I lost my new blades to a wind storm.

[bigrockcandymountain]

Well, I've been slow to post about this but my nice new blades are no more.  The turbine ran away in a really fierce winter storm and they all broke off at the roots.  It ran for awhile making tons of power, but they started making bad noises and i knew i should go shut it down.

That means climbing the hill and crawling to the base of the tower, all while being beat on by 100km/h winds and -30c.  I was sick that day and feeling like crap, and i put it off too long.  Anyway, here is a picture of what is left.

I tipped it down today, and put the old rotor back on.  I tipped it back up, but then noticed the yaw bearing must be jammed sideways in the yaw tube.  The head is sitting up higher than it should and not wanting to yaw at all.

I think the head jumped when the blades let go and jammed the yaw bearing.  Anyway, i guess I'll order a new yaw bearing.  The head is a pain to slide off to change the bearing, mostly because of the terrain. It was way too icy up there today to attempt it.  Hopefully it will bare off a bit and i can get it fixed up.

I didn't dial the motor shaft, but it seemed smooth and undamaged. 

Back to the drawing board to build a new set of blades again.  It's a good thing that carving blades is fun.  I put a lot of work into getting a very nice finish on the last set.  I'm quite sad that they didn't last longer than they did. 

I'm not exactly sure what was keeping them from furling and being ok.  I'll have to think on that one.  My best guess is what adriaan calls the "self orientating moment" 
Another factor might be that the yaw tube is full of grease and it would have been very stiff to yaw in that temperature.  Stiff to yaw means stiff to furl. 

Hopefully i can get this back running in a couple weeks .  I hate having no turbine, especially in the winter.

[kitestrings]

Sorry to hear and see this.  I know you put a lot of work into the new rotor.  I suspect you are on to the likely main causes.  That, and simply Mother Nature is flat out brutal at times.  It's a wonder anything survives.

We shut ours down just last week when storm "Elliott" rolled in.  In our area, which was not the worst, we had peak gusts of nearly 70 mph.  We survived it fine, but I'm glad we weren't running.  We had about 75,000 folks without power across the state, just in time for the holidays, and then the temps plummeted for the next 5-days.

Good luck with the repairs but do be careful.  Sounds like challenging conditions.  ~ks

[Adriaan Kragten]

This result proves that the safety system isn't working properly. The safety system should limit the rotational speed and thrust to a safe value at any wind speed. I expect that your wind turbine uses the inclined hinge main vane safety system which is described in my public report KD 431. You should change the construction such that the rotor turns enough out of the wind at high wind speeds. The moment equation around the hinge axis shows that this is realized if the vane becomes lighter or if the angle in between the vane axis and the vertical becomes smaller. You should also check if the eccentricity e is large enough for the chosen rotor diameter D (e must be at least 0.08 * D for a fast running rotor). If e is too small, the so called self orientating moment will keep the rotor too long in the wind.

The fact that the rotor becomes very noisy at a high rotational speed can be an indication of flutter. Flutter is a combined torsion and bending vibration which is mainly caused by a too low torsion stiffness. The bending and torsion stresses are very high during flutter and therefore a blade won't live long. The torsion stiffness can be increased if a wooden blade is covered with a layer of glass vibre imbedded in epoxy or polyester.

[mbouwer]

@ Bigrockcandymountain,

Shall I make you 3 new blade roots of thin steel sheet so that you can use the same blades again?

[bigrockcandymountain]

Adriaan, it is not a inclined hinge system but a simple side hinged design with tension springs.

The offset is not 08(D).  It is approximately .05(D). That is probably the biggest factor.  I'll look in to changing it.  It won't be easy though.

I also agree that flutter was occuring at high speeds.  Would an airfoil with a smaller chord at the tip and higher thickness help with this problem?  The blades were a 17% thickness ratio at the tip.

mbouwer, thanks for the very generous offer.  Unfortunately, the blades were all shattered and there isn't much left except toothpicks.

[Mary B]

Blades hit the tower then. You need to do a full inspection of everything up top to make sure the force didn't bend anything...

