time for repairs

[kitestrings]

We're getting closer to getting our turbine ready.  The magnet rotors cleaned up really well.  Aside from the one magnet rust/swell issue the other appear to be in good shape.  These are now ready for sandblasting and paint.

One question: I'm curious what folks have used for glue/sealant to attach magnets to the rotor plates?  We will be using an epoxy primer (Concept DCC) after sandblast.  I've heard super-glue, I've heard JB Weld, but I want to make sure it isn't something that will soften the primer.  Perhaps a water-based sealant?

[kitestrings]

Downwind magnet rotor:


Upwind magnet rotor.  Gorilla duct tape worked well to remove all the magnet filings.  Acetone removed the dirt and grease residue:

[SparWeb]

I have access to a variety of structural adhesives - when they expire at work I get to take them home.  I've used System-3 T88, Hysol EA9309.NA and 3M EC2116AB recently, and years ago used Hysol EA934.NA to bond the magnets to the motor-conversion's rotor, and they are all unbelievably strong.  Not much point in singing the praises of these adhesives when they're all expensive, and you don't need a superglue for this anyway.

The magnetic attraction of the magnet to the steel means that the adhesive not actually needed to hold them together.  The coefficient of static friction between the two parts X the force of attraction gives you a friction force so high to overcome I don't think the alternator could spin fast enough.  So the adhesive is acting more like a sealant the way it's being used in the rotor.  When that's done properly, it prevents water from getting between and causing havoc.

Most bonds fail because of improper preparation of the surface, not the adhesives themselves.  We'll never really know what went wrong with your one errant magnet, but looking at the others, I would say the process you used when you originally built the turbine is sound.  So is there really a need to do anything different?  It is unlikely that you'll allow a contaminant to get into the joint this time.

Epoxy can allow some moisture to infiltrate the joint, but not much, so if you paint it afterward you're safe.  JB Weld is fine.

[DanG]

My experience with JB Weld was watching 1 x 1/2 x 1/4 N42's cleave loose and jump free of it first chance they got on a tightly packed motor conversion.

Blue water mariners count on 3M 5200, it is one of the few things trusted below the water line. It is about as forever of a caulk & adhesive as can be produced and is a good vapor barrier to keep from feeding corrosion on bedded parts.  I've potted impossible to stop leak areas on vintage Airstream trailers - and was able to forget about them until a minute ago, thanks.  Using it would make the next repair cycle 'interesting' as you make up brass gouges, chisels and wire saws to get anything freed of it...

[kitestrings]

Yes, I'm not concerned about adhesion, what we've done in the past has worked well.  I really was just thinking to have a sealant between the magnet and the primed plate before finishing.

I suspect one possibility is that there was some small void, and this magnet was likely the where the turbine preferred to settle on (6-O'clock position).  Moisture could have been behind the plating as an OEM defect, but more likely it got a foothold through some path behind the retaining ring, and in-turn the magnet plating.  Anthing I can do to reduce the possibility of that being repeated would seem wise.

I'll take a look at the 3M 5200. I've had good luck with the JB Weld Clear-weld on minor blade repairs.  As long as it doesn't soften the epoxy primer - a test area would be easy to try - this may be fine.

[Ontheronix]

Beautiful build (I like the white finish and the way the stator is built with the seperate coils that are held in between the polycarbonate sheet) and pictures!
How high is that tower? You'd want one for the view alone 

[bigrockcandymountain]

I am pretty sure jb weld is an epoxy itself, so should adhere well and not soften your primer.  I also like the t88 spar mentioned but it is expensive and maybe overkill for what you are wanting.

Glad to hear you are getting close.  I have been wondering how things are going with no turbine.  I would be sad without mine this fall.  Lots of wind and not much sun lately. 

