Author Topic: Tail failure (Read 7144 times)

DanB · « **on:** February 25, 2012, 10:43:15 AM »

The day before yesterday the tail broke loose on my 20 foot turbine. It appears to just be hanging from the gusset beneath the tail boom, I'll know more when I take it down tomorrow or Monday. 6 weeks ago the 'other' 20 foot turbine also had a different sort of tail failure. The notch that sets the furling stops tore through (1/2 inch of steel), and the tail got into the blades. One blade was damaged. Funny... I just fixed that machine Monday only to have mine break down 3 days later! Overall both of these machines have been well behaved for a while, mine is almost 5 years old, the other one is going on 3. Both of these tail failure happened when the machines were shut down.

It has been a terribly windy winter. Middle of this week we had 3 days with turbulent gusts up to 70 mph. In Nov we had one day with gusts of about 120! A lot of large old trees in our area are snapped off.

jlt · « **Reply #1 on:** February 25, 2012, 12:25:05 PM »

Dan

Looks like it is fixable. Strong wind here also. My camper shell was blown across the field
about an hour ago.

Glad your home to repair it before it does more damage.

ChrisOlson · « **Reply #2 on:** February 25, 2012, 12:40:38 PM »

That's because the tail hinge pin and brackets themselves should not be required to absorb the full torque from the restoring moment of the tail. It will break it eventually. I calculated the stresses on it and even 1/2" mild steel will only absorb a finite number of cycles before it fails.

Might want to look at using a damper that transfers ~ 1/2 the stresses of the tail's restoring moment from the hinge assy to the main yaw tube, thru the damper. Sure, you can throw raw cross section at the problem and try to get it to hold up, but I've always preferred to use sound engineering principles.

It's kind of like your car - how long do you think it would last banging steel on steel with no springs or shocks on the wheels?

--
Chris

DanB · « **Reply #3 on:** February 25, 2012, 02:57:07 PM »

Chris - I agree, a damper would go a long ways towards solving this and it is a common way of dealing with this.
although ~ I am inclined to go with a simpler solution. This time (Because I dont have a lot of time to mess with it) Ill probably just throw more steel at the problem, although using a chain for the furling stop seems a simple and strong solution (not very graceful perhaps but a chain would hold it nicely).

oztules · « **Reply #4 on:** February 25, 2012, 04:07:34 PM »

Dan,
Do you have any idea of the KWH these things put out on a daily basis ( average, peak, poorest=0?)

.............oztules

dinges · « **Reply #5 on:** February 25, 2012, 04:15:22 PM »

Quote from: DanB on February 25, 2012, 02:57:07 PM

Chris - I agree, a damper would go a long ways towards solving this and it is a common way of dealing with this.
although ~ I am inclined to go with a simpler solution.

I think an elastomer damper would be a more simple, more reliable solution; something like this:

Even an aluminium plate (1cm thick) could be enough to reduce the peak-forces during bumping to something more sustainable. The aluminium 'bump' plate would be a wear item though. I expect the rubber to wear out over time too.

midwoud1 · « **Reply #6 on:** February 25, 2012, 04:35:00 PM »

Hey Dinges.
Mogelijk contact ?
Groeten Midwoud1
Frans

ChrisOlson · « **Reply #7 on:** February 25, 2012, 06:22:48 PM »

Quote from: dinges on February 25, 2012, 04:15:22 PM

I think an elastomer damper would be a more simple, more reliable solution; something like this:

I expect the rubber to wear out over time too.

It's my opinion that "simple", "reliable" and "wear out" don't go together.

It's easy to build a hydraulic damper. All you need is a cylinder, a piston with a metered leak, a chrome shaft so it don't rust, a lip seal that fits the shaft and the cylinder bore, a little bushing for the shaft to slide on (GM starter bushing), and a spring to return it to full stroke. It took me all of about an hour to build the custom hydraulic damper shown above out of "junk" I found around the shop.

A hydraulic damper is impervious to sunlight (UV), impervious to water, it works in sub-zero weather without becoming brittle and breaking, and it don't wear out because it's oil lubricated. I mean, I suppose it COULD wear out, but it will last for the useful life of the turbine. Look at the abuse the shocks take, and the conditions they run in, on an automobile - those are nothing but hydraulic dampers.