[mbouwer]

@ Bigrockcandymountain,
Since you have such an ideal site for a windmill and can use the energy so well,
I would like to support you to get the stuff running again.

[DanG]

I congratulate you on no heroics, coulda lost the both of youse…

[Adriaan Kragten]

Adriaan, it is not a inclined hinge system but a simple side hinged design with tension springs.

The offset is not 08(D).  It is approximately .05(D). That is probably the biggest factor.  I'll look in to changing it.  It won't be easy though.

I also agree that flutter was occuring at high speeds.  Would an airfoil with a smaller chord at the tip and higher thickness help with this problem?  The blades were a 17% thickness ratio at the tip.

mbouwer, thanks for the very generous offer.  Unfortunately, the blades were all shattered and there isn't much left except toothpicks.

In figure 4 of my public report KD 213, I have compared the moment supplied by the thrust to the self orientating moment for a ratio e / D = 0.1 (both moments are made dimensionless to cancel the influence of the wind speed and the geometry). The thrust moment is maximal for a yaw angle delta = 0° and decreases according to a cos^2 function. The self orientating moment is maximal for a yaw angle of 30°. The final rotor moment is the thrust moment minus the self orientating moment. In this figure it can be seen that even for a ratio e / D = 0.1, the self orientating moment has a rather large influence on the final rotor moment. If you use a rato e / D = 0.05, the influence of the self orientating moment becomes twice as strong and this means that the final rotor moment will decrease strongly at increasing yaw angle. So the rotor will turn out of the wind but this will require a strong increase in wind speed. If the ratio e / D is chosen 0.02, the line for the self orientating moment increases by a factor 5 and it will intersect with the line for the thrust moment. This means that the yaw angle will never be larger than 30°. Long ago a Belgium guy has built a wind turbine with such a small ratio of e / D and his rotor didn't turn out of the wind.

If you use a constant chord blade, the torsion stiffness and bending stiffness are also constant and rather low for slender blades made out of massive wood. If you use a tapered blade, the torsion stiffness and bending stiffness are increasing at decreasing radius r. If you compare blades of the same design tip speed ratio, the same tip radius R and the same airfoil, both blades will have about the same chord half way the blade length. So both stiffnesses are lower for the outside part of a tapered blade and higher for the inside part of a tapered blade. The root of your blade is clamped in between a rather stiff hub assembly and so I expect that the inner part if the blade is almost not twisting if the blade is fluttering. So increasing the chord at the blade root and decreasing the chord at the blade tip may even strengthen the sensibilty for flutter. Increasing of the torsion stiffness of the outer part of the blade has the strongest positive effect, also because the aerodynamic forces on this part of the blade are strongest.

[SparWeb]

Oh no!
It's a terrible feeling (I know too well)

[SparWeb]

My small suggestion is to ensure some "nose-up" angle in mounting the generator, which keeps the blade tips away from the tower.  If you already did then, then a moot point.
Torsion and flutter are relevant, for sure, but flexing can't be ignored.  Besides, you can tes them all at once.  Hugh P. gave detailed instructions, back in the day, and maybe still can be found on his site.

[bigrockcandymountain]

I congratulate you on no heroics, coulda lost the both of youse…

The older I get, the more I realize that heriocs are usually a bad idea.

I don't think the blades hit the tower.  All the evidence suggests flutter causing failure at the blade root, and then tower contact after root failure.  Evidence to support this:
1.  The blade roots were indented from the hub on both the back side and the front side, suggesting flutter.

2.  One blade was found mostly intact, and there is no scraping at the tip where tower contact from flexing would have happened.  I know what this looks like because my first set of blades flexed and hit the tower.  They had a very defined scrape at the tips. 

After those first set contacted the tower, i made wedges and gave the generator a nose up angle for better clearance. 

I guess now I just need to decide if moving the generator farther out to increase the offset is worth the trouble, or if I should rely on my manual furling like i have been doing. 

I also think there is an assymetric thrust component to all of this. I changed the rotation on this set of blades from the last ones.  I think the next set will go back to the clockwise rotation of the set that were the most successful. 

