15' footer upright and turning

[kitestrings]

Hi Folks,
 
I've been wanting to post an update on this turbine.  It's finally up and running.  For now it is mounted on just a guyed stub-pole tower section, about 25' up.  It has been nice to be able to keep an eye/ear on it, and make a few adjustments without having it 100' off the ground. So far I'm only running it when I'm home to monitor things.
 
Briefly, here's the overview:
 
15' dual-axial, 16-pole

Mag rotors: 32 - 3x1.5x.75 N42 Ndfeb

Coils: 12 coils – 2-in hand, #14 AWG, 53 turns
Cut-in: ~92 rpm, 58 VDC
No slip-rings, SO “twist cord” with strain-relief

 
Here's what is a perhaps a bit different:
 
Coils are individually potted with T& G sides interlocking the 12 sectors.  We used a ‘thermally conductive' epoxy encapsulant; coil cores are open for cooling.

 
During assembly the magnet rotors are gapped and mounted to the spindle, then the stator is positioned and centered.  Conceivably the stator can be removed without jacking apart the rotors.
 
The stator bracket is stainless and is fastened to the spindle using a keyless bushing.  It can be removed if needed.



 
The blade hub plate (1/2” A656) is also the upwind magnet rotor.  There are stainless magnet rings that position and assist with holding the magnets in place.
 
The tail-boom is telescoping AL tubing, so the length can be adjusted (3 positions).  The tail hinge has bronze graphite radially bearings and a recessed greasable, roller pin thrust bearing.
 
The yaw bearing is a graphite/alloy composite thrust washer.  The radial fit between the yaw-head assy and the stub tower is greasable (I know, a bit of a contradiction in objectives).
 
The tail can be manually furled with a hand winch.  4 mm dyneema line (thank you Frans) runs thru a pivoting pulley head.  The line is held in tension with a ‘sash weight' in the stub tower.



 
Rectifier:
 
There are two isolated, full-wave stud mount diode bridge rectifiers.
RPM gauge (Brian Smith's)
 
 
Controllers:
 
Midnite Classic 250 (2)
“Waste not” diversion to water pre-heat
Homemade ‘clipper' and high-limit voltage sensing relay


Still have to lots of tweaks to make  - dialing in the curves, adjustments on the Aux settings, furling adjustments - but, so far looking good.  And, still too soon to know which changes work and which we'll live to regret.  For now this is where things.  Many folks to thank here for helping me get this far.
 
prior post on potting coils & mag rotors:
http://www.fieldlines.com/index.php/topic,147846.0.html

kindly, ~kitestrings

[kitestrings]

Missed a couple:

The gauge reads rpm (x100).  Rough early observations: cuts-in about 90 rpm, 125w@100rpm, 160w@110 rpm, 300w@120rpm, 500w@130, 600w@140rpm.  I'll try to chart better data soon.


Classis snapshot:

[Boondocker]

Looking Good.  Nice to see another turbine in operation. 

Any idea of the wind speed when you snapped the photo of the Classic displaying 1100 watts?

[fabricator]

Very nicely done.

[SparWeb]

Nice work all around.  Not just the nifty trick you pulled with the stator coils, but also the details in the tower and tail furling. 
I'm sure by now you look forward to getting home to see "what it's been up to today".   

[charlie_ruizpr]

Can you tell us how to make the homemade clipper?

[kitestrings]

Any idea of the wind speed when you snapped the photo of the Classic displaying 1100 watts?

I don't have a very good way of capturing wind speed yet, other than thru observation.  My guess would be in the 17-19 mph range.  The rpm gauge is a pretty good indication, although it too is instantaneous.  I hope to get a data logger on either the rpm output or my anemometer.  The local app is also pretty cool, because you can watch it from most anywhere in the house - just wish it had wind speed:


It seems like I spent an inordinate amount of time on furling design, but I've seen quite a few turbine failures where the tail was underbuilt.  I also have never liked having the tail lower position stop in at the hinge-point.  It just seems like having one of those door stops that slips over a hinge pin, and eventually breaks the whole door casing.   I still may not have it exactly right.  My fear is the vane is still a bit too heavy, and I may go to AL.