[kitestrings]

Well, the intended advantage of the separate stator sectors is to allow the stator to be removed without taking the whole unit down.  The disassembly worked well…I'll keep you posted on the reassembly.  This is what's left up top:


Things have taken a little longer than expected.  No surprise there, but it has allowed us to work through a few improvements that I've been contemplating for a while, namely:

Resolving the magnet corrosion issue
   We're epoxy priming the magnets before setting this time… five sides, then they'll be flipped onto the already primed magnet rotors.




Refinishing the blades
   I've been wanting to take the trailing edge down to near knife edge at the outer stations.  This was something I hadn't dared on the initial build.  The blades are primed and ready for top-coat.


Better balancing of the rotor
   We've got a second spindle shaft this round, mounted on an 8' stand.  I'm hoping to balance by moment individually , but then as an assembly on the stand.
Replacement of the stator rings
   We opted for G10 / FR4 stator rings to replace the polycarbonate.  The material is .236” instead of 3/16” and the rings are 2.3125 wide instead of 2”.  All of the fasteners now have bushing stops and will be sleeved with shrink-wrap to allow for expansion and contraction.  Some discussion on this:
https://www.fieldlines.com/index.php/topic,149818.0.html

[Bruce S]

With all those mags on that bar, I'm surprised you don't have tools flying off the shelves.

Nice set of edges on the blades too!

Bruce S

[kitestrings]

I always breath a sigh of relief once I have the things on a piece of steel.  The scary thing, to me, is having one in free air for any reason.  I've found that carpenters' wood screw clamps work about as well as anything for handling them.

A word of caution: you can very easily change the polarity of a compass with an NdFeb.  Be sure and check it before orienteering.

[SparWeb]

ou can very easily change the polarity of a compass with an NdFeb. Be sure and check it before orienteering.

Not a problem; just be sure to view your map in a mirror at the same time!

[kitestrings]

I had to increase the width of the slots in the magnet retainer rings by ~.003 to account for primer


The magnet rotors are primed


Here is the upwind rotor with the hub removed and the magnets in position.  Note that the rings are spaced off the plates this round to help eliminate hiding spots for moisture.


These are the new G10 stator rings.  They notably stiffer than the polycarbonate rings they are replacing.  The spacers and shrink wrap fit snug-tight.  Does anyone have a suggestion for UV primer/paint for these?  I'm thinking it might make sense for the exposed sides.


Here they are fitted against the stator sectors


Finish painting is most of what's left, then a break in the weather?

[SparWeb]

Metal spacers to clap-up against - exactly what I was hoping to see.

UV resistant paint/primer - there's a lot to choose from.  All aircraft primer and most automotive primer has UV protection.

[kitestrings]

Yes, that's right.  There is only minimal (length) clearance between the thickness of the G10 and the spacers.

I'm having second thoughts on the smaller, barrel bolts.  They would require me to put in about three shim spacers to get the right fit, and juggling them up on the tower seems an unnecessary complication.  I think I will probably just use 1/4-20 hex bolts, with the shrink wrap and nylocks.

The weather will be a factor now.  I'm continually reminded

[Bruce S]

Is it me or does that one coil to our left of the barrel through hole look broken? or at least out of the molding?

Bruce S

[kitestrings]

It does appear that way, but it is just that the wire closest the center not covered with epoxy at the bottom, and the one above it is just the opposite (making it look like a break in the same strand).

[SparWeb]

Did you get it raised back up after these repairs?

How is it running now?

[kitestrings]

No.  We seemed to have one obstacle after another, and in the end we opted to wait.  Nothing too serious... initially it was just that we started a bit too late, then made the decision to take down the alternator and to refinish the blades.  Eventually weather became a factor, and then (maple) sugaring season, which with global warming now starts in January!  Go figure.

I'm not in any way discouraged though.  We've hit most of the alterations I'd hoped to get.  And, the PV output has been very steady this year, so pretty limited run-time on the generator.

We've totally refinished the magnet rotors.


They are now re-gapped and torqued


We've replaced the polycarbonate stator rings with G10/FR4.  Those are now primed and painted.


The blades are primed, and prepped for top coat.

