Preventing & Identifying Failures of Dead Break Elbows Wind - PowerPoint PPT Presentation

Preventing & Identifying Failures of Dead Break Elbows Wind Farm Applications 1

Introduction • Brian Peyres, Senior High Voltage Reliability Engineer, EDP Renewables. • Level 2 Thermographer with 20 Years in the generating and utility industry. 2

Introduction • Dead break elbows are commonly used by utilities and wind energy generators. 3

Introduction • Dead break style connections are a reliable connection if installed and maintained correctly. • This presentation will highlight nearly 6 years of work investigating failures and developing preventative maintenance plans to eliminate termination failures. 4

Wind Farm Collection Systems • 10-17 turbines with a generating capacity up to 28 MW per circuit are typical. 5

Wind Farm Collection Systems • Up to 165 high voltage terminations per circuit. 6

Wind Farm Collection Systems • Average of 1300 high voltage terminations per 200MW wind farm. 7

Faults 8

Faults 9

Faults 10

Faults 11

Faults 12

Faults 13

Faults 14

Faults • At our peek we were losing on average 2% of our assets due to termination failures. • This means roughly 35 circuit failures per year. • Considering generation losses, damaged equipment and labor each circuit failure could result in nearly $ 80,000 in total losses. 15

Maintenance • In early 2010 we began an aggressive Infra-Red inspection approach. • The results of the IR scans were quite alarming. 16

The finds over the years 17

The finds over the years Damaged Stud and surface pitting 18

The finds over the years 19

The finds over the years 20

The finds over the years 21

The finds over the years Screwdriver used during construction to twist and form cable termination into position. 22

The finds over the years Cross threaded stud resulting in metal shavings and poor connection. 23

The finds over the years End cap could be removed by hand when this unit was repaired. 24

The finds over the years 25

The finds over the years 26

The finds over the years Notice the anomaly at the Loose connection was back of the T body? found internal to the transformer under oil. 27

Why? 1. Connection not properly tightened. a. Ensure termination is free of defects and mating surfaces are parallel to each other. b. Tighten the end plug as much as possible by hand. c. Tighten to 55 ft-lbs. d. Check fitting by moving T-body front to back. e. Torque again to 55 ft-lbs. 28

Why? 2. Contamination a. Watch for metal shavings, caused by the termination resting on the stud as it is placed into position. 29

Why? Metal shavings resulting in poor contact and damaged threads. 30

Why? 3. Environmental Stresses a. Water collecting in the lower end of the vault can freeze and add weight to the cable. b. Thermal cycling is common on wind farm equipment. This continuous thermal movement could have effects on termination reliability. 31

Why? Support bar and cable clamps added to help mitigate movement and support weight. 32

Why? 4. Studs ! Watch out for imitations ! Studs supplied by transformer manufacturer. Not uniform, various lengths and poorly machined. Studs supplied by T-Body manufacturer. Uniform in shape, same length. 33

Why? Shown here is a stud from the T-Body manufacturer 34

Why? Overall Flange Threads "E" "A" "B" "C" "D" Origin Diameter Length Length Depth Bottom Top T-Body Manf. 15.25 61 14.3 1.25 23.5 23.5 D T-Body Manf. 15.25 61 14.3 1.25 23.5 23.5 T-Body Manf. 15.25 61 14.3 1.25 23.5 23.5 T-Body Manf. 15.25 61 14.3 1.25 23.5 23.5 B A Tranformer Manf. 15.5 63 10.25 1 26 26.75 Tranformer Manf. 15.25 63.5 10.5 1.25 26.5 26 Tranformer Manf. 16 65 10.25 1 28 26.75 Tranformer Manf. 15.25 63 10 1 26 27 C Tranformer Manf. 15.25 66 9.75 1.5 29 26.5 Tranformer Manf. 15.5 63 8 1.25 29 26 Tranformer Manf. 15.5 63 9.5 1.5 27.5 26 Tranformer Manf. 16 62.5 10.5 0.9 25 27 E Tranformer Manf. 16 63 10 1.5 27 26 Not a single stud supplied by the transformer manufacturer were machined to the same dimensions as the T-Body manufacturer. 35

Why? Damaged or poorly End of the stud at a machined studs will wider diameter than not fully seat, this will the inner. result in a false positive torque. 36

What it should look like 37

What it should look like This area must be free of debris, secure and tight. 38

Experiment Time! This area must be free of debris and secure and tight. 15 minutes into 100amp test. Bare connector 44.2 ° C Shielded boot 26.9 ° C ΔT 17.3 ° C 39

Experiment Time! This area must be free of debris and secure and tight. 45 minutes into 100amp test. Bare connector 69.6 ° C Shielded boot 35.7 ° C Δ T 33.9 ° C 40

Experiment Time! This area must be free of debris and secure and tight. 75 minutes into 100amp test. Bare connector 72.3 ° C Shielded boot 40.9 ° C Δ T 38.4 ° C 41

Experiment Time! This area must be free of debris and secure and tight. 75 minutes into 100amp test. Bare connector 72.3 ° C Shielded boot 40.9 ° C Δ T 38.4 ° C Good rule of thumb, add a multiple of 1.8 to what you read outside to estimate what your generating inside. 42

Why? • False positive torque or poor craftsmanship = 43

Questions? 44