I tried instead extending the top plate to form the stop.

Can you tell us how to make the homemade clipper?

Charlie, take a look here:
http://midniteforum.com/index.php?topic=7.msg20#msg20

We using a SS relay with a low voltage DC control input, but otherwise similar to Halfcrazy's approach.  I hope to change this to a load bank though, because I don't really like braking using the stator.  It is pretty abrupt, and IMO puts heat where you've otherwise worked hard to avoid it.

Thanks for the comments.  This place is been pretty quiet.  Chris had said earlier, "Not many folks building turbines here lately."  I suspect the cost, simplicity and comparatively lower regulations on PV has most folks questioning the sanity of this stuff.  Still, watching my solar panels make power is pretty dull .

Kind regards, ~kitestrings

[midwoud1]

Kitestrings -

Good luck with your new windturbine ,  with new ideas.

Rgds. -  Frans  -

[Boondocker]

I agree, it is fascinating watching those blades spin compared to panels.

[tecker]

Look like all that design came together nicely It's been fun and informative watching your progress thanks for posting

[clockmanFRA]

Looks good Kitestrings.

Regards Tail stops. I Love that top mounted low end stop bracket.

I have 3off 12footers to Hugh Piggotts design. This last Christmas blow, My No 2 Turbine had its tail come off when fully furled and broke a blade as it descended, but had been up nearly 3 years and was out of balance!.

So for me the fully furled stop is my main concern. Luckily with my No 1 turbine I hit the stop arrangement and welded the stop tab just right.
But the No 2 turbine was 20mm out, and there is a tendency to raise the tail off its hinge pipe spigot. Hugh in his 2011 edition mentions adding a 'safety catch' to the yaw pipe above the point of contact of the furling stop.
 
In the picture you can see my new stop bracket at a new angle and welded to the old high end/furling stop.

Real nice to see you getting things spot on.

[kitestrings]

clockman,

You may want to consider some sort of shock absorbing material.  I assume the curved cut-out mates against a steel tube or frame piece(?)  We  fitted a piece of adhesive backed neoprene rubber in the lower rest. 

For the upper stop/limit I tried to have the main tail-boom section align with the back of the spindle in the furled position.  This too has a rubber cover.

~ks

[Boondocker]

Kitestring,

I'm curious about the manual furling.  What is the routing of the cable down the tower, and is there a swivel in it to prevent twisting?

bd

[kitestrings]

Bd,

The winch cable goes thru a fixed block in the base of the tower - I think you can see that part - then up a few feet into the stub section I have a length of smaller pipe.  I forgotten the diameter, maybe 1 1/4", but it is essentially just a bit of weight to keep the line in tension.  At the top of this pipe is a small thru-bolt with a thimble.  The dyneema line terminates around the thimble.

At the center of the yaw head, just beside the SO cord & strain relief (which is dead center), I have a pulley that rotates to face the direction of the pull to the tail termination.  It is the same pulley used to actuate the clutch on many PTO winches if that helps (the pull there is on the opposite side, but it's the same idea).  The cord terminates on an eye-bolt on one of the tail thru-bolts where I can adjust the length of the boom.

There's no swivel above the pipe/weight, but that would be a good spot for one if needed.  Hopefully can see it here:

[kitestrings]

Pretty cool.  Notice that the bottom photo has the airfoil at the tip in complete shadow and if you look closely you can see the curve of the trailing edge, and deep cut, going towards the root of the blade.  Purely happenstance (my photos are sometimes hit -or miss), but I couldn't have planned the lighting any better.

~kitestrings

[kitestrings]

We've been stuck in a relatively   l o n g   cold snap and haven't had much for wind, or sun, until today...