It's Dan's turbine and he can do what he wants. But on a large turbine like that I wouldn't mess around hanging a chunk of log chain on the tail or screwing rubber door bumpers to it. There's a way to cobble stuff, and there's a way to design it so it lasts.
--
Chris

breezyears · « **Reply #8 on:** February 25, 2012, 06:26:58 PM »

The damper that crisolsen shows looks pretty dam simple to me???
I would go with it.
You can even purchase them and design a way to incorporate it.
Its a no-brainer... i will be adding one to mine.

ChrisOlson · « **Reply #9 on:** February 25, 2012, 07:30:22 PM »

I went to NAPA and tried to find a shock off a little Japanese crapbox that might be weak enough to work. But I couldn't find anything. So I decided to roll my own. The piston in it is nothing but a 7/8" USS nut that I ground the hexes off. I kept grinding on it until it fit and leaked the right amount of oil past when I pushed it into the cylinder (which is just a piece of 1" x 1/8" wall tubing).

I pressed a GM starter bushing into a piece of steel, put the shaft thru it, slid the lip seal on it, and welded the piston to the end of the piece of 4140 shaft with machinable nickel. Then I slide the piston into the bore, pressed the seal in the bore, wrapped a wet rag around it and welded the bushing end on. Then I threw it in the snowbank to cool it off so it wouldn't damage the seal. I filled it up with oil, and welded a cap on it, then threw it in the snowbank again to cool it off right away.

Finally, I put a pretty stiff return spring over the exposed shaft and welded a washer to the end of the shaft. It's made out of "stuff" I found around the shop - even the seal is a shifter shaft seal out of a Muncie 4-spd transmission.
--
Chris

dinges · « **Reply #10 on:** February 26, 2012, 04:04:57 AM »

Quote

It's easy to build a hydraulic damper.

We must have different ideas of what is 'easy'.

I think I would've looked for bicycle shock absorbers instead of those for cars; bicycle ones come in sizes that make more sense for smaller windturbines.

http://www.google.com/search?hl=en&psj=1&gs_sm=3&gs_upl=2641l6764l0l7047l22l20l0l6l6l1l262l1531l10.2.2l14l0&bav=on.2,or.r_gc.r_pw.,cf.osb&biw=1136&bih=733&um=1&ie=UTF-8&tbm=isch&source=og&sa=N&tab=wi&ei=ge9JT8mDKJCXOr2hhMoE&q=bicycle%20shock%20absorber

The more luxurious variants can even be varied in damping rate at the twist of a button, if you feel like optimizing the mass-spring-damper system for critical damping....

Though I've got much more preference to go with the elastomer solution and simply replace it periodically. Good enough in practice, I expect.

As someone once said, 'things should be made as simple as possible... but no simpler!'

Considering the extremely simple solution DanB got away with for many years, I think the only-slightly-less-simple solution of an elastomer damper would be good enough for even more years of reliable service. And if UV exposure bothers you and you don't want to do periodical maintenance and replacement of wear-items of your turbine, it'd be easy enough to shroud it to prevent direct exposure to UV.

I like the KISS-idea. What isn't there can't go broken....

ChrisOlson · « **Reply #11 on:** February 26, 2012, 12:48:58 PM »

Quote from: dinges on February 26, 2012, 04:04:57 AM

I like the KISS-idea. What isn't there can't go broken....

Well, there's many ways to look at the problem. I'm well aware of it and that's why I incorporate a damping system on the tail by design, not as an afterthought. Things like hanging a chain on the tail for a furling stop? Been there, broke that. The chain is dangling most of the time in the wind and it wears the links out. Then what happens? The wind is blowing at 50 and the turbine gets snapped around, the chain goes TWANG-ANG-ANG-ANG! SNAP! and you got the tail in the blades.

Door bumpers? They don't last on doors, much less on wind turbines. The first time it gets down to -10F with the wind blowing at 40 and I find pieces of my tail bumper laying in the snow under the tower because it got brittle - do you think I'm going to lower the tower to fix it? Ain't gonna' happen.