What's left is re-balancing and reassembly.  I'm also considering adding a weather shroud above the alternator to just deter rain & snow from entering the thing unchallenged.

In truth, the days are long enough that there is no pressure now.  And, the Covid-19 issues have shifted priorities... this all seems small right now.  Still fun, but just less pressing.

[kitestrings]

Oops, posted the wrong picture there.  Here's the one  of the stator rings that I'd intended:

[SparWeb]

Well with spring around the corner, good weather will be calling you back to it.

My wife now reminds me (more often that before) to work carefully because I don't want to hurt myself and end up in the hospital NOW!

[kitestrings]

Been awhile, but we finally got the blades primed and painted.  I did a bit of reshaping... mostly taking the trailing edge down a bit sharper and rounding out the transition from air-foil to square stock at the hub.

Albert, my friend & neighbor in town does this for a living.  I teased him that my painting was different.... i usually end up with a run or two.  He makes it look easy.





I since have applied 3" leading edge tape.


And started the final balance process.  My plan this round has been to calculate the moments and adjust for anything over ~2%.  I re-checked after paint application and we are within 1.5% (highest to lowest spread).

blade#      distance from      weight      moment
      center (in)      (lbs)      (in-lbs)
1      30.0625      23.50      706.469
                  
2      30.6875      23.38      717.320
                  
3      30.3125      23.50      712.344
                  
               1.5%   hi/lo spread

This worked well. The edge is locked on my saw top, so if it tips it can only go a few inches in either direction.  Works better than my vice, which is what I've used in the past.


I now have an extension on my floor stand, and we mounted a second spindle to it.  So, I will attempt to do tracking, 120 deg check and final balance with that in play.  The only catch is that I don't have a space inside quite large enough to be able to spin this (15') rotor.  It will have to be done outside, and that means a totally still day.  ~ks

[SparWeb]

Around here, perfectly still doesn't happen ever, so I don't know how much luck you'll need to do a whole-rotor balance outside. 
For that last 2% there's very little force needed to upset it one way or the other. 
If you're only tracking, then it shouldn't be so hard.

[kitestrings]

A calm day during this time of the year shouldn't be too difficult to find.

A couple minor tasks have gotten checked off my list; this was one.  I installed these neoprene load-rated bumpers stops.  Previously we just had some 3/8" carriage bolts with a small stack galv. washers.  The washers pinched tight against the tube, but I never liked that it was metal on metal contact.  It's an area that could introduce rust, but it also may be a source  of some noise thru vibration at/near cut-in.




This plate just serves to transfer a bit of the flex out of the tube at its mid-span to within the guide portion of the tower.  You can see it here:

[kitestrings]

Worked on final balancing over the holiday break.  Here's the extended test stand.  It worked very well.


Here with the blades mounted.  Pretty easy reach off the tailgate of my truck.


I put the blade with the lowest moment on the "ear" of the hub-plate that was heaviest.  With weights near the blade tips it only required two or three 1/2" washers, but I hated the thought of drilling holes in them.


To get the same moment... approximately 10 in-lbs., I needed a 9" piece of 1/4" x 1.5" bar stock on two of the blades. This was what I leaned toward.  The only trade off is that they are not at the plane-center of each blade, if that makes sense.  I suspect Flux might have cautioned about introducing a potential balance imbalance, or "couple".  Hopefully negligible.

I also checked spacing (120 deg.) and added some minor shimming to true up the tracking.


Here are the primed and painted blade plates and the two balance weights that I fabricated.  Yes I can paint, half well, when I need to....


I think this is the last of the bits and pieces.  It's time to start thinking about getting things put back together back on the tower.

[SparWeb]

Looks like the math worked out for you, if you used 10 inch-pounds and that's what your measurements of the individual blades predicted, two weeks ago.

[kitestrings]

Pretty close, yes.

