Ferrite Gen with a Classic 150

[ChrisOlson]

I got a Classic 150 controller in the FedEx truck on Monday and I couldn't take it.  Had to test it.  So late this afternoon I hooked it up to my test rig.

The best way to test it so I know what I got is in the shop driving the generator with my hydraulic motor.  This way, the only thing I can blow is a hydraulic hose - and those all held fine.  I did heat that Classic 150 right up though   

It's a long ways from my shop to the utility room in the house where the battery bank is.  But with the help of about 100 feet of #4 welding cables, plus another 100 feet of 8/4 old silo unloader cord I got the test rig in the shop hooked up to the bank in the house.  I tested the same generator a few weeks ago with the same setup, same wires, everything.  The below chart is from when I tested it:



When I tested it before, my bank was charged up pretty good and the voltage came right up.  But today it didn't.  The bank was down to 23 point something volts because my wife was running something big in the house.  I think it was chocolate chip cookies in the electric oven that was doing it.  Even with 90 amps going into the bank from the test rig, it never got much above 25.0.  I don't know if that skews the results or not, but it's what I had to work with.  The below chart is what I got with the Classic 150 all hooked up to the thing:



Just a couple general observations; on the bottom end I don't think there's much difference.  On the top end it takes a LOT less power to drive the generator to get 90 amps into the batteries than it does to get 60 amps with the turbine hooked direct.  The little GX270 Honda on the hydraulic pump really didn't even act like it was working that hard to get the thing up to 90 amps.  But the generator has to run a lot faster with the Classic hooked up than without it.

That's all I know for now.  I'm building a set of high voltage stators for the thing that will run it up 110 volts and I'll be able to use higher voltage right at cut-in (hopefully - I have to figure out how to make another one of them 16 step power curves for the new stators).  This test kind of gave me an idea of what it does.
--
Chris

[wpowokal]

Chris could you add a column that highlights the watts lost in the stator, I know it's in your info but for new members it seems to me a good opportunity to highlight losses. You being such a cleaver chappie and all.

allan (retired) now of the jungle

[Flux]

With the high voltage stator you will be able to loose most of the line cable loss and that is quite a bit. Letting the volts rise and stator current fall will also reduce the I^2R loss in the stator so with no other changes you should see a big difference.

Now you are no longer tied up with blade matching you need a different approach for the alternator design. I know you tend to just make the thing stiff enough to get the blade matching at present, but with the classic the thing to go for is the highest possible alternator efficiency and that effectively means making it as stiff as possible. With a transmission you have more options, but I would expect you to be able to increase alternator speed to advantage. All your other approaches to keep winding resistance to a minimum are still valid.

Another thing that may be in your favour is that you can probably get a better space factor in coils wound with thinner wire, large round wire doesn't stack well and rectangular wire is difficult to get and doesn't entirely solve the problem.

You may not see as dramatic improvement as some others, as you seem to have mastered the art of keeping your blades on maximum power with direct connection, but there is still a lot to gain from the electrical end. You may find that the high torque blades you use at present may not be the best here. If they will run with tip speed following wind speed they will be fine, if they want to work at lower tip speed on full load then a change of blade may be a benefit.

Having gone further than anyone else I know with improving the results with direct battery connection, it will be interesting to see how much more you can gain.  If you can't get any gain at all from the blades then taking the electrical efficiency from near 40% to possibly over 70 will show a considerable improvement.

Look forward to the results.

Flux

[ChrisOlson]

I am glad I didn't start the actual winding the stators before I tested my new controller.  I learned from it by seeing what it does and how the controller adjusts the "Steps" in the power curve.  If I wouldn't have done this I would've wound the stators wrong.  I also learned that I will need two controllers networked for this size turbine if I intend to push it very hard.  At the amp limit of the controller I was not anywhere near what the turbine could put out yet with bone stock stators designed for direct hookup.  I do not intend on doing that.  High outputs in high winds are useless for battery charging turbines unless you have a way to control it so you can use it to bulk and absorb charge the bank.  If you just have dump loads, then the dump loads are on all the time with those high outputs.

I am interested in seeing if I can get significant gains at 5-6 m/s wind speed.  I think I have a large leeway to increase the voltage a lot to gain efficiency there.

I never believed in this for a long time because I think it is a high wind thing only.  I am out to prove myself wrong, I guess.  Plus, as I found out, the Classic has a lot of real nice features in it with the computer networking capability and power logging and so on.  It appears to be a solidly build controller and well thought out.

--
Chris

[Volvo farmer]

I highly recommend you build a clipper for that Classic or you're going to let the smoke out of it.  One good episode into the 175 volt range and you'll be sending it back for repair.  Also, if you bump up against the amp limit, the controller will start disconnecting itself from the turbine and the voltage will rise quickly.

I find Flux's comments interesting. Next time I take down my 10 footer, I might remove all the washers that I used to increase the air gap back when I was battery tied.  I wonder if I would get more power at lower RPM if I did that?

[Flux]

I am sure reducing the air gap will gain you a fair bit as long as you re programme the curve. The smaller gap will give you more volts for the same speed and the converter can phase back to convert that extra voltage to useful current.

Flux

[ChrisOlson]

I highly recommend you build a clipper for that Classic or you're going to let the smoke out of it.

