Wind Turbine repair

[Islander71]

Last September I lowered my 5Kw wind turbine tower as it was taking stronger wind to get turning than it should, though strangely once spinning it would keep going in very little wind. With the hub pointing down and tail vane pointing up, it span without restraint, but lifting a blade and trying to turn it revealed it was rubbing, so the main bearing had failed. It was commissioned 19 years ago, and the main bearing went after 9 years, and the main hub/shaft was returned to the current manufacturer for renovation as they thought it not possible for home repair.
This time I thought I would do the repair. The problem was the bearing with housing and shaft that needed replacing as a unit was unavailable.

I perhaps too hastily had taken an angle grinder to the fibreglass hub cowling to get better access..... The blades and tail vane were taken off, then the main hub unit unbolted from the large swivel bearing at the top of the 12m tower, and taken back to work on.
The rotor consists of 2 large 15mm plate discs each of which has 16 rare earth very powerful magnets. The stator sits between the 2 rotor/magnet discs and the coils are set in a resin disc, with about 4mm gap between stator and magnets. I was told there is a 4 ton magnetic force pulling the 2 rotor discs together!
Eventually the bearing/housing/spindle assembly became available and sent to me, so I set about dismantling the hub. 3 of 8mm, 8.8 grade studding was screwed into the inner rotor and nuts and washers used to gradually wind the rotors apart. Had to be careful as if the spanner got a little too close to a magnet it would try to snatch it out of my hand. Once separated by about 7in, it was possible to carefully lift the rotor off and place it down on some wood.....with nothing ferrous close.
The stator was held onto the shaft with a Tollack fitting, which was new to me, but reasonably straightforward to remove.
The old shaft/bearing unit.

It weighs 62lb. Much of the difficulty was the sheer weight of everything as I was working on my own with occasional assistance from my wife.
Then I encountered another problem...

 

[bricol]

You mean a "tollok" locking assembly? Excellent devices - use them a lot. Ringfeder a more expensive version of the same sort or thing, usually, but not always, with greater capcity on the same shaft/hub.

If it's up in the air and annoying to get to, make sure you follow the torquing instructions to make sure it stays there.

 

[Islander71]

Looking at the inner rotor, it was obvious 2 of the magnets had lifted and the inner plate was heavily rusted.

Had to use a chipping hammer then angle grinder to de-rust the rotor and under the lifted magnets. It had to be taken onto the kitchen table as the glue required a minimum of 15*C for 24 hours.
In order to fit the 2 new magnets i had to make a jig and slide to position the new magnets, as otherwise they would leap onto the other magnets and be almost impossible to remove.
Glass beads were put where the magnets were to be bonded to give the correct glue film. The other magnets were covered with wood. As soon as the new magnet was placed on the slide it leapt into position!

The cleaned inner area and around the new magnets was painted with Zinga paint, then fitted to the new shaft/bearing, then smothered with marine waterproof grease before the rest was re-assembled.
Now I had to sort out the hub cowl I had cut in half for ease of access.....

Used aluminium plates and fibre-glassed an inner lip.
Now it was just necessary to get the hub onto the front loader forks and taken back to the tower. Getting it perfectly to line up with the tower bearing ring while placing a spacer ring in place and getting the mousing cord ( for the wiring) through took some patience!

Then just a question of putting everything back together, a new tail fin was fitted as the old one had rotted in places, setting the pitch angle of the 3 blades and then setting the blade pitching torque correctly. By the time that was done it had become too windy to risk raising the tower, so it was lashed down for the night.

The next morning it was time to see if it all worked. First it was an hours hard labour operating the Tirfor winch to raise it.... that used to be a lot easier when I was younger! Lifting the first bit with the front loader helped.

Eventually it was up, that was yesterday and it seems to be working as it should.....

 

[Gareth J]

Brilliant write up, thank you. I have been told that replacing the main bearing isn't a DIY job so hats off to you for doing so much. Good that those loose magnets hadn't chewed into the resined in coils, that could have been expensive.

The inside of your machine looks very clean! Compared to mine anyway, which is a dog's dinner of brown dinitrol and rust.

I'll be giving those magnets a close inspection next time.

What are your thoughts on greasing that main bearing? Obviously there is no grease nipple but I confess to have snuck some past the seal with a flattened hypodermic needle....then second guessed myself and convinced myself I'd trashed the bearing for sure!

 

[Islander71]

I'm unsure whether adding more grease would prolong the life of the bearing - the danger is damaging the bearing seal. If you can get the right grease in without seal damage I guess it would help. It could well be dirt getting past the seal that causes the wear. Personally I suspect that a single bearing for such a prolonged high loading is a mistake, but cann't see how a second bearing could be added to the existing design. It only takes slight wear at the bearing to be magnified by the distance to the magnets to cause a problem. I would have thought a roller bearing set a distance back down the shaft and adjustable taper roller bearing at the front would have been a better design.

 

[Gareth J]

I've got some hyper expensive "wind turbine" specced grease that I use for it, well the yaw bearing really. Hopefully appropriate.

I'd agree that just the one bearing feels a bit vulnerable, especially considering the knock on consequences of failure. Then again, mines done pushing 150k kWh on its original so can't be all wrong.

What glue/system did you use for the magnets?

Am currently messing about with a pair of sma sb3300 inverters, trying to get them to run in wind turbine mode as a bit of insurance against the current inverters failing (again).