Despite everything being symmetrical on the alternator, there is still some minor imbalance there.  If you spin the thing, it will always tend to land with the heaviest section down, right.  We had corrected for this with a minor weight directly opposite this "ear" where one of the blades mounts.  I experimented a bit with - and without this weight.  By putting the lighted blade on the heaviest ear, I believe it cancelled out a small bit of imbalance.  The remainder I made up for through trial and error on each of the two lightest blades.  I was working within the ~1.5% differential; it is definitely closer than our first attempt (where I ignored this delta).  Now each blade will hold a horizontal position on either side.

[SparWeb]

Kudos for not sprouting mushrooms all over it, too.  You did better than me.

[clockmanFRA]

Hi Kitestrings,

I see you are having low speed centre off balance, because of those balancing weights being near the centre of the hub and the weight having to be substantial.

There must be a set of figures when the blades overall weight can be affected by a offset weight that has a mass against the blade weight.

Since reading your balancing post here, i wonder if anyone out there has actually used small weights drilled into the tips of blades successfully over several years. ?

Perhaps a new thread specifically for weight adding and securing at the tips of blades.

As i said recently on my fiberglass blades that are about 2/3rds heavier than my wood cedar blades i do not have slow offset vibration due to the weights near the hub. While my wood cedar blades definitely oscillates to the point that I may have to drill and add weights at the tip.

Suggestions. ?

My Cedar wood blades weight added on just 2 blades.

[bigrockcandymountain]

I wish i had a picture.  My one blade is quite a bit lighter than the other 2.  I think it is a different species.  It is very soft. 

Anyway, when i balanced it, i had to use a piece of 1" thick steel 6" wide and 4" long.  That is about 7lbs or 3kg. It works ok and doesn't have a preferential spot to stop, but oscillates pretty bad at low speeds.

Next time i carve blades, i will be careful to get similar weight wood. 

[mbouwer]

Would it be possible to include a small u-shaped ss steel plate, thickness 2 mm, in the blade tip leading edges.
With a threaded hole in the leading edge. For example M8
Then neatly filling the hole with sealant that does not become too hard.
Whenever you notice an imbalance, you can scratch the sealant out of the hole and screw in a short or longer M8 allen screw.
And seal again.
Or remove the screw and seal the hole.

[kitestrings]

My fiend Neil, who's done quite a lot of this work, has said,

"My personal preference for weighting is to drill a 3/8" hole into the leading edge of the blade, put in the appropriate length of 3/8" round lead and add West system epoxy to bind it together.  After the leading edge tape and paint goes on over it, it's aerodynamic, secure and protected from the weather."

This seems like a good approach.

We tried to keep the weight of the blades very close as we carved them, but the finish, tape and in our case any minor  imbalance in the alternator itself plays in to it.  We got a really nice finish on the blades, and I just couldn't bring myself to drill into them.

While it may seem like a lot, in this case the weights we added were less than 1 lb. (.45 kg), considerably less than the blade plates at roughly the same location.  I guess if I had it to do again, I might have balanced the magnet rotors first, but I'm really not expecting any problems.  Statically, we're pretty spot on, and we're going to know soon....

[kitestrings]

I like mbouwer's suggestion.  Seems simple enough, and I like methods that would allow one to mkae adjustments.  I think it was Zubbly that had the threaded rod and adjustable weight 180 deg. opposite the heaviest blade.

[SparWeb]

Since reading your balancing post here, i wonder if anyone out there has actually used small weights drilled into the tips of blades successfully over several years. ?

I can claim 9 months.  I found that putting long bolts into the blade tip works well, but it was limited in scope.  I could put only a certain quantity of metal into the tip, before so much wood at the tip had been removed that there wouldn't be enough strength to hold the metal inside.  At the time I did not think of putting a rod on the leading edge.  That's a great idea and I hope I can remember it years from now when/if I need to consider it again.

I haven't needed to adjust the balance since raising mine last year (I had an opportunity to check on it a few months ago) but it would be very had to change the balance now.  I opted to epoxy all of the balance weights in very thoroughly.
If I was unsure, I could have done as Mbouwer suggests - a plate or fitting that would accept more or less bolts until the balance adjustments were satisfactory.