Oil filter analysis (OFA) on continuously operating wind turbines The black holes of oil analysis

Size: px

Start display at page:

Download "Oil filter analysis (OFA) on continuously operating wind turbines The black holes of oil analysis"

Simon Sparks
6 years ago
Views:

1 Oil filter analysis (OFA) on continuously operating wind turbines The black holes of oil analysis Gerrit van Middelkoop Bianca Coolen Page DEKRA

2 Contents Company introduction Condition Based Maintenance and oil analysis Effectivity of lab methods Interpretation of oil analysis results Black holes of oil analysis and their consequences There are alternatives! Conclusions and recommendations Page DEKRA

3 Van NS naar DEKRA Rail 2014 Page DEKRA

We will be the global partner for a safe world Prevent accidents to protect lives In transport, at home, at work Global partnering with all stakeholders Global thought leader in safety Because safety

4 We will be the global partner for a safe world Prevent accidents to protect lives In transport, at home, at work Global partnering with all stakeholders Global thought leader in safety Because safety pays out DEKRA is An expert organisation with people active in over 50 countries focussing on Safety, Quality and Environment presenting a service portfolio in three Business Units: DEKRA Automotive DEKRA Industrial (a.o. DEKRA Rail) DEKRA Personnel Page DEKRA

DEKRA Rail Utrecht Company 65 employees 3 labs & field measurements Top 3 (often #1) market position Central location in NL (Utrecht) From insight to implementation Customers Infrastructure &

5 DEKRA Rail Utrecht Company 65 employees 3 labs & field measurements Top 3 (often #1) market position Central location in NL (Utrecht) From insight to implementation Customers Infrastructure & transport, including heavy rail, tram, metro Construction & maintenance Asset managers: logistics, energy, civil engineering Services Testing Inspection / Monitoring Certification Applied Contract Research Fields of expertise Materials science Mechanics Tribology Electrical Engineering Equipment Calibration Page DEKRA

repair and logistics of electrical and geometrical parameters Testing & analysing of metal/plastic constructions and components, like

6 Laboratory Facilities Complementarity for effective system analysis Chemical lab Electrical & geometrical lab Materials lab Field lab Tribology: analysis & testing of oil, grease and fuels, tens of thousands samples for year well equipped lab Equipment calibration, repair and logistics of electrical and geometrical parameters Testing & analysing of metal/plastic constructions and components, like wheels, couplings, rails, and weldings On-site measuring & monitoring of EMC-levels, forces, vibrations and noise Page DEKRA

Five Competence Areas Coherent, multidisciplinary expertise Testing Electric EMC / cross-talk Detection ERTMS Dataengineering Calibration.

7 Five Competence Areas Coherent, multidisciplinary expertise Testing Electric EMC / cross-talk Detection ERTMS Dataengineering Calibration. Electrical Physical Geometrical Logistics Mechanics Forces & Acceleration Noise & Vibrations Modelling helping customers with optimising the exploitation of their technical, moving systems Materials Degradation Conservation Welding Wear NDT Tribology Lubrication Degradation Adhesion Life span Inspection Certification Applied Research Page DEKRA

8 Customers Diversity within but also outside Rail Transport Infrastructure * Industry 8 Page DEKRA

9 PDCA-circle for the asset manager services: condition management 9 Page DEKRA

10 Classical oil analysis for Condition Based Maintenance Benefits Effective for general purpose Cheap and quick Plenty lab capacity available Page DEKRA

11 Pitfalls of oil analysis for Condition Based Maintenance - Technicians are not aware of black holes in classic oil analysis - Interpretation lab data into actual machine risks not straightforward - Advice on effective maintenance actions not straightforward - Different labmethods produce incomparable results - Insufficient info on construction and operating conditions - Bad sampling and frequency Page DEKRA

Optimal design of condition monitoring Condition of

limits interpretation machine condition Target

Hitrate R/I = 100% sufficient Miss Rate M/I = 0?