Did you manage to sort out your cutting out issue? I recall you were having problems. Mine's started doing it quite a bit during squally showers. I changed one set of pitch bearings last summer, I thought there was only one offending notchy bearing - I should have changed them all but was in a rush. I presume my issue is another bearing that needs doing anyway.

 

[jmp49]

Nice write up. Does this have variable blade pitch too?
I have a wind turbine of dubious parentage and a tower to go up bought on a whim to complement solar, but its in storage for now as we have to have a "mot" (CT) for wind turbines here (to check the blade condition isnt dangerous) to be legal and I havent wanted to face the nonsense of a inspection yet, and the blades are not in the best of health after some of them exited the chat during a overspeed on the previous owner and have been repaired with glassfibre patches. Its not variable pitch but it does have a dc brake that works.
It too is a massive lump with heavy construction on the housings, I can just about drag along the floor the main unit sans the tail fin section and to lift it, the forklift is used. Maybe I should consider stripping it and checking the bearing states as well as replacing the blades.
If there's any hidden damage, I think I'd be tempted to get a replacement 5kw one and just use the tower elements as I think things have come a long way since it was designed...

 

[Islander71]

Hi Gareth, Britwind supplied the glue with the new magnets. Once I was eventually able to get in contact with them, they were very helpful. The 2-part glue was Permabond TA4202, but putting down the fine glass beads is essential to stop the force of the magnet squeezing out all the glue.
I am still running the original SMA Windy Boy inverters now 19 years old, though they have become irritatingly noisy. I think Britwind have found a suitable replacement now that the SMA ones are obsolete.
The cutting out issue I had before turned out to be an iffy corroded connection at the bottom of the tower to the brake switch, that must have broken connection whenever it got windy and probably shaken by the tower. Nearly drove me up the wall having to keep going down to switch the brake off several times a day!
You can check the blades are pitching out properly by disconnecting the mains simulating a power cut when there is sufficient wind to trigger the pitching mechanism - you can easily hear it - and noting the voltage which corresponds to the rotor speed. It should be between 420-470v.
My turbine is on its third set of blade pitch bearings, but at least they are easily available and inexpensive, and easy to replace.......unlike the main bearing.

 

[GaryS65]

Do you have a drawing or pictures of the main bearing assembly? Anything put together by man can be taken apart by man.

 

[Morris]

Do you have a drawing or pictures of the main bearing assembly? Anything put together by man can be taken apart by man.

You've never owned a French car have you

 

[Gareth J]

Hi Gareth, Britwind supplied the glue with the new magnets. Once I was eventually able to get in contact with them, they were very helpful. The 2-part glue was Permabond TA4202, but putting down the fine glass beads is essential to stop the force of the magnet squeezing out all the glue.
I am still running the original SMA Windy Boy inverters now 19 years old, though they have become irritatingly noisy. I think Britwind have found a suitable replacement now that the SMA ones are obsolete.
The cutting out issue I had before turned out to be an iffy corroded connection at the bottom of the tower to the brake switch, that must have broken connection whenever it got windy and probably shaken by the tower. Nearly drove me up the wall having to keep going down to switch the brake off several times a day!
You can check the blades are pitching out properly by disconnecting the mains simulating a power cut when there is sufficient wind to trigger the pitching mechanism - you can easily hear it - and noting the voltage which corresponds to the rotor speed. It should be between 420-470v.
My turbine is on its third set of blade pitch bearings, but at least they are easily available and inexpensive, and easy to replace.......unlike the main bearing.

Excellent information, thanks. I'll take a look at the brake box too.

Yeah, I've got a nice old analogue multimeter for monitoring DC with the inverters off. Really let's you watch the pitch mechanism's operation. Never seen it overvolt but maybe it just sticks a bit randomly.

The inverter settings appear to be changing nicely. Let me know if you ever get in need of an inverter, if these sb3300 ones work, I'll have several spare sb2500 ones that should pretty well be a straight swap.

 

[Gareth J]

Do you have a drawing or pictures of the main bearing assembly? Anything put together by man can be taken apart by man.

Something about special alignment jigs and press tools.

 

[selectedgrub]

You've never owned a French car have you

Has anyone outside of France?

 

[GaryS65]

Something about special alignment jigs and press tools.

A press sure. Alignment jigs, can't see it myself

 

[Islander71]

A press sure. Alignment jigs, can't see it myself

I was told the force on the bearing to press into the outer housing was monitored and if not within a tolerance was rejected, apparently very critical. The same with the inner of the bearing fitting the shaft, and not to re-use a shaft as they can plastic deform over time. The same with the bearing housing. It is a high load for a single bearing - the blades have a 4.8m diameter, the complete rotating hub assembly at a guess is weighing maybe 250-300Kg and when any wind, is spinning at about 250 rpm, and lasted 9 years of doing this! Obviously it is not turning when calm, but probably the equivalent of continuously running for 7 years.

 

[Gareth J]

Another thought for the future, you could make a fitment for the loader to, once the lift of it is exceeded as standard, attach onto the headstock, giving you another length of prodding height that will allow you to assist the tirfor through the heaviest part of the lifting.

However, don't attempt to lower it using the loader instead of the winch - a neighbour did that and it swiftly picked the front of the tractor up and dragged it along, crashing the turbine into the ground!

 

[GaryS65]

It's got to be a double angular contact to take those kind of loads.

 

[GaryS65]

Which model is it? A Britwind one?