That's why I designed a system that you don't have to fix on a regular basis, and requires no normal maintenance. If you use KISS and it sure looks good, but it fails when you need it most, then you didn't accomplish anything. I been a mechanical engineer for too many years to know that using a stick to prop it up, just because it's simple, is not always the best solution. And my advice to anybody reading this thread is that when you got parts - heavy parts - falling off wind turbines, that you'd best attend to the problem the right way, and don't patch it.
--
Chris

Mary B · « **Reply #12 on:** February 26, 2012, 03:46:57 PM »

How about an air cylinder and just use a needle valve on the input to control the amount of air escaping.

RP · « **Reply #13 on:** February 26, 2012, 04:56:50 PM »

I've used micro pneumatics at work for years. The problem with air cylinders is that air is compressible so you end up with a spring instead of a damper.

ChrisOlson · « **Reply #14 on:** February 26, 2012, 05:46:18 PM »

There's a number of different ways the problem can be addressed - the important thing is that is addressed. I'm glad DanB posted it because it shows people the results of improper design and testing. And that a design is never proven until it has adequate tower time.

Tails breaking off wind turbines is nothing new. The Jacobs turbines marketed by CDC had problems with the tails breaking off. I don't know who came up with the new heavier duty bracket design - whether it was CDC or Marcellus and Paul. But it's a well known fact that Marcellus, especially, was not happy with CDC marketing these turbines without what he considered to be adequate testing (this was after CDC acquired controlling interest in the company). Experience had taught Marcellus Jacobs that paper and computer simulation engineering rarely works in the real world of wind power, and he was not afraid to tell book-learned engineers that right to their face.

As long as you incorporate some method to absorb the energy of the tail restoring moment, it is way better than relying on a notched hinge tube that just slams steel on steel.
--
Chris

halfcrazy · « **Reply #15 on:** February 26, 2012, 06:45:27 PM »

Chris
Just thinking out loud but could you use a heavy spring inside a tube with a smaller solid rod slid in for the tail to hit against. It would take a little machine work to keep the solid rod captured but would keep it a little more DIY for some?

Ryan

ChrisOlson · « **Reply #16 on:** February 26, 2012, 07:55:17 PM »

I think that would also work. Anything that absorbs the energy of the tail, and provides a decent amount of "cushion" to absorb it instead of merely instantly transferring it someplace else, is good. The reason I didn't use purely a spring is because of the rebound tendency. I'm afraid that the rebound of a big spring might bounce the tail and rotor back to partially furled and cause erratic operation.

The key, to my way of thinking on it, is to make it so the operation of the furling system is smooth. If the wind catches the tail when it's in a partially furled condition, and slams it up against the stop in the full flying position, I think you want it sort of like dropping a bowling ball in the sand box, as opposed to having it spring back, to keep the stresses on stuff down to reasonable limits.

The best furling system I ever developed is the spring loaded vertical hinge type used on my 12G turbines. Those are so silk smooth that it just doesn't get any better. That's because it's dampened both ways and the furling rate is constant to 70% furled, instead of progressive, and drops off as the linkage approaches over-center. This brings the rotor back into the wind very gently and very smooth instead of a radical return like an angled hinge system has. I've only recently gone back to the old angled hinge system because the linkages on the type used on the 12G won't fit around the large ferrite generators I'm using now.

I've been working on different furling systems for the better part of three years. Ultimately, I'm looking at methods to eliminate it all together because turn-it-to-the-side furling is not exceedingly reliable no matter what you do. It's pretty much stupid to impose huge gyroscopic forces and torque loads on the turbine head structure in high winds when there are better, and proven, methods to control power output that don't require swinging the rotor around the yaw axis.
--
Chris

halfcrazy · « **Reply #17 on:** February 26, 2012, 08:00:44 PM »

Chris
Excellent explanation as usual, I agree with the rebound issue that make a lot of sense I guess I was thinking the bounce back of a spring may be the lesser of 2 evils when compared to the hard bang of steel on steel. Thank you for the descritions and keep up the great work it sure is fun to watch all the great work you do come together.

Ryan

ginger48 · « **Reply #18 on:** February 26, 2012, 09:36:17 PM »

Most, if not all, waterpumping windmills made have a rebound spring on the tail. It's been a success for them for over 100 years.

