US Offering $45M For Huge Wind Energy Test Bed

coondoggie writes "On a day when one of the largest wind farm plans bit the dust, the US Department of Energy is offering up a five-year, $45 million grant to design and build a large dynamometer facility for testing 5 to 15 MW rated wind turbines and equipment. The DOE says such a facility is needed as the US has fallen behind other countries in the race to build ever-larger wind turbines for energy production. According to the DOE, the average size of wind turbines installed in the United States in 2007 increased to roughly 1.65 MW. Additionally, turbines already developed range in the 2.5 MW to 3.5 MW capacity sizes; with plans being developed for even greater power ratings. The larger wind turbines have outpaced the availability of US-based testing facilities, the DOE stated."

Comments

Go for it.

[Ethanol-fueled]

Build it, I say. That'll teach those birds to crap on my car!

Re:Go for it.

[XPeter]

Build it, I say. That'll teach those birds to crap on my car!

Your worried about bird crap? Oh man, your going to be in for a big surprise when a turbine hits the bird and it's guts are on your car. That's when the shit really starts to hit the fan.

Re:Go for it.

[j0hnyquest]

"For its part, the DOE has ambitious plans saying it expects wind to provide up to 20% of the nation's total electricity needs by 2030." There is no way that will happen. But this will definitely help with getting rid of those pesky birds and bats.

Re:Go for it.

[Anonymous Coward]

Well, here in Spain we already produce 20% of our electricity needs with wind, and it wasn't very hard or ultra-expensive:

https://demanda.ree.es/demanda.html

Re:Go for it.

[Gravatron]

Solution: Turbines disguised as ceiling fans over the congress chambers, to capture all the hot air.

This...

[incognito84]

This blows!

So if I understand this correctly...

[Romancer]

So if I understand this correctly...

We are looking for an artificial environment to test devices that specifically will be used in the natural unpredictable outdoor environment as their sole purpose?

Why not put them in a large windy area and map out their performance with actual gusty conditions and directional changes like they will be subject to in practice.

You'd get better data by skipping the artificial step.

If you really need the extremes to be on demand for destruction testing then put a big fan in front and a shroud around the device to be stress tested. Ramp it up and see how she performs.
Cost wise you could be selling all the energy that the time tests generate to pay for the spot testing and cleanup of the stress tests that fail.

Why do we need a giant test facility to create what's out there already and is the final place these things will be operating in anyway?

Re:So if I understand this correctly...

[girlintraining]

Why do we need a giant test facility to create what's out there already and is the final place these things will be operating in anyway?

a) To catch obvious design flaws early,
b) To test the device over the entire range of possible operation,
c) To provide a benchmark that remains static from one test to the next,
d) To control all external variables so as to create a consistent frame of reference,
e) To save a few bucks because it's really f----ing expensive to test every design as a full-scale prototype. ... Or to pull a page from our own industry, what's wrong with the following statement: "It compiles, ship it!"

Re:So if I understand this correctly...

[frosty_tsm]

You raise good points that I won't argue with, so I'll raise my own.

Most wind farms are in the middle of nowhere (for example, the one an hour outside the LA area on the I-10). Building a test location here requires a) the R&D staff relocate or b) the R&D staff drive potentially hours to work and then back. Neither of those is attractive to potential hires. If you were generous and decided to include their commute as paid time out of their 8 hour day, this could result in 4 or 6 hour work days. To offset the reduced productivity (in pure hours), the team needs to be doubled. Then there is more overhead (meetings, admin, recruiting, lower hiring standards, etc) due to the higher headcount. As painful as it might sound, $45M to build this in a population center might actually be cheaper.

Re:So if I understand this correctly...

[markk]

How much do you think it will cost to fully instrument an experimental Turbine in the field, then tear it down and build a different one? Now, how much for the equipment to stress the turbine at various loads, to manufacture wind speed conditions that mimic many different places around the country, and different loadings, look at various types of network interconnects... We might as well build a testbed location to do this. It might cost 40 or 50 million even eh?

Re:So if I understand this correctly...

[willy_me]

Why do we need a giant test facility to create what's out there already and is the final place these things will be operating in anyway?

A static environment is required to observe the effects of different designs. Tests in a real environment are also important - but they do not negate the need for a static test environment.

Re:So if I understand this correctly...

[Anonymous Coward]

Having just visited a wind farm today, I can tell you that the gearboxes are what fail most often. The wind facility I visited had 44 older wind turbines (starting from 1998) and those gearboxes cost around $150,000 to $200,000 plus a significant amount for the cranes and man hours ($1000/day for the large crane required + $10,000 setup and $10,000 tear down, and they can only operate in low wind conditions). Those gearboxes, however, are supposed to last, IIRC, 10 to 15 years, but typically last less than 5, simply because of the stresses caused by starting and stopping (according to the mechanic there).

