Tricky wind turbine question 5

[Windar]

Hello everyone,

I have a difficult question regarding wind turbines that I was hoping to get some feedback on.

Situation: There are a number of ~90 kW turbines mounted on 60-foot lattice towers. Due to the design of the machine, it occasionally happens that the turbine will go into what's called runaway mode. This means that there is no effective way to stop the turbine blades from spinning faster than they should be. We are looking at ways to bring a runaway turbine to a stop.

First of all, I realize the common answer is: you don't do anything. However, assuming that something HAD to be done, do people have suggestions for ways to bring an out of control wind turbine to a stop. The rotor diameter is about 17 meters, three-bladed, mounted on a 60' tower.

Any action taken would have to be done from the upwind side because these are downwind machines.

Thanks!

 

[Eltron]

Sounds like you need a better yaw control mechanism.

Dan

http://www.eltronresearch.com

Dan's Blog

 

[rb1957]

a different angle ... it sounds like you've experienced over-speeds already. how do you verify that the structure is still sound ?

 

[1gibson]

Brake that is actuated by a spring loaded mechanism, employed when centrifugal force exceeds a certain value (speed exceeds a rated value). Think the opposite of a non-reverse ratchet on an electric motor.

Retracts automatically as speed is reduced, no controls required. Comments above indicated a disc brake could already exist, although the concept might be more straight forward to apply to a drum brake.

 

[MiketheEngineer]

Can you swivel the blades around 180 degrees so they are now pointed opposite the wind??

That would stop them - but then they would also start spinning in opposite direction.

Some of these hubs are 200-300' in the air. Not too many water cannons and such can reach such a height.

 

[ivymike]

these ones happen to be on 60' towers, per the post.

 

[moon161]

Turn it off of the wind. If you have active yaw control, you should have redundant wind sensing and yaw control.

If it pivots freely, Bergey, I think uses a steering vane that deflects in an overspeed situation, causing the rotor to turn off of the wind.

 

[verymadmac]

You could try mounting a drag parachute in each blade tip that would deploy at a set centrifugal force.

 

[tbuelna]

Here's the wind turbine industry's standard approach to the problem:

http://svendborg-brakes.com/en/products/rotor-lock.aspx

 

[ScottyUK]

DC injection brake? Possible, if the generator design is favourable.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[berkshire]

Windar (Mechanical)
There used to be a system called "reefing" a wind mill, AKA wind turbine, where the yaw control turned the rotor 90 degrees to the wind.
Is your system not capable of doing this?

B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[MikeHalloran]

I think the challenge presented assumes that _all_ of the controls have "gone off to see God", and are therefore unresponsive to external stimuli, and not behaving as they should.



Mike Halloran
Pembroke Pines, FL, USA

 

[Windar]

MikeHalloran (Mechanical)
26 Sep 11 9:51
I think the challenge presented assumes that _all_ of the controls have "gone off to see God", and are therefore unresponsive to external stimuli, and not behaving as they should.

This is correct. Significant retrofits to the turbine could obviously address the issue, but if this is not possible, what about when the worst case occurs?

What about a large net propulsion system? (the blades are fairly easily replaced if they were to fail).

 

[rb1957]

i don't think any system is designed that way ... not aircraft, not missiles, not nuclear power stations.

if all your controls are assumed to have failed then the machine becomes uncontrollable.

rational design works to levels of probability of failure, with redundant systems of control.

 

[Tmoose]

I think when the first blade is lost at high rpm the fact that blades are easy to replace may not be all that reassuring.

 

[rb1957]

unbolt the tower base, install new tower, install new blades ...
oh, and pick the pieces of broken wind turbine scattered about.

 

[btrueblood]

...and call the insurance company to handle claims from homeowners 1 km away who have a blade slice through the roof.

see

http://www.caithnesswindfarms.co.uk/accidents.pdf

"Pieces of blade are documented as travelling up to 1300 meters. In Germany, blade pieces have gone through the roofs and walls of nearby buildings."

Seriously, you'd need to run some kind of testing to verify that the various blade-braking schemes suggested would not just become blade-breaking schemes. The blades aren't really designed to take any impacts more significant than a frozen chicken when at operating speed. Even the reefing ideas would have issues with slew rate and gyroscopic forces if not controlled precisely.

A more reasonable method might be a means of erecting a series of poles/masts or a more-or-less solid "wall" upstream of the troubled rotor (I'm thinking of a D-8 cat as an anchor, and a couple of big cargo parachutes), in an effort to block all or most of the wind, allowing the rotor to spin down to a lower, safer speed before attempting some method of blade breaking/braking. At least the lower speed will reduce the ballistic distance of the blades...

 

[rb1957]

"and what is the flight speed of a frozen chicken?"

actually they use thawed chickens, since frozen are too solid.

and i wouldn't've thought bird strike was much of a design criteria. a bird, even a 12ld-er, impacting at 15ft/sec (10mph), should be pretty easy to absorb ... at the root the blade is very rigid, at the tip very flexible.

maybe a design requirement is to safely depower if a blade is shed

 

[MikeHalloran]

The tip speed can be much greater than 10mph, and they don't make a lot of noise, which is why the birds don't always steer clear of the blade disc.

;--

The problem I see with physically snagging the blades with nets, ropes or parachutes, is twofold.

If the rope wraps around the hub, there may be enough inertia and mechanical advantage to lift and maybe even swing a Cat D8 as if launched from a trebuchet. Modern ropes can do amazing things like that.

If the rope managed to snap off a blade, I'd expect the resulting imbalance to take out the tower pretty quickly. Maybe the towers are designed for a missing blade imbalance at above design speed; they don't look like it to me.

;--

Deploying multiple parachutes or a tall screen upwind might slow the rotor a bit, but you don't want to accidentally snag the spinning rotor. You still face the problem of stopping the rotor, but intervention becomes a little less dangerous.

;--

IF the nacelle is free to swing during control failure (not revealed so far), then pre-deploying a weighted rope from its upwind end might allow one to yaw the rotor away from the wind, with the Cat D8.



Mike Halloran
Pembroke Pines, FL, USA

 

[rb1957]

tip speed sure, but bird strike is governed by forward speed (hence the need to accelerate the bird in a cannon, to get it up the speed of the airplane ... in real life, the plane flies into the bird).

i was thinking of a bird flying into the plane of the turbine, bending about the weak axis of the blade. thinking more, the bird could be unlucky enough to be in the prop-plane at the wrong time whihc would resemble a more typical bird strike on the leading edge of the wing. still the tip speed would be much less than an airplane's forward speed, the L/E radius of the balde would be much smaller (for the blade) ... factors which decrease the impact loads, but the section is much smaller than a wing which would increase the effects.

 

[1gibson]

Mechanical tachometer that triggers an emergency brake? And I'm still partial to the idea of a brake actuated by centrifugal force.

I can't tell if some of these suggestions are jokes, or if I missed part of the original post about keeping the machine in one piece not being a requirement.

That's not entirely true, the "strongly worded note on the drawing" was obviously a joke. But some of the others still have me confused.