Ep 51 Teton Dam Failure
Engineering News – CO2 Capture Membrane (0:50)
This week's engineering failure is the Teton Dam Failure (6:15). The dam catastrophically failed during its initial fill (17:15) leading to assessments of other similar dams (40:00).
Teton Dam Failure
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 a 6 pack of wine glasses from IKEA (and definitely some other stores), you can hear us talk about more interesting engineering failures!
This week in engineering news, a new membrane to better capture CO2.
Researchers at North Carolina State University developed new membrane technology to allow for more efficient carbon dioxide capture from mixed gases.
Carbon dioxide and nitrogen dioxide mixtures are fairly common in the greenhouse gas emissions of power plants. And carbon dioxide and methane are common in the natural gas industry.
The membrane is designed to remove the carbon dioxide from these mixed gases, which honestly sounds like a challenging task.
Another application is to scrub carbon dioxide from air inside submarines or other isolated vessels.
Membranes as a means to remove carbon dioxide are appealing due to scalability, and ease of replacement.
Another solution from membranes is chemical absorption which requires bubbling the mixed gas through a column but that has a larger footprint and can be toxic and corrosive. A membrane, assuming the manufacturing of the membrane is reasonable, can be a much greener solution.
Ok so initially reading this article, I assumed that the membrane was trapping the CO2, but it's actually capturing everything else and the CO2 is what comes out the other side and is then captured.
One of the challenges is permeability vs selectivity. If you have a really selective membrane that captures everything you want it to, it likely takes longer for the CO2 to pass through it. Whereas the more permeable the membrane, the less it captures. It’s a fine balance.
If you want to read more on the new membrane technology, check out the link on the web page for this episode at failurology.ca.
Now on to this week’s engineering failure; the Teton Dam.
Suffered catastrophic failure on June 5, 1976 as it was being filled for the first time. The dam failure killed 11 people, 16,000 livestock and cost $300 million in claims, with the total damage expected to be up to $2 billion. The failure was so bad, they never rebuilt it.
The area downstream of the dam suffered a severe drought in 1961 and then severe flooding in 1962, so a dam to manage spring runoff was a really good idea and made a lot of sense. That said, it’s only one side of the story and there was a lot of opposition from environmental and conservation groups regarding impact on fishery and other wildlife habitats. There was also a concern of economic return on investment.
Construction of the Teton Dam started in 1972 and took four years and cost $100 million.
The dam was built by the Bureau of Reclamation, one of eight federal agencies that are authorised to build dams. The Folsom Dam that we covered in episode 26 was built by the Army Corps of Engineers before being turned over to the Bureau of Reclamation for operation and maintenance.
Earthen dam on the Teton River in Idaho, located about 1000km drive south from Calgary, Alberta or 2,400 km from Calgary, Texas. And a long airplane ride from Calgary, Scotland. The purpose of the dam was for flood mitigation and irrigation as well as hydroelectric power. The dam was 520m wide at its base, 940m long and 93m tall. It created the Teton Reservoir with a total capacity of 355 million cubic metres or just over 142 thousand olympic sized swimming pools.
The dam site was made up of basalt and rhyolite, both of which are highly permeable and less than ideal for dam construction. The test cores also revealed fissures, especially on the west side (where the dam eventually failed) that the bureau planned to seal by injecting grout at high pressure to create what's called a grout curtain. During the excavation process they found a lot of fissures, some of which were so large that a person could walk inside. This is probably where they should have stopped construction or at least significantly revised their plan. Some of the fissures were grouted and some were deemed beyond the boundaries required for grout. The total grout was about 36,000 linear metres of drilled holes and double the amount of grout expected.
As well, the dam was located in an area of reasonable active seismic activity. Five earthquakes had occurred within 50km of the dam site in the previous five years.
The dam failure was actually so eminent that they strategically placed cameras around the dam to document the process. Which is why the failure is so well documented.
After completing the dam in November 1975, they started filling it, as one does. The standard rate of filling was 300mm a day. But after heavy snow over the winter, they doubled the filling rate in the spring to accommodate the additional runoff. During the filling period, workers continued to inspect for leaks. The filling rate was doubled again to 1.2m per day 6 months after filling started.