Isaiah · « **Reply #19 on:** February 27, 2012, 02:24:33 AM »

Why dose the tail furling have to climb? if it doesn't have to climb one could use a crap car spindle and hub for the tail pivot.
May even be able to build something with the spindle and ball joint Assembly?
Another option on the damper and what Chris made would be a two way hydraulic cylinder with a air bleeder valve in one side you used to be able to buy these bleeder valves at the farm store but I don't know today .
Or probably make your own out of stainless steel or brass so it doesn't rust. Its nothing more than a bolt with a small hole in it.
I'm in the process of building a small mill and am just about to the point of the tail. I have a 1937 GMC 1 1/2 ton truck out back I'll take a look at the shocks on that as back then you put the oil in them and they are totally different than today's shocks.
One could hook up two small hyd. cylinders fill one with oil the other empty and let them work it out. just some thoughts.
Dose anyone know how much force is on the mill head when it furls there is none on the tail as the tail is always down wind.?

DanB · « **Reply #20 on:** February 27, 2012, 07:42:35 AM »

Good discussion. Even a rubber bumper could go a long ways, I've done that in the past.
There is no doubt that the very small radius of the stop and such a heavy tail is a problem here. We also have unusually turbulent winds at this site. Another problem that's lead to this is my yaw bearing, without getting into detail let's just say its a poor design that wears out every 2 years and gets very stiff (it's worn out now) - such that, when the machine is shut down (not vibrating) - it won't yaw, so the tail gets slammed around a lot more than it should.

Yesterday was too cold/ windy to lower the tower (the tower would've been fine but the humans were not in the mood). Maybe today. Probably my solution this time will be to weld the tail back as it was, and use a chain for the furling stop, this will move the load out to a more reasonable radius away from the tail pivot.

wildbill · « **Reply #21 on:** February 27, 2012, 10:31:11 AM »

I don't have a water pumper, but i have a air pumper with eight blade fan 6+diameter ft. I used a car shock on the tail, but found out i needed a spring to bring the tail back to center,so added that. after four month's i noticed the tail stays 12" off center. Okay it works, spins in low wind turns out in 20+m/hr wind all is good. I look out this morning i see my spring hanging down, tail is now 24" off center. Pumper is still working, but i can't take it down today,grand kid duty, and a storm of 6+" of snow coming, so I'll just have to cross my fingers for a few days.

windvision · « **Reply #22 on:** February 27, 2012, 11:36:22 AM »

Dan:

You mentioned that it was also vibrating. Do you know what is causing that? Thanks.

ontfarmer · « **Reply #23 on:** February 27, 2012, 11:39:09 AM »

I use two inch hydraulic hose on my 2o footer for the bumper works good for long time.

Ungrounded Lightning Rod · « **Reply #24 on:** February 27, 2012, 12:10:21 PM »

The lever arm of the tail can produce ENORMOUS forces on the pivot and stop if they're close together. So put the stop a significant distance from the pivot.

I like the rebound spring rather than bumper - either solid or cushioned: First it applies the stopping force gradually. Second, it doesn't degrade from UV. (If I did a bumper I'd pick a material designed for long rough service - like those used in automobile suspensions - and try to shade it.) Third it doesn't absorb energy itself: Instead it bounces the tail back and lets the vane-vs.-air interaction provide the damping. (Absorbed energy generally produces wear. So absorb it into the air, which - even if it DID wear out - is constantly renewed.)

Downside of metal-on-metal impact is noise. A bit of exterior-service rubber over the impact point would quiet it. Wearout in that mode would just make it noisy (telling you to replace it) rather than make fail (telling you to make more blades and live without the mill's power until they're done).

I'd be tempted to deal with the "miss the target" issue by putting the spring on the tail and providing a large target for it on the head assembly.

The furled-stop is intermittent duty so it should last a long while. It gets hit hard and exists to prevent mill destruction so it needs to be reliable. If you do a fully-extended stop - continuous duty but just to improve tracking by making the tail-head angle constant up to the start of furling - it might have different requirements.

Quote from: Isaiah on February 27, 2012, 02:24:33 AM

Why dose the tail furling have to climb?

If you're talking about a tilted-pivot system, the tail climbs because the restoring force of the tail's weight working against the tilted pivot acts as a "spring" that provides a restoring force that varies with tail angle, to meter the furling angle of the blade assembly. A physical spring would wear out, detuning the system and eventually failing. Gravity doesn't wear out and a pivot can last a VERY long time.

News:

Author Topic: Tail failure (Read 7144 times)