This morning the wind was howlin' again and when I heard those familiar sounds - creaking of the wood shed (we have two live maple trees growing thru it), off-key broken wind chimes, compression of the window sashes in their frames - I came off the bed with a bit of a start.

It was only 4-5 AM and I couldn't find my glasses right away, but it became pretty clear when I did that we were having a 'weather event'.  Lot's to try and in before any coffee.  I took a number of screen shots, trying to record the rpm at each one.

The first Classic clicks on at 59V; the second awakens at 85V, but doesn't output anything until ~100V IIRC.  I got to see the first Classic hit its current limit, and the second one start to load up.  I had the setting on the 'clipper' relay intentionally low; about 155V.  Before it ended I got to witness the a short period where the relay was getting pulsed.

I also ran out to watch the furling a bit.  It's working, but still a bit later than I'd like.  Generally I want to try less vane weight, and a bit more boom length.  The latter could be just aesthetics, but the prop just seem to have a bit more of the battle.

By the time things settled down, we'd hit a high of 2,200w on the first unit and another 1,300w on the second.  Our first real test.  Exciting.  Terrifying.

~ks

[Mary B]

With gusts over 50mph here today(my weather station said 60) I would have been on pins and needles. Gets like this where the wind howls for a couple days I even begin to worry about the solar panels taking a beating.

[SparWeb]

Congratulations.  It passed the shakedown cruise.

Sunday night a front blew through my place, too, but by morning it had mostly died down.  It must have wandered on down south to visit you next.

[kitestrings]

Just a quick update on this project -

I replaced the tail-vane around the first of this month.  The original was made of #20 GA stainless.  As I'd mentioned the weight of the assembly (boom, inner boom, vane and strut) was over our target weight by quite a bit.

The new one is maybe not quite so pretty , but it is much closer to the 'design' weight/restoring moment.  Below is a the new vane.  It is .063 (~#16 GA) 5052 AL, mill finish.  The strut on the back is .090" AL.  It is rubber isolated form the boom.


We were able to drop about 7+ lbs. in total.  I also started this time with the boom extended out to it's center position - about 6" beyond where it had been.  Based on Hugh's most recent formula, this puts the restoring moment at about 65% of the rotor thrust at 10 mps (22 mph).  All theoretical, but closer to a reasonable starting point I think.  The verdict is still out, but it seems to be furling much closer to where I'd hoped it would.  And, I've still seen close to 4 kW out of it in some recent but not overwhelming winds.

I thought I'd share one other thing.  There's been some discussion here on how much 'skew', or angle is developed off the true wind direction particularly in low to -mid wind speeds.  I think it may be more than some folks have suggested.  We put the lower tail-stop at about 105 degrees as I recall.  Below are some pictures in 10-15 mph winds.  It's interesting.  Could be me, but it seems to track closer as the winds increase and the tail begins to lift off the lower stop.



Kind regards, ~ks

[SparWeb]

I tried that once too.
At the time the angle of the tail was just 90 degrees, and after I attached the ribbon, I adjusted it outward to about 110 degrees.



PS
Never put christmas lights on your tower.

[kitestrings]

Never put christmas lights on your tower.

I remember seeing them when you posted a picture a few years back (?)  Were they hard to maintain, or did you get flack from neighbors who want "property rights (theirs), but also height restrictions (for every one else)?

~ks

[SparWeb]

Actually the worked just fine, and the neighbours had way more lights on their houses than I did on my tower so nobody complained, or even mentioned it.
But over the years I read a few articles and a book by Paul Gipe that all said that one of the most annoying things on towers are lights.  And I have to agree, every time a cell phone tower goes up, there's one more stupid blinking light on the horizon.  So I wasn't helping the cause.

I probably live in a "regulatory" heaven here, where there are currently no rules controlling wind turbines, and various sorts of towers are permit-free uses on my property zoning already.  I phoned the municipal offices when I started the project, told them a bit about it, and they just said "it's fine".  How long this nirvana will last is hard to guess.  Someday, one level of government may pass a feed-in-tariff act, but 12 minutes later another government will pass a zoning restriction bylaw.  Until then, I just let my nearest neighbours know if I have plans, and so far they've been mildly interested and amused but not opposed in any way.  We all have our crazy projects here.