Unjustified rejection: false alarm R: Justified

12 Optimal design of condition monitoring Condition of Machine population Sampling+Lab tests + rejection limits interpretation machine condition Target situation A: Legitimate approval: no action required Hitrate R/I = 100% sufficient Miss Rate M/I = 0? M: Unjustified approval: misser I:inadequate F: Unjustified rejection: false alarm R: Justified Rejection: maintenance required F low R = I Page DEKRA

13 Effective lab methods for condition monitoring of oil lubricated systems Diameter of wear particles µm Hydraulics Engines Compressors Gearboxes test method metal content: ICP/Rotating disc XRF Wear particle amount and diameter : Particle counters/ferrography Wearscan/microscopy Magnetic Plug Analysis MPA Oil Filter Analysis OFA Page DEKRA

14 Crucial: correct interpretation of machinecondition Many small particles Normal wear Abnormal wear Machine construction Operating conditions Small particles Help Wear! Extreme wear Sample frequency Service life Big particles high Ironcontent Very big particles remnants processing materials Thick/thin particles In general: big particles <- problem Page DEKRA

15 Wear material in practice Page DEKRA

16 Black Holes wear particles: Magnetic plugs in gear boxes Full flow filters in gear boxes Filters in engines Additional filtration for oil conditioning Page DEKRA

17 Black hole wear particles: magnetic plugs Page DEKRA

Changing Problem into Solution Magnetic plug analysis Problem: Most wear indicators lost in oil sample No effective condition monitoring with classical methods possible Solution: Use debris on plug

18 Changing Problem into Solution Magnetic plug analysis Problem: Most wear indicators lost in oil sample No effective condition monitoring with classical methods possible Solution: Use debris on plug as indicator in microscopic analysis 25 year of proven effectiveness Hitrate ~ 100% Restrictions: Normalization of subjective measurements Interpretation requires much expertise and experience Page DEKRA

Example: Operation of windturbines Consequences of failing gear boxes Damage probability = 10% Average down time in 10 years > 4 months Consequences (despite current inspection techniques) Estimated

19 Example: Operation of windturbines Consequences of failing gear boxes Damage probability = 10% Average down time in 10 years > 4 months Consequences (despite current inspection techniques) Estimated unnecessary anual production loss in Holland 30 M Significant loss op production (per turbine: 350 MWh, ~ 25k ) availability of spare parts (leadtime 2-6 months) High repair costs (up to 400 k /defect) Each year more than 5 M unnecessary repairs Reputational damage Page DEKRA

20 Black hole wear particles : full flow filters Page DEKRA

21 Effect of full flow Oil Filter No indicators left in oilsample Diameter of wear particles µm Hydraulics Engines Compressors Gearboxes test method metal content: ICP/Rotating disc XRF Wear particle amount and diameter : Particle counters/ferrography Wearscan/microscopy Magnetic Plug Analysis MPA Oil Filter Analysis OFA Page DEKRA

22 OFA How it works Advanced test of oil filter representative sample selection sent filter Identification wear particle quantity, shape and size Determination chemical composition wear particle components by XRF Interpretation (damage type) Based on in-depth knowledge of damage mechanisms Based on fingerprinting of hundreds of thousands of previously analyzed gearboxes Assesment: current technical status and prediction Utilizing trending, statistics and engineering expertise Reporting: clear and simple Page DEKRA

Therefore Oil Filter Analysis (OFA) OFA replaces all conventional condition monitoring methods for Windturbine gearboxes and provides reliable results Methods Hit rate

23 Therefore Oil Filter Analysis (OFA) OFA replaces all conventional condition monitoring methods for Windturbine gearboxes and provides reliable results Methods Hit rate Conditiemonitoring End of warranty discussion Damage investigation OFA high Oil analysis low Endoscopy average Vibration analysis average Page DEKRA

Possible OFA Benefits Lower costs, increased revenues Significantly lower costs and increased revenues Increased early detection probability: less frequent and shorter down-time Less repair and crane

24 Possible OFA Benefits Lower costs, increased revenues Significantly lower costs and increased revenues Increased early detection probability: less frequent and shorter down-time Less repair and crane costs Less production loss for jam recovery Optimizing Maintenance Extending maintenance intervals Less maintenance Less production loss for scheduled maintenance Fewer unplanned maintenance and repairs Functioning wind turbines: increased business reputation Page DEKRA

25 Black holes of oil analysis Conclusions Oil Analysis is a valuable condition monitoring technique But : interpretation of test results requires thorough expertise filters and plugs can ruin effective condition monitoring knowledge of machine construction is essential for detection Magnetic Plug and Oil Filter Analysis are effective solutions for black holes Page DEKRA

26 Black holes of oil analysis Suggestions 1. Evaluate possible black holes before starting programs 2. Choose effective analysis methods 3. Keep the score: Check predicted wear versus machine inspection results (Hit rate) Page DEKRA

27 End of presentation Questions? Page DEKRA

Improving predictive maintenance with oil condition monitoring.

Improving predictive maintenance with oil condition monitoring. Contents 1. Introduction 2. The Big Five 3. Pros and cons 4. The perfect match? 5. Two is better than one 6. Gearboxes, for example 7. What