Right now, my plan is to build another set of blades very similar to that successful set, but better balanced and finished.  The balance and noise were my main complaints with that set, and that is easily fixed. 

I'm also going to extend the hub about 4" (100mm) farther out along the blades.  I'll just weld on some steel plate ears.  I always thought that hub was a little small, and I'll feel better with a bit better support. 

A new yaw bearing is in the mail.  I'll machine an adaptor for the tower top to hold it in place better.  Right now its just kind of sitting there.  I've never really been happy with that setup, so now is my chance to make it right. 

[kitestrings]

We all get to Monday morning quarterback things, while you get to do all the hard work, right, but if you are looking at some changes, perhaps a furling actuator?  Crawling out to the tower in the middle of a storm is not ideal.  I didn't like this approach either.  Our solution was a ~$150 furling actuator at the base of the tower.  One of the best things we ever did.

I believe you are also using the MS Classic controller, is that correct?  In our case, we used the non-PWM relay, Aux 1 for a simple close or open contact signal to the actuator.  You can start or stop it manually, from the house, I do that almost every night.  By using PV V on Hi, you can also program the voltage and duration it sits there at which point you want an automated shutdown.  This works well when we're away, or just when when there are really turbulent winds.  Maybe something to consider?

[MagnetJuice]

I have done a lot of reading on ways to stop a wind turbine. Shorting the phases, whether shorting all three phases at once or in two steps, using a low ohm resistor before shorting all three phases, works most of the time. Sometimes it doesn't stop the turbine.

It depends on many things, how ferocious are the winds, the efficiency of the alternator, the weight of the blades, etc. etc.

kitestrings linear actuator is one of the best foolproof ways I have seen.

Here is a link to his thread:

https://www.fieldlines.com/index.php?topic=149230.0

It is a little more work than a simple brake switch, but it can save the turbine from destruction.

Ed

[kitestrings]

I'd meant to ask: I wonder if the yaw bearing issue could have been developing earlier, and perhaps contributed to this event?

Regarding offset.  In his "Wind Power Workshop", Hugh suggests 4% as the minimum offset "as a rule of thumb".  I think I've seen multiple sources cited here that suggest 4-6%.  Ours is 9.5" (241 mm) on a 15' (4.6 m) turbine, so just over 5%.  Hugh commented on the difficulty of calculating furling in this post some time back.  I do like, and used, his approach - balancing the calculated thrust and restoring moments - which seems to better account for the length, weight and position of the materials.  I found it interesting that he had refined his calculation a bit based on real-world experience.  See:

https://www.fieldlines.com/index.php/topic,146065.msg997305.html#msg997305

[SparWeb]

Mind if I ask about a few more dimensions?

With a diameter of 13 feet, then the "rule of thumb" wants the chord to be 13 inches.
With a 13" chord, then a 15% thick airfoil would be 2" thick.
Starting with a 2x8 board wouldn't allow it to be that thick once finished.
Was it a rough on-size or a S4S board only 1-3/4" thick to start with?

The thickness of the blade is what resists torsion, so you need this much airfoil thickness to tolerate high-speed strong winds.

Any taste for laminating two boards together to gain thickness?

[bigrockcandymountain]

Ok, I have been reading responses here and absorbing the knowledge and weighing it against time and money constraints as well as effort required.

Lets start with adding an electronic actuator to the manual furling.  That would be my number one upgrade and something that has been on the to do list for a long time.  So why haven't i made it happen?
The turbine base is 400' from the house.  Actuators are only readily available in 12v and 24v.  The 24v ones draw about 2a.  Even at that low amperage, i think i should run a 12/2 cable to keep the voltage drop within reason.  I also would need to trench this in by hand.  Between the $500 for wire and the no fun install, I just haven't got excited. 

I have thought about a wireless remote with an old 12v battery and a small solar panel up at the base.  All the cheap remote controls don't have near the range to work.

Maybe someone can tell me that an actuator will work just fine with 18 awg cable at 400'.  That would help.  And yes, i still have one unused aux1 on the wind classic that could control the furling. There is an old actuator on a big satellite dish in the junk pile.  No stickers left on it.  Would it be 120vac somehow or would there have been a low voltage dc in the dish reciever to control it? 