The NWTC is putting in a test gearbox at the same site to collect data for (hopefully) a year, but I really don't know anything about that facility. I guess they don't have what the DOE is looking for here?

Re:So if I understand this correctly...

[tibman]

it WILL be tested in real life.. AFTER it's been tested in a smaller controlled environment. Half-assing stuff is building expensive systems and full-scale deploying them as a test phase. Guess what, if they work but have some problems.. that company won't be addressing those problems because they aren't worth the redeployment costs.

Also, a real-life environment won't go through the full range of capable scenarios during the limited test phase. You need to try out all kinds of odd-ball stuff that happens in real-life but just not very often (ie: hurricane).

Being able to install a prototype drive-train and go through the motions of testing without lengthy installation/setup times is important!

Re:So if I understand this correctly...

[Anonymous Coward]

Disclosure 1: I read the article.

Disclosure 2: I work at a wind turbine blade testing facility.

First point: the article speaks of the drive train, so the blades are not attached. The blades are *HUGE*, so that approach would be insanely expensive compared to the usual approach.
So, the usual approach is to test the parts separately, and to test the connections of the parts. When the parts can handle the load, and the connections can handle the load, and the support can handle the load, then the load can be safely transferred from the parts to the support.

a) The smart approach is to apply loads equivalent to that of the uncontrolled environment, increased with a safety factor. You can do that in the controlled artificial environment. This way, you abuse the parts more than what nature will deal out.

b) The tests should include fatigue tests. Since it is economically unfeasible to do a 20 year experiment, the testing frequency and amplitude are increased such that the cumulative damage in the critical spots is equivalent. Then the test is executed at a level augmented with a safety factor. So, yes, the varying aspect is tested.

c) I am not in the business of testing the drive train; however, depending on the construction of the turbine, the hard gusting winds from off angles that you speak of, are considered. I expect most of that load is carried by the bearings that carry the turbine hub, and that the remaining part is taken by the drive train. These loads are known, and thus can be tested. Moreover, risking the dyno as a test subject is kind of the purpose of the test: If it breaks then there is a flaw that needs addressing.

d) I agree that the test conditions are not the real world conditions. I do not agree that testing can be done only under real world conditions and that results obtained under testing conditions cannot tell anything about real world behaviour. The essential part here is the mapping between the real world and the test conditions. For instance, one could observe from material tests that doubling the load reduces the number of load cycles to failure by a factor 8 (I am pulling figures from thin air here to support my reasoning, however, one may look up the literature to find the proper figures. They'll support my reasoning). So, by doubling the load, we can test 8 times as fast. However, it is known that the UV light from the sun degrades the resin over time. So, in reality, the blade may not be as strong at the end of its life as the test specimen at the end of the test. These kind of influences must be taken into account. And again it is done with *rimshot* a safety factor.

e) As far as I know, every design is tested. Most of them by simulation, and some (including all designs going in production) of them by physical test.

The usual cycle is to do the engineering design, do simulations, improve, rinse, repeat, until a satisfactory design comes up. Then build an expensive prototype, which most likely passes the test, as the failure modes are known from the simulations and designed to be outside the operating envelope. This adds the following purpose:

f) Test the prototype to show it complies with the requirements.

One of the great experiences of my job is to see a blade fail at the predicted spot at the predicted load in the predicted way. Another is to see it fail in another way, and being able to spot what cause was missed. It is akin to the satisfaction of discovering and eliminating an important bug.

Re:So if I understand this correctly...

[TheRaven64]

There's a great video of a wind turbine exploding which you can probably find if you look. Once it went past a certain speed, the tensile strength of one of the blades was exceeded and it split. The turbine then became unbalanced and quickly pulled itself apart.

This turbine, if I remember correctly, had been in use for two year when it happened. It only broke because the winds were much higher than average for the area. If you're testing in a wind tunnel, you can keep turning up the wind speed until the turbine explodes and get an accurate measure of how much energy it produces at each wind speed and how much it can take so, when you deploy it, you can shut it down when the wind speed approaches the maximum. If you test it in the real world and 'skip the artificial step', you may need to wait several years to get wind speeds that high.

From your post, it seems like you've never designed anything for real-world deployment. You always want to control the test conditions so you can see exactly which variable is causing failures in your prototype.

Why?

[Dripdry]

IANAE (Engineer, yes) however I seem to recall the energy generation from wind turbines being a fairly simple function of the size. Although I understand there is an acreage issue is it truly necessary to develop bigger and bigger turbines? Can someone explain this? Is it simply that we should optimize the land useage?

Also, bring on the inevitable "ditch wind, go nuclear" stuff. I can has mod points now?

Re:Why?

[MichaelSmith]

Maybe they want to test wind turbines to destruction, or model their behaviour in different weather conditions. For example: how does ice deposition on turbine blades affect efficiency? Do this introduce any dangerous operational modes?

Re:Why?

[XPeter]

Also, bring on the inevitable "ditch wind, go nuclear" stuff. I can has mod points now?

