series of research projects were approved to consider energy savings as a factor in oil selection. A project was initiated to test a range of candidate gear oils in a laboratory environment and later take this into an in-service trial. Eskom s Demand Side Management (DSM) sponsored the laboratory testing. Eskom Research Division funded the in-service trial. Eskom Matimba power station conducted the inservice oil testing.

Options to optimise Energy from Lubricants Ensure the correct viscosity is used (use minimum viscosity acceptable). Use high Viscosity Index lubricants (they thin out less with increased temperature). Use high performance oils designed for energy efficiency.

Lubricant selection 19 Oils were identified for the laboratory trial. The tests were conducted by the Tribology Laboratory in the Department of Chemical Engineering at Pretoria University. The test method included a modified FZG machine which could measure heat loss and mass loss. Effectively a heat balance model was developed to identify oils with friction lowering properties.

Lab Test Variables The laboratory test is set up to change speed, load and temperature as variables. Oils were selected to represent: Different viscosities (oils of ISO VG 220,320, 460). Different oil types (mineral oils, synthetic oils and one engine oil).

Schematic of FZG Machine Standard FZG layout Modified FZG method using forced cooling to maintain oil Temperatures.

Laboratory trial results Total Heat Removed (kj) 15000 14000 13000 12000 11000 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 Total Heat Removed per Test 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Lubricant No. Test1 Test2 Test3 Test4 Test5 All lubricants were subjected to the same test sequence in which either speed load or temperature where changed. Test gears were not removed between tests.

Oil type versus energy. Heat (kj) Total 16000 14000 12000 10000 8000 6000 Samples in red were chosen for in service testing Samples in red were chosen for in service testing 4000 2000 0 Synthetic (PAO ester ) Synthetic (PAO ester ) Synthetic (PAO) Mineral Oil (Semi - synthetic) Mineral Oil (Diesel engine oil) Mineral Oil Mineral Oil Synthetic (PAO) Synthetic (PAO) Synthetic Synthetic Synthetic Mineral Mineral Mineral Oil (plus moly additive) Synthetic (PAO) Synthetic (PAO) Synthetic (PAO) Synthetic (PAO ESTER Blend) Lubricant type Most synthetic oils generated less heat than mineral oils in the lab test. Heat Removed (KJ)

Heat Generated vsoil Type and Grade Heat Generated vs Oil Type 16000 14000 12000 Heat (KJ) 10000 8000 6000 Heat (kj) PAO Heat (kj) Mineral Heat (kj) PAG 4000 2000 0 0 50 100 150 200 250 300 350 400 450 500 PAO Polyalphaolefin PAG Polyalkylene Glycol Viscosity Grade (Centistokes) Oils chosen for inservice trial

In-Service trial 13 gearboxes in the Matimba Air Cooled Condenser (ACC). 2 different designs from 2 different manufacturers (Flender and Hansen). Gearboxes are double reduction gearboxes with vertical shafts. Gearboxes drive fan of 9m diameter at about 14 rpm. Absorbed Power 220 kw per box. 288 gearboxes in total drawing 57 600 kw. Baseline test were run on an ISO VG 320 mineral oil. Final testing with an ISO VG 320 synthetic oil (PAO). Air Cooled condenser (ACC)

Data Gathered Electrical current (Amps) on each gearbox is measured continuously. Temperature probes and data collectors were installed to gather ambient temp and bulk oil temps. 3 refurbished Flender gearboxes were used. 10 new Hansen gearboxes were used. Flender Gearbox Hansen Gearbox

In Service Variables ACC fans are subjected to changes in: Wind speed and direction. Air density (temperature). Air Humidity. This results in fluctuating loads which require compensation in results. Compensate for air density. Compensate for change in motor power factor.

Bulk, Ambient and Delta T 100.00 90.00 80.00 3 Flender Boxes Bulk, Ambient and Temperature Rise ( C) 10 Hansen boxes Temperature 70.00 60.00 50.00 40.00 30.00 20.00 Temp Mineral Ambient Mineral Temp Rise Mineral Oil Temp Synthetic Ambient Synthetic Temp Rise Synthetic Oil 10.00 Temp rise synthetic oil - 1-22-50 2-22-30 3-33-20 3-22-20 3-22-50 3-23-20 3-23-60 3-24-20 3-24-40 Gearbox number/ location 3-31-10 3-31-20 3-31-40 3-32-60 Temp rise mineral oil

300.00 250.00 200.00 150.00 100.00 50.00 - Power Consumption Power Consumption (kw) 1-22-50 2-22-30 3-33-20 3-22-20 3-22-50 3-23-20 3-23-60 3-24-20 3-24-40 3-31-10 3-31-20 3-31-40 3-32-60 3 Flender Boxes 10 Hansen boxes Gearbox No/Location Power Mineral Power Synthetic Power (kw)

Power Consumed VS Temperature Rise 70.00 Bulk Oil Temp Rise VS Power Consumed Flender Gearboxes Temperature Rise (degrees Celcius) 60.00 50.00 40.00 30.00 20.00 Note shift to left and down showing drop in temperature and drop in power consumption Hansen Gearboxes Temp Rise Mineral Oil (Flender) Temp Rise Synthetic Oil (Flender) Temp Rise Mineral Oil (Hansen) Temp Rise Synthetic Oil (Hansen) 10.00-150.00 170.00 190.00 210.00 230.00 250.00 270.00 Power (KW)

Results Average Temperature drop of 8% for Flender gearboxes and 11% for Hansen gearboxes. Average Power drop before adjustments of 6.62% for Flender and 5.82% for Hansen gearboxes. Nett savings after adjustments 4.97% for Flender and 4.38% for Hansen gearboxes.

Conclusions A saving of at least 4% equal to 2304 KW is possible at Matimba should all gearboxes be converted to synthetic oils. Synthetic oils correctly selected can reduce energy consumption in helical boxes. Further work required to determine the impact in industry but it is estimated that in Eskom alone more than 10MW is available in other gear applications. Eskom lubes are less than 5% of RSA industry, which implies that as much as 200 MW could be available via a simple oil change! At R0.60/KWH and a 90% load factor, 2304 KW represents a saving of R10.9m per year.

Acknowledgements Eskom Research and Eskom DSM for funding the testing. Prof Philip de Vaal at Pretoria University for the Lab Trials and general advice. Mr Hein Goldshagg (System Engineer at Matimba) Mrs Andra Geldenhuys (Matimba Tribology) Mr Christo Greyling (Medupi Mechanical Engineer) All staff at Matimba who contributed. All Oil Companies who willingly participated in the trial.