Kitestrings, thank you very much for the suggestions and especially for the data points on offset of your turbine and hugh piggot's etc.  I think I'll leave my offset where it is.  The yaw bearing issue could have been a contributing factor to this.  I'm not 100% sure.

Ok sparweb, I know I should just bite the bullet and build some proper laminated, thicker blades.  I'm kinda lazy and cheap and that's a deadly combination that usually causes one to end up working harder and spending more money in the end.  Hmm.

All 3 sets of blades so far have started as 2x8 planed spruce so 1-1/2" × 7-1/4".
They have all been full 7-1/4" at 12" radius tapering down to 4-1/2" at the tips.  So the thickness at 15% is about 5/8" at the tip.  I usually do a bit thicker than 15% so about 3/4" to 7/8" at the tip, so more of a 20% thickness ratio.

My plan right now is this. 

Put up my old set of blades that are badly balanced and loud.  They are already up there, i just need to fix the yaw bearing and get it stood back up.  This will buy me some time.  It sucks having no turbine this time of year, so whatever it takes to get it back producing quickly.

Once that's done, I'll carve a new set of blades and think about other improvements.  I think the new set of blades should be almost identical to the old set that are going back up, just better finished and balanced.  Those blades worked for a long time and my only big complaints were the noise and rattle from being unbalanced. 

[joestue]

run the cheapest wire you trust and juice it with however much voltage you need to push 2 (or whatever its rated for) amps through the motor. it won't burn out.

don't bother with a limit switch. get a 4 amp circuit breaker. apply juice till the breaker trips, it will be fully actuated.

[MattM]

The shear scale of your rotor made those old blades look paper thin.  Mother nature sure did a number on it.

If only there was a way to adhere foam to the new blade to increase chord without the weight penalty of going thicker blades.  That extra weight just scales the forces through the roof as rotational speeds increase, so I'm sure wind direction changes twist those blades more than they are simply flexing.  Would be so much easier to shape foam than wood, and you could protect the leading edge with metal tape just the same.  But that twisting under extreme force would be greatly reduced.

[Adriaan Kragten]

I have described the quasi stationary behaviour of different safety systems (see report KD 485). Quasi stationary means that it is assumed that the wind speed varies only slowly and that the reaction of the safety system is that slowly that the dynamic effects can be neglected. For most safety sytems it then can be predicted how the vane will move with respect to the head frame and how the head will turn out of the wind. However, the real situation can deviate very strongly from a quasi stationary behaviour and systems which move stable under quasi stationary conditions can become instable at strong fluctuations of the wind speed and the wind direction.

When I worked at the University of Eindhoven, I have designed a small 1 m diameter windmill called the WESP which was equiped with the inclined hinge main vane safety system and which had an eccentricity of 10 % of the rotor diameter. The generator was a hub dynamo of Sturmey Archer used at 12 V DC after rectification. The main vane was pointing upwards under an angle of 45° and was designed such that it could rotate 360° without touching the rotor. The head bearing and the vane bearing had only very little friction. The windmill was placed on the top of a 10 m high building. If the wind speed increased slowly, the dynamic behaviour of the system was fluently. However, during heavy wind gusts, the rotor turned out of the wind very fast and then it swinged back also very fast. So sometimes the rotor was perpendicular to the wind for a short time even at high wind speeds. This made that the tower was shaking heavily. This problem was mainly caused by the low moment of inertia of the head around the tower axis. This moment of inertia was increased by adding a iron rod to the head frame in parallel to the rotor plane and by putting a weight at each end of the rod. So the moment of inertia of the head was increased strongly and this prevented fast swinging of the head. Now the system was stable, also during heavy wind gusts.