[kitestrings]

We had generally strong winds over the weekend, but on Saturday we had a strong snow squall, or 'clipper' as the weather folks like to say roll in.  The snow was heavy and the winds were hellish for maybe an hour or so.  It was a little scary to watch, but things seemed to work as hoped.  Makes me wonder what it'll be like when it's up in the air.  It's going to be like taking a solar panel out of a box.

The machine definitely furls sooner with the new vane, so I'm happy with that change.  I think this may be the highest instantaneous output we logged to date:

[SparWeb]

Nice numbers!

So if the battery side sees ~3300 W, then that's about 60 Amps into a 55V bank.  On the turbine side of the controller, what voltage do you think it's operating at?
If it's 110VAC line-to-line on the 3-phase side, then the coils are taking about 30 Amps.  I'd have to go back to your post last year about potting the coils to get the details, but I could just ask now: do you think that's close to peak power output before heat affects the stator?  Or do you think there's still a good safe margin yet?

[kitestrings]

SparW,

I'll check the logs when I get home.  I think it was up around 160V.  IIIRC, I have the clipper circuit at or about there, so that is the point where it starts to short the windings, and I believe it just started to tickled that relay.  My guess is that this was a pretty short-lived event, but you bring up a good point.  What I'm most concerned about is the capacity of the Classic.  It is a 250, so the max amp rating is 55-60A:

http://www.midnitesolar.com/pdfs/classicPowerChart.pdf

We're supposed to be current limited at 46 amps on Classic 'A'.  That is the 16th step on the power curve for the first one.  Classic 'B' comes to life at 85V, but it is set to pick up at the ending step of 'A'.  In round numbers, I'd we'd have peaks of ~5 kW, with each unit picking up potentially up to 2.6 kW.

The stator shouldn't be seeing more than a little over 20A (here), maybe 25A at peak, so we should be okay there, but tell me if I'm overlooking something.  The coils are 2-strand, 14#GA.  I'd hoped to run this by Ryan at MS.  Just haven't had a chance.  Maybe he'll chime in.

With a proper resistor load bank, I think we should also be able to influence the pace and magnitude of these highs.

~ks

[SparWeb]

Rule of thumb for wire gauges, doubling the number of wires takes you up 3 gauges.
So:  14 gauge X 2 wires = 11 gauge (give or take)

I sort-of remember that for 10 gauge the maximum current is about 30 Amps.  Check me on that.  The NEC numbers for household wiring don't really apply to the heat-loss conditions in the stator.  But using that ballpark figure then the equivalent 11 gauge would be okay for 25 amps or so.

And correcting my guess for 160VAC on the stator, then I get 20 Amps, like you got.  I'd say there's margin there, and that's all I was curious about.  I wasn't worried 

[kitestrings]

Rule of thumb for wire gauges, doubling the number of wires takes you up 3 gauges.
So:  14 gauge X 2 wires = 11 gauge (give or take)

I've seen this before.  What I never can remember is whether it is equivalent, whole number gauges, or what's readily available.  Here in the states anyway if I go into my local electrical supply house - we all know one another by name - they will look at me sideways if I ask for some #11 copper.  So I prefer to use my EE tables.

In this case 14# in code applications is rated for 15A (not more than three conductors in a raceway or cable...), but if you take the allowable temperature rating up to 85C it goes to 25A.  Alternatively, single conductor, free air rating is 20A and at 85C goes to 30A.  Either way at 25A we should be well below the point where insulation would start to fail.  And, I don't expect to be at that level for prolonged periods.  Does this seem like a logical assumption?

The diodes are rated at 85A.  The only heat meaningful that I can detect on the rectifier sinks seems to be from the one SS relay, which is normally energized.