Let’s pause a second to talk about water. It’s a powerful, powerful thing. Like in an extremely impressive way. In a reservoir type setting, the pressure exerted on the dam at the surface of the water would be neutral. There would be a dynamic impact from the water coming into the reservoir, but let’s consider it static for a second. So the pressure at the top is 0. The deeper and deeper you go, the pressure starts to increase. And we call this head. I have never built or designed a dam, but for piping within a building, the head is measured by the vertical column of water that sits in the piping. So in a 5 storey building that is 15m tall, the pressure exerted on the fittings on the bottom, when all of the pumps are off, is 15m. Now remember the Teton dam is 93m tall, so the pressure exerted on the bottom of the dam is 93m. Probably more once you factor in dynamic forces of currents and bypasses.
So we talked about the fissures and how the owners should have probably stopped there, but I think there is another huge, huge oversight that was made. When they started filling the dam, the main outlet works and spillway gates WERE NOT YET IN SERVICE. Those gates were blocked off with steel walls while they were painted. There were emergency outlets, but they were only sized for 24 cubic metres per second.
Over the course of June 3rd and 4th, three small springs started downstream of the dam on the west side. I don’t know how, but they seemed surprised that this was happening. And even though the dam was almost at capacity at 73m deep, it doesn't appear that they thought there was a problem or that they started releasing water to lower the level in the dam below the leak.
By 730am on June 5, 1976, the leak was discharging 0.57-0.85 cubic metres per second. They used bulldozers to plug the leak, but were unsuccessful.
By 11:15am evacuations started downstream and work crews were forced to flee as the leak was now larger than a swimming pool and it started to swallow their equipment. In fact, two bulldozer operators were pulled to safety with ropes.
By 11:55am, the crest of the dam sagged and collapsed and two minutes later, the west third of the dam wall disintegrated releasing over 57,000 cubic metres of water per second. This is almost 3x the flow of Niagara Falls. And I'm talking about the Canadian Horseshoe falls; the good ones.
By 8pm, the dam was empty.
Investigators believed that the permeable materials and general flaws in the earth filled dam allowed water to seep through and cause internal erosion that eventually washed out the dam.
An investigating panel was unable to determine if the erosion was caused by the flow of water under highly erodible and unprotected fill or the hydraulic fracturing of the core material.
“The fundamental cause of the failure may be regarded as a combination of geological factors and design decisions that, taken together, permitted the failure to develop.”
The Teton Dam failure caused the Bureau of Reclamation to assess all other similar dams which identified two other dangerous dams.
The Fontelle dam which was filled very similarly to the Teton Dam. In May 1985 they drained the entire dam and reinforced an area of seepage to address a long running leak.
The Jackson Lake Dam which was susceptible to failure from an earthquake with a magnitude of 5.5 or greater. The dam was upgraded to withstand a 7.5 magnitude earthquake in 1986-1989.
This reactionary process has been the norm in other dam failures. The Folsom dam gate failure that we covered in episode 26 initiated a safety review and repair of several other dams throughout the U.S. and the Tous Dam failure in Spain also initiated dam reviews and repairs around the world. We covered the Tous Dam in a mini failure on our patreon page.
The entirely preventable (at multiple stages) dam failure impacted thousands of people and livestock throughout the Teton Canyon about 10km downstream of the dam site.
The American Falls Dam downstream of Teton had a bit of time to prepare; by a bit of time, we mean hours not days. They increased discharge by less than 5% before the flood arrived. Which doesn’t sound like enough, but the dam held and stopped the flood from progressing.
So there you have it, the Teton Dam. Yet another dam failure that was completely preventable. This failure demonstrates what happens when you completely disregard all of the evidence in front of you that says “this is a bad idea”. Trust your spidey sense folks, it knows what it's doing.
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 into the next episode where we will talk about the Luna Park Ghost Ride fire in Sydney Australia. Inadequate fire protection systems and a very combustible building allowed a fire to spread quickly throughout the ride.
Bye everyone, talk soon!