Several of these WESP windmills have been built in the Cape Verdian Islands where the wind speeds are very high and they worked nicely. But I found it a disadvantage that extra material has to be added to the head frame to increase the moment of inertia. Therefore I have designed the new hinged side vane safety system for which the main vane is a part of the head frame. The only moving part is the vane blade. This vane blade reacts very fast on wind gusts but the whole head reacts only slowly on variations of the changed position of the vane blade. This prevents oscillations of the head and it reduces the maximum rotor thrust but also the gyroscopic moment in the blades and the rotor shaft. Recently I have designed a similar wind turbine but now with a hub dynamo of Nexus and with the hinged side vane safety system (see manuals VIRYA-0.98 and VIRYA-1.04)

If the head bearings or the vane hinge bearings have a lot of friction, it will also flatten fast movements of the head and the vane arm but it will also result in the wrong stationary position of the rotor with respect to the wind direction.

You may think that your blades are broken because of a too high thrust but it might also be because of a too high gyroscopic moment. The gyroscopic moment in a blade is proportional to the product of the angular velocity of the rotor, the angular velocity of the head and the moment of inertia of a blade around the rotor shaft. The gyroscopic moment can easily become larger than the bending moment of the thrust. 

[mab]

I think the 'skywaller' (USA brand?) dish actuators are 36v, although not particularly well made IME.

+1 what Joestue said: use a thin wire and run with extra volts - you're not using it often/long enough for poor energy efficiency to be an issue. Although i would personally stick with a limit switch unless i could find a breaker that would reliably trip quickly before the motor cooked - with 500' of thin wire the stalled current may not be much higher than the running current.

[bigrockcandymountain]

I pulled apart that dish actuator today and found it is marked 36v.  It is made in usa, but not super high quality.  I think it will work though.  It has a nice adjustable limit switch that i think is a good thing.

I'll probably run 48v to it because that's what my system is.  Ill use whatever wire i can find in the 14 awg range. 

I will run it in a conduit of "siphon string" and bury it a few inches deep. 

Siphon string is free around here.  I think i have about a mile of the stuff.  It is 1-1/4" od 1/2" id hdpe.  Very stiff and nasty to work with.  It should last a couple thousand years as conduit though.  It comes from shallow gas wells. 

It's looking like this project will get done after all.  That's what i love about fieldlines.  You guys give me the push i need sometimes.  I'm even thinking i might do it right away and just lay it on the ground.  I can bury it in the spring when the ground thaws.

[bigrockcandymountain]

You may think that your blades are broken because of a too high thrust but it might also be because of a too high gyroscopic moment.

That is a possibility I wouldn't have thought of.  It kind of makes sense though.  They all failed at the same time, or within a few seconds i think. 

[Mary B]

Option 2: use a battery at the tower(actuator will run fie at 24 volts or even 12, just slower and less force. You don't need 2,000 pounds of force those actuators could push!) Mount it motor up so it pushes down pulling your furling cable. That will keep water out. Add a relay up top and run small control cable down to the house. There are 433mhz relay systems that would have that range line of sight, to many trees and it would block the signal though! The actuator should have built in limit switches, before trying to adjust one look  it up online, they are a major pain in the ***!

If you want a 24 volt actuator I have several 18"...

[Adriaan Kragten]

You may think that your blades are broken because of a too high thrust but it might also be because of a too high gyroscopic moment.

That is a possibility I wouldn't have thought of.  It kind of makes sense though.  They all failed at the same time, or within a few seconds i think.

If one blade fails, the rotor gets a terrible imbalance. I have seen a tower jumping up and down because of this imbalance. This tower pipe had only a flat sheet at the bottom and was pushed downwards by its weight and by the pulling force in the guy wires. The rotor had two blades made out of Roofmate, glass fiber and epoxy. The next blade failed within seconds after the first one failed. The broken blades were that light that they whirled downwards like leaves. A similar but stronger rotor is described in report KD 532 but such a rotor should never be used without a proper safety system.

[kitestrings]

We used one of these with the "dual potentiometer control":
https://www.progressiveautomations.com/products/linear-actuator-with-potentiometer?variant=18277322326083

Ours is 12V, 150# and 18" stroke IIRC.  Our tower was quite a bit closer so that helped, but the route is a bit circuitous so we ended up putting a battery in the shop, but closer to the tower base.  I had considered, however, using a rechargeable 12V (14 or 18V was what I look at) Li-Ion pack like used on a cordless drill.  If you could knock that off somehow, then you only need a signal wire.  What I like with this control & actuator is that there is a simple "ignition" wire.  Remove 12V from it and the thing moves to pre-set1; apply 12V and it moves to pre-set2.  The keypad lets you set the limits with no programming skills, and there is an "auto-retract" safety if for any reason the thing meets more resistance than it should.