IANAA (Adult, yes) Nuclear is much more efficient when compared to wind farms, but nuclear energy hasn't been developed enough for it to be used as a main energy source. There are many advances and safety precautions to be made before nuclear goes to the big leagues. Intermittently though. we need to drastically cut our addiction to oil and go after power like wind and solar.

Re:Why?

[morgan_greywolf]

IANAA (Adult, yes) Nuclear is much more efficient when compared to wind farms, but nuclear energy hasn't been developed enough for it to be used as a main energy source.

Someone should tell that to the French. Nuclear reactors provide more than 75% of France's power requirements..

Re:Why?

[maxume]

If you look closely, you will see that they also provide the U.S. more electrical generation than France has.

Re:Why?

[samkass]

And Brazil gets only 3% from nuclear, has only slightly less power requirements than France, and yet is largely independent of foreign oil, while France is not.

Re:Why?

[TubeSteak]

Maybe they want to test wind turbines to destruction, or model their behaviour in different weather conditions. For example: how does ice deposition on turbine blades affect efficiency? Do this introduce any dangerous operational modes?

A test chamber big enough to do all that sounds like the type of thing the military would be interested in. Why not build it at one of the Army Proving Grounds? The road/rail infrastructure already exists for handling extremely oversized loads and they operate wind tunnels.

Re:Why?

[MichaelSmith]

Maybe. Years ago when I worked on the FA18 there was a team working on crack growth analysis on (I think) the Mirage or the F111. They had stripped an airframe down in a test chamber in the US, cooled it down until the metal became brittle, then stressed it with hydraulic rams until cracks started to grow out of control. The resulting model was used to predict catastrophic failure in operational aircraft.

I can imagine similar things being done on turbines, for similar reasons.

Re:Why?

[T Murphy]

Brazil has sugarcane to produce ethanol where it cannot make enough oil. France is small, has little (no?) oil to produce for itself and no economical source for ethanol. The only way this has to do with nuclear is France's lack of plentiful domestic fossil fuel makes nuclear more attractive. While nuclear is generally more expensive (less attractive) than fossil fuel power, oil isn't a direct part of the equation. We (US) would probably have more nuclear plants if it weren't for the plentiful coal- I don't think that picture will change because of any fluctuation in our oil imports.

I find this hard to believe...

[Anonymous Coward]

I used to work for an aircraft engine company (Pratt & Whitney). They had lots of test cells for engine testing & research. This was a big heavy block of reinforced concrete with lots of instruments attached, and you bolt the engine to it.

I really doubt a wind turbine generates more power. I'm sure you could build one on the edge of a cliff so you don't need to worry about the wind turbine blades hitting something.

Re:I find this hard to believe...

[Anonymous Coward]

I suspect the power output is only a part of it. More important (as pointed out above);

- at what wind speed does it break down?
- if you run it at X high speed for Y hours, and small cracks form, how many more hours can you run it before it breaks? This is important e.g. if there's been a storm - how quickly do you need to send out a maintenance crew, or switch off the turbines?
- how is it affected by ice? how is it affected by flying plastic bags, or birds?
- if you want to compare 25 different designs for each of the above, how can you make sure they all experience the same input, so the output is comparable?

Testing in the wild is sensible but doesn't provide the quality of data needed.

Insert Taco Bell joke here

[crunchly]

Wait for it...

And

[Sycraft-fu]

So a gigantic blade doesn't go flying in to someone's house.

When you are talking machines as big and as heavy as this, you want to test outside conditions in a safe environment to make sure it won't fail. You do not want to discover later that oh, maybe it WASN'T as strong as we thought.

Same reason why the bend wings on an airplane. No, they will never face stresses that high in the real world. However, we don't want to just fly it around and say "ok, that's probably good" only to find out later that no, it really isn't. You test an outside case, and you do it somewhere that nobody gets hurt.

$45 million over 5 years?

[Animats]

It's so government.

I once worked in an R&D facility for heavy hydraulic equipment. They had about fifty test cells of different sizes, the largest of which was used for hydraulic transmissions for medium-sized locomotives. Those test setups used a big motor and a water-cooled brake; the hot water went through a cooling tower, and then to sprinklers in what appeared to be a decorative lake out front but was really a heat sink. That gear was in the 5MW range, somewhat smaller than what's being described here, but not a lot smaller.

That setup was where it belonged, near the engineers who designed the things and the machinists who built the prototypes. When the big test cell was put in, it took a few months to build. Not five years.

Dynamometer != Wind tunnel

[drewm1980]

Most of the posters seem to be under the false impression that this will be some huge wind tunnel facility. One of the difficult problems in designing a wind turbine is that the shaft turns very slowly, but electrical generators operate much more efficiently at higher shaft velocities. With the sort of dynamometer they are talking about, you use a very large motor to spin the generator (and possibly the attached drivetrain) and measure how its efficiency throughout its speed range.