Gearbox Failures in Wind Turbines

[MolaKule]

Wind Turbine systems were designed to operate over a 20 to 25 years period.

Here is an animation with the gearbox explanation starting at ~ 2:10

However, recent analysis of over 3,000 onshore wind turbines show they will only generate electric energy for 12-15 years.

Referring to the references below, what is the major failure mode within these gearboxes?

https://www.nrel.gov/docs/fy07osti/41548.pdf
https://www.nrel.gov/docs/fy16osti/66593.pdf


This Question of the Day is Open to all BITOG members.

 

[ABN_CBT_ENGR]

you want all of them?

NREL is slightly "conservative" on their published findings

 

[JHZR2]

Looks like bearing-induced failure resuting in FOD damage in various places.

Seems like a design flaw. We can build tens of MW ship propulsion reduction gears that last 30-50 years... granted mass may be less of a concern

 

[MolaKule]

you want all of them?

NREL is slightly "conservative" on their published findings

i am looking for the Major failure mode.

 

[ABN_CBT_ENGR]

It would be more correct to say its a design that currently technology and materials science cant quite overcome.

The "major failure mode" is fatigue ( mostly axial) brought about by in determinant bend moments and work hardening of gear teeth and races. This allegedly contaminates the oils and degrades them. It gets "blamed' on all kinds of stuff.

Factually there are conflicting failure modes and mechanisms depending on what methodology is used.

That's not coming from the net or a conference, that's from actually doing it CONUS and OCONUS under auspices of several major EPCM firms and even Sandia.

 

[Kestas]

I see and analyze many of the bearing failures in windmills. Much of it is related to poor lubrication and possibly hydrogen damage.

 

[WyrTwister]

Poor lubrication ? As in , wrong lube , low on lube , luge not being maintained / changed or poor / badly designed lube system ?

Hydrogen damage ? In the manurfacturing process of gears ? Bearings ?

 

[MolaKule]

Okay folks, we're not asking for a dissertation or asking you to filibuster with your responses, nor how you would design or redesign a wind turbine's transmission; that is not the intent of this QOTD and is off topic.

Furthermore, Finger pointing at one or one or more bearing manf. is silly because all turbine bearings are facing the same problem. Finger pointing posts will be removed.

Argumentative posts will be removed; Extraneous posts not deemed on-topic will be removed as well.

Referring to the references below, what is the major failure mode within these gearboxes?

https://www.nrel.gov/docs/fy07osti/41548.pdf

https://www.nrel.gov/docs/fy16osti/66593.pdf


This Question of the Day is Open to all BITOG members.

 

[ArrestMeRedZ]

I only read the first source, but it states plain as day the main failure source is bearing failure. Bearing surfaces start to fail, design tolerances are exceeded, contaminants (bearing fragments) enter the system, etc. It also mentioned bearing manufacturers aren't in the loop for application specific information - housing stress and other application related necessary information needed for bearing design.
It seems to me these turbines are assembled components and not a systems designed unit. As such, the "design life" specification is more of a "desired life" based only on individual component lifespan, and not a realistic overall unit life expectancy. Kind of like "We want these things to last 20 years for planning purposes, so we are going to pretend they do."
To fix the problem in future units design the things properly using systems engineering concepts. In the meantime, increase the maintenance of existing units and sample, sample sample. Oil analysis should reveal early stages of bearing failure so units can be rebuilt before they are run to catastrophic failure.

 

[MolaKule]

I find it more convenient to now place the answer in PDF format.

 

[ragtoplvr]

There was recently a Drenchearo that passed thru windmill country of Iowa and Illinois. Anyone know how the windmills did or did not survive.

Those members in the area might watch in the coming months and see if a lot of turbines are shut down or under repair as it seems reasonable that design limits were approached if not exceeded. 100MPH wind with rain, hail, farmers roofs, etc.


Rod

 

[wpod]

My uneducated guess would be under engineered or low quality made is China parts . These wind mills have been failing faster than expected to quite a while.

 

[JustN89]

I honestly have no clue, but I'll hazard a guess because guessing is free.

"4. The majority of wind turbine gearbox failures appear to initiate in the bearings. These failures are occurring in spite of the fact that most gearboxes have been designed and developed using the best bearing-design practices available. Therefore, the initial focus of this project will be on discovering weaknesses in wind turbine gearbox bearing applications and deficiencies in the design process."

This is what stands out to me. If the majority of failures are initiating within the bearing, and the "best" bearing designs and practices are being used to this day, then something is occurring that we did not account for. My guess would be lubrication within the bearings- considering the load changes in some places of the gearbox compared to others, my guess would be that a certain load would need to be maintained to ensure lubrication, particularly in high load/low wind speed environments. Whether or not a bearing design change would fix this, or if the solution is tribological in nature, I couldn't tell you. Like I said, it's just a guess.

 

[andyd]

What is WECS? TIA. Wild guess is the gearboxes to control blade pitch are too small, The main shaft bearings and planetary gears etc are too small. Shafts cant handle side loads because it is too small. Blades are 150 ft long, that's a huge amount of leverage. Some one mentioned that it's a bunch of off the shelf assemblies.Not designed to specifically function as a self contained unit Job is only 1/2 done. Now de-bug it.

 

[Kestas]

WEC is White Etch Cracking. This is a phenomenon of subsurface microstructure decay from Hertzian loading. When it progresses, it develops suburface initiated spalling. Introduction of hydrogen into the material hastens this process.

 

[ragtoplvr]

They are building out a large windmill farm near me. The nacell that houses the generator, gear box and pitch mechanism looks very small compared to the size of the blades. It offends my sense of scale. Making things bigger adds mass and wind resistance, so it does not always help. However the ratio seems quite high and suspect they are simply too small.

Rod

 

[MolaKule]

I honestly have no clue, but I'll hazard a guess because guessing is free.

"4. The majority of wind turbine gearbox failures appear to initiate in the bearings. These failures are occurring in spite of the fact that most gearboxes have been designed and developed using the best bearing-design practices available. Therefore, the initial focus of this project will be on discovering weaknesses in wind turbine gearbox bearing applications and deficiencies in the design process."

This is what stands out to me. If the majority of failures are initiating within the bearing, and the "best" bearing designs and practices are being used to this day, then something is occurring that we did not account for. My guess would be lubrication within the bearings- considering the load changes in some places of the gearbox compared to others, my guess would be that a certain load would need to be maintained to ensure lubrication, particularly in high load/low wind speed environments. Whether or not a bearing design change would fix this, or if the solution is tribological in nature, I couldn't tell you. Like I said, it's just a guess.

The answer was given in the PDF 10 days before you posted.

 

[JustN89]

The answer was given in the PDF 10 days before you posted.

To show my intelligence, I never found the answer.

 

[sloinker]

Time to get rid of those newfangled bearings and get some babbitt and a shaft grounding system for those stray currents.

 

[blingo]

Observations linking to tribolayer thicknesses were also mentioned some time ago (Mr. Mayer): https://www.lubesngreases.com/magazine/wec-killer-at-large/