This is a typical morning start up from the house; a shut-down is just the reverse of this:
https://www.youtube.com/watch?v=tecGNiydJd8

[bigrockcandymountain]

This is the actuator i will try to use. It is made by JGS Engineering, St. Louis MO  It is 36v and about a 20" stroke.  I think 14 awg wire and 48v feed should be ok.  Thanks for the offer Mary, but I'll give this one a try first.


I also started carving blades yesterday.  I got the front flats finished.  Now i need to thickness the tips down and make the airfoil on the backside.  Maybe another 3 hours or so. 

I'm still waiting on a yaw bearing and for the ice to clear off the hill a bit. It has been pretty sunny here the last 2 weeks, but it still sucks with no turbine.  There are lots of windy evenings that i would normally go work in the shop.  With no generating happening, i don't really like to beat up the batteries. They are starting to show their age, and I would like to limp them along for another 3 years if possible.  That would be 10 years, which is pretty good for a family of 5 on 8x L16s. 

[MagnetJuice]

WOW BRCM, you are moving fast! Like they say, necessity is the mother of invention.

With 48 volts, that motor should get moving, even in cold weather. Make sure no water gets inside and keep all connections tight. Myself, I like to solder my connections, I don't trust spade lugs for long-term use.

It looks like your shop is like my basement; I can find anything down there, like an old fashion hardware store. 

Ed

[bigrockcandymountain]

I finished the tower top bushing.  It turned out pretty acceptable.  The bearing showed up, and the weather was nice today, so we put it all together with the old rotor and stood it back up. 


Now my progress will slow down a bit.  It is definitely a good motivator to rely on it for power. 

And ya, my shop fills up with projects and junk at a rate that almost defies the laws of the universe.  Every horizontal surface accumulates things. 

[DamonHD]

That junk thing is the second strongest force in the universe, the strongest being the attraction of tomato in any dish at a buffet to any white items you are wearing.  Even if you don't eat the tomato thing, or go near it.

None of this Higgs Boson footling stuff (I used to pass his class while he was teaching at uni, BTW)!

Rgds

Damon

[SparWeb]

...Oh, those light cuts when the chuck has nothing but a wee nub to hold on to....

Git'R Done: That's a quick turn-around!  I'd still be fiddling with it by the time the snow melted. 

[kitestrings]

Wow, you're knocking it out.  Hopefully this bridges you thru the winter anyway.  I'll be curious to hear how the furling is working - or if there's any change - with the new bearing.

I can relate to the lack of free horizontal space; it's magnets of some sort... but they pull in everything.

[SparWeb]

Workbench magnets....

They start by drawing in random wrenches, loose bolts, and cans of WD40.  No surprise at first, magnets do that.  Every bench seems to have them, but you can't see them.  Soon, paint cans and margarine containers with odds and ends begin to appear, and you start to wonder about these workbench magnets.  Stuff covers less that half the bench top so it doesn't bother you much.

Months layer, you walk into the shop and realize the workbench magnets have accumulated a half-finished roll of poly tarp, a power drill battery charger and a box of hardwood molding offcuts.  You now question if these workbench magnets are breaking some law of physics or an invention you should patent.  There is still nearly 2 square feet of workspace, so it slips your mind after you finish whatever project you are working on, like painting a doghouse or sharpening the lawnmower blade.

After a year you have learned to live with your workbench magnets, and you manage to complete a couple of new chairs for the kitchen using the 13 square inches of flat space that remain.

Workbench magnets seem to be installed in every bench and table in my shop.  Even sawhorses and 10 gallon pails seem to get Workbench magnets after spending time in my shop.

I visited my Dad last week.  Ever since he retired the Workbench magnets have gotten into his boat and pickup, because they are now filled with random stuff too!  We have to find a way to remove or block these Workbench magnets.