I already got "clippers" on my turbines.  These things have dual stators with only two leads from each one.  So I've been using regular old off-the-shelf Ford starter relays to short the stators one at at a time on the AC side.  Those starter relays are cheap (15 bucks), they handle 600 amps with ease and you can't weld the contacts in 'em with a wind turbine.  The coil in the starter relay draws more amps than what controllers can put out, so I've been using pilot relays operated by the controllers that in turn operate the starter relays.  My Morningstar RD-1 has been set up to "clip" them off during power surges when the bank voltage gets too high in danger of kicking the inverters out.  I short out one stator first and if that don't do it and the voltage goes .2 higher it shorts the other stator.  The RD-1 shorts them only long enough to "clip" off that top end, then lets them go again.  It's been working really good.

I think I can do the same thing with this Classic.  I just have to figure out if the Aux 1 or Aux 2 port on it can trigger the pilot relay based on input voltage to the controller.  I haven't gotten that far with it yet.
--
Chris

I just happened to see this:
The leading horse is white
The second horse is red
The third one is a black,
The last one is a green

LOL!  That was my chuckle for this morning.  Good one.  LOL!

[jarrod9155]

My old turbine had a very close air gap , with my new direct drive I opened the air gap up about 3 millimeters and wound the stator with 18 guage instead of 17 . I wish I would of stayed with the close air gap and  bigger  wire . I don't have a classic but do have mppt with 16 steps and have had a problem with the goe222 blades it seems the blades want to run away and the mppt applies load to the point of stalling the turbine , this mostly in 15 mph + winds  and then partially furling really gets the inverter searching for the proper load . Opening the gap did help in my case in 6 mph to 15mph .

[ChrisOlson]

jarrod, the blades are more than likely tending to want to run too fast because of higher internal resistance in your new stator and a bigger air gap, plus the MPPT trying to speed them up.  That's my guess anyway.  Those blades are a little weird because they don't like to run too fast.  Right around TSR 5.7 is pretty ideal, but you don't really lose much if you lug them down.  You lose big if you run them too fast.  And they will run VERY fast, but they don't make as much power to the shaft because it gets eaten up in aerodynamic drag.

My test with the Classic showed me that having decent electrical efficiency in the generator is still important with a MPPT power curve.  You can improve the electrical efficiency by bumping up the voltage, but that requires more rpm's.  If you get the rpm's too high for the blade profile, it's not going to work.

I think there's a balance between squeezing the most out of an MPPT design, just like there is in squeezing the most out of a direct hookup to battery design.  I think at present, most builders are putting MPPT on a machine that was originally designed for straight up battery charging, and enjoying the efficiency gains simply due to the fact that the original design was poor to begin with and just ran the rotor blades into a stall situation under heavier loading.  I'm going at it a little different and trying to optimize a design specifically FOR MPPT.  What I'm coming up with is that this is going to require ridiculously high voltage with internal resistance as dead low as you can make it be.

I've been talking some with Hugh and Flux and I think we sort of came up with a single stator ferrite design that will have the required attributes for good MPPT performance - ability to wind for really high voltage with really low relative internal resistance.  The more I look at it, the more I like it for MPPT, after trying out the Classic to get an idea of what it takes.  I'm looking at cutting in at really low wind speed on a 24 volt system at 40-50 volts, then using the MPPT power curve to keep the rotor at TSR 5.7 all the way to about 10 m/s.  I'm pretty excited about this - I think it will work.
--
Chris

[Volvo farmer]

I am sure reducing the air gap will gain you a fair bit as long as you re programme the curve. The smaller gap will give you more volts for the same speed and the converter can phase back to convert that extra voltage to useful current.

Flux

Thank you! As long as I have your attention...

What would be the effect of putting eleven foot blades on this thing? It's a standard Danbuilt 10 footer. I know there would be more power to be had in all wind speeds because of swept area. I also know that it's not likely for me to burn up a stator using MPPT. Do you have an opinion on whether the framework and furling system would be overly stressed by adding 10% to the swept area, or is there something I haven't taken into consideration?

[ChrisOlson]

on whether the framework and furling system would be overly stressed by adding 10% to the swept area, or is there something I haven't taken into consideration?

By going from 10 feet to 11 feet you're actually adding almost 22% to the swept area.  That's a lot.
--
Chris

[boB]

  I also learned that I will need two controllers networked for this size turbine if I intend to push it very hard.  At the amp limit of the controller I was not anywhere near what the turbine could put out yet with bone stock stators designed for direct hookup.

--
Chris

Networking for multiple controllers on one larger turbine is coming very soon as a software update.
This should work fine but will require one bridge rectifier per Classic but at least
diode drops are not an issue for AC output turbines.  Hopefully this networking and semi-parallel
operation will be ready as soon as you are ready to try it which should be a month or two at the
latest.  This has been tried already without the networking and worked pretty well but coordination
between 2 or more controller will work much better.

boB

PS.  My funny for the day.....

I'll never forget those eyes of hers....
One eye was green and one eye was gray...
The other eye was brown.

Spike Jones and his City Slickers

[Flux]

I would seriously think whether you have enough alternator offset. Unless you can increase it be careful. There could be more tendency to wind seek with the higher tsr of the classic compared with a machine designed for stall loading.

If you have enough offset or can increase it then fine go for it.

Flux

[kitestrings]

Chris,

Nice to see this.  I can tell you we'll be most interested in your findings, as we're move much slower on our turbine, and few of us have the resources and know-how you muster.  I'm anxiously waiting to see more.

I'm confused by one comment from Flux,

All your other approaches to keep winding resistance to a minimum are still valid.

In an earlier post I was exploring what design changes one would make knowing that MPPT was to be incorporated, as we intend.  I think you said effectively that you would build stiff, and not skimp on magnets, and that there was little benefit from [lowering] line resistance.  I took this to mean that more copper was also benifical as the Classic (and clipper/deversion) could preumably limit the current to reasonable levels.  Maybe I missed something?

regards,  ~kitestrