Hi and welcome to Failurology; a podcast about engineering failures. I’m your host, Nicole
And I’m Brian. And we’re both from Calgary, AB.
Thank you again to our Patreon subscribers! For less than the cost of almost anything, because inflation sucks, you can hear us talk about more interesting engineering failures! Yes $5 Canadian per month to get more interesting engineering failure episodes.
On this episode of Failurology, we have a guest!! We are joined by Leann, a mechanical engineer in the renewables industry, Welcome Leann!!
This week in engineering news, staying on theme, boosting wind farm output, without new equipment.
This one is from MIT news.
Wind farms produce more than 5% of the world’s electricity.
Each wind turbine in the farm is controlled as an individual, free standing unit; even when there are dozens or hundreds in the farm.
Engineers at MIT found that just by modelling the wind flow of the entire farm, they can optimise individual units.
The increase in energy output is modelled at about 1.2% overall and 3% for optimal wind speeds.
This seems small. But the algorithm can be implemented at any wind farm. If it was implemented at every wind farm in the world, it would be the same as adding 3,600 new wind turbines or enough to power 3 million homes. And an increase of about $3 billion to producers.
In order to get the most out of the wind farm area, the turbines are spaced close together. Which means that the turbines in front send turbulent wakes to those downwind. Something that the current method of controlling turbines individually does not take into account.
The algorithm uses physics-based, data assisted modelling that learns from operational wind farm data to find the optimal orientation for each turbine at a given time, depending on the wind conditions. Turbines adjust their vertical axis position to align with the incoming wind direction and speed.
The article includes more information about how they developed and tested the algorithm in real wind farms If you want to read more about it, check out the link on the web page for this episode at failurology.ca
Now on to this week’s engineering failure; wind turbines.
We are going to do something a little different on today’s episode. Where normally we would narrow in on one failure or maybe two similar failures for the episode. Today we are going to talk about the main causes of wind turbine failure and discuss some examples of each. This because;
There is limited information on individual wind turbine failures.
Information available does not always include a cause.
Even if we had detailed information for a specific failure, it would likely fall into our mini failure category
Wind farm impact in Alberta
The local Light Rail Transit in Calgary, known as the C-train, is 100% powered by renewable energy. It’s the only LRT system in North America to use wind generated electricity to power the system, which I find incredibly interesting as Calgary is the corporate oil and gas hub of Canada, the Houston of Canada if you will.
There isn’t a dedicated group of turbines that are connected to the train system, but the City purchases enough wind power to match the LRT use.
I think there is a way to use the air generated by moving trains, specifically in a tunnel, as an energy source. But I have not done the maths to see how viable that would be. Something like that would probably see the most bank for its buck in places like New York City with an underground tunnel system, the subway, and a high frequency of trains.
Wind farm impact in Germany
Germany is the largest wind power producer in the European Union by a large margin; they have over 64,000 MW installed and in 2020 (which was their best year on record) they produced 132 TWh of wind power.
20% of the electricity produced in Germany last year was from wind power compared to 6% in Alberta. Germany produced more wind power last year than all of Alberta’s electricity generators combined.
It took the last 14 years of product development, industry development, government initiative, and growing pains for wind power to climb from 6% of the electricity market share to 20%. In that same span of time, the cost of wind power has been reduced by half.
There are four types of wind turbine failure that we are going to discuss today.
This sounds like pretty much every component of a wind turbine.
Failure 1 - blade failure
The main function of the blades is to boost energy production - either by increasing the length of rotor blades to catch more wind or by adjusting the blade pitch to account for variable wind speeds.
In 2021 the largest operating turbine was 14 Mega Watts with 107m long blades - one blade is the length of a football field! 15 and 16MW prototypes are scheduled for construction this year.
A longer blade adds stress along the entire length of the blade from the sheer weight and momentum. The adjustable blade pitch mechanism adds complexity and moving parts to the highly stressed connection point at the root of the blade.
Consider factors like wind magnitude, temperature, density, contamination, etc. are all varying wildly throughout the year, and even hour by hour. The blades take the brunt force of these wind variations. In 2018 it was estimated that 3800 blades are damaged each year on some 700,000 installed blades. The damage could be anything from cracked gel coatings to complete and rapid destruction.
Debonding of fibreglass layers
Joint failure at the blade root
Splitting along fibres
Gel coat cracks allowing for moisture ingress
Erosion at the leading edge
Excessive blade flex resulting in a blade striking the tower or extreme load buckling where the blade folds back on itself.
Inclement weather such as lightning strikes, rain/ice/hail, high winds
Material or controls or braking failure
Poor design that doesn’t account for high wind condition or fatigue design
Examples of wind turbine blade failure.
Perkins High School located in Sandusky Ohio on the southern shore of Lake Erie had three 20kW turbines installed January 23rd 2009 and connected to the power grid February 4th.
The weather on February 7 was reasonably windy with 35-40km/hr winds gusting up to 85km/hr. Note that the typical cut-out wind speed is 25m/s (90km/hr).
We weren’t able to find any official reports but the pre-investigation speculation was that one of the blades flexed in a gust of wind and contacted the tower. The rotor became unbalanced with only two of three blades spinning and the other two blades contacted the tower shortly afterwards.
To make matters worse, a second failure occurred at the Perkins High School installation less than two years later on November 29th 2010.
Again we didn’t find any official investigations but there were rumours of sheared bolts from the connection at the root of the blade. In this instance the turbine was brought to a stop before the other two blades failed.
It was reported that this was the third set of blades to be installed on the turbine - presumably both blade swap activities were done proactively as a result of the 2009 failure.
Both of these failures fall in the “infant mortality” range of the reliability bathtub curve and were likely either a design error or maintenance error.
In the first case either the blades were too flexible or perhaps the control system wasn’t well tuned to handle gusts of wind. If it was the first time the machine had been turned on it could have been a badly executed start up sequence.
The second case could have been either design error or maintenance error. Either the wrong bolt size or grade or torque could have been specified for the new reinforced blades or perhaps the wrong size or grade of bolts were installed or improperly torqued.
A similar bolt shear failure occurred in Sidinge, Denmark on February 23, 2008 where the root cause of the sheared bolts was deemed to be insufficient bolt torque that allowed some of the bolts to fail in fatigue which caused a domino effect for the remaining bolts.
It is possible that the bolts had been insufficiently torqued since the turbine was installed in 2000 and the deficiency was not found in any of the annual inspections.
Following this incident it was recommended that the Danish wind turbine certifying body produce a guideline for ongoing service and maintenance of wind turbines in Denmark including such critical items as bolt torque specifications and checks.
We really enjoyed all of our conversations and side tangents with Leann, so we didn’t want to edit those out. We’re going to pause here and bring you the second half of the wind turbine failures in our next episode.
So there you have it, the most common cause of wind turbine failures is blade failures. We also talked about optimizing wind farms and how wind energy has impacted Alberta and Germany.
For photos, sources and an episode summary from this week’s episode head to Failurology.ca. If you’re enjoying what you’re hearing, please rate, review and subscribe to Failurology, so more people can find us. If you want to chat with us, our Twitter handle is @failurology, you can email us firstname.lastname@example.org, you can connect with us on Linked In or you can message us on our Patreon page. Check out the show notes for links to all of these. Thanks, everyone for listening. And tune in to the next episode, our second part in this wind turbine failure series, where we’ll examine generator, gearbox and foundation failures and share more wind turbine failure examples.
Bye everyone, talk soon!