Lastly, I did get some feedback from MS.  He said he this was pretty much the behavior that they'd expect.  We could reduce the current limit on 'A', but that would in turn limit the top end overall.

~ks

[CBabcock]

Beautiful looking blades and nice looking machine!

How has the furling line held up with the twist cable?  I'm thinking about doing that same thing but have some concerns that the furling line will be twisted up with the power cable and not operate when you need it (when the wind is blowing mach 90 and you want to shut the machine down manually)

Corey

[kitestrings]

Thanks.  So far it seems to be fine, but it also isn't mounted up very high.  It is really the first turbine that we've had that has not had slip-rings, and there was considerable back and forth as to whether or not to incorporate them on this one.  Enertech actually had some fairly large turbines with twist cords.

It's obviously simpler, but not without challenge.  I fretted over the furling details.  Originally I'd hoped to use the SO cord to actuate the manual furling, perhaps through a hydraulic pac, but I eventually gave up the notion (for this round anyway).

On this machine the twist cord has a strain-relief and it is dead center through the yaw head.  The furling line is maybe an inch or less off that hole with a short bushing welded though the cap.  The pulley pivots on this bushing.  It is connected with dyneema line which seems to be pretty low friction, low stretch stuff.  We also have a 'sash weight' on the line near the which just to keep it taught.  Mainly this ensures that it stays seated in the pulley, and does sag or catch on anything when the machine furls on it's own.

The manual furling is also not the only means of shutting things down, but I do like it because it is very 'soft' on things and doesn't rely on electronics.  We also have a braking switch, but I mainly added that as a parking brake if I needed to work on something and it is closer to the base of the tower.

 

~kitestrings

[kitestrings]

Not really what it is intended for, but it occurred to me that a switch like this - in this case it is a 20A DPST - would probably handle anything the turbine could throw at it.  That is, in this application where you are shorting the phase leads there is not the typical load-break, and associated arc.  My hunch is then, the limit of a switch in this configuration may be well above it's rated capacity, and instead dictated by what the amperage capabilities of the terminals and mechanism are in a closed-blade position.

Counterpoints are welcome.  ~ks

[ontfarmer]

Reading about your turbine a couple things I done on mine a few years ago. I put in a switch
 that if it over speeds it comes on putting extra load on it to help control it. I also made a latch that when it furls in high winds it hangs on to the tail this has worked very well.
                                                                                 Grant

[SparWeb]

Not really what it is intended for, but it occurred to me that a switch like this - in this case it is a 20A DPST - would probably handle anything the turbine could throw at it.  That is, in this application where you are shorting the phase leads there is not the typical load-break, and associated arc.  My hunch is then, the limit of a switch in this configuration may be well above it's rated capacity, and instead dictated by what the amperage capabilities of the terminals and mechanism are in a closed-blade position.
Counterpoints are welcome.  ~ks

That's roughly what I think, too.  A switch, with ratings given for (usually) 125VAC or 250VAC at 60 Hz, being used at voltages down below 30VAC, and below 30 Hz, either as an AC shorting brake or a dynamic brake switching in resistors, will experience arcs, if any, that do not stress or damage its contacts, even if the "mathematical" analysis of the resistance in the windings would lead you to think the shorting current is many tens of amps above the nameplate rating.  I don't push my luck though, and use a pair of 30 Amp 3-pole switches to brake.  If any one contact breaks, the other 5 are still shorted.

Some switches have different ratings for inductive, capacitive, and resistive loads (also referred to as "lamp loads" or "motor loads") which give an idea of the switch's ability to withstand both arcs AND continuous currents.  You only get that for expensive types of switches that have reliability specs, though.


...

Ontfarmer,
What do you do to release the turbine once the storm's over?  (release cable?)

[ontfarmer]

I have a small gear head motor mounted up there to release it. Today it is windy here and the turbine is running with the tail latched. The stop on the tail is set at near 60 degrees.
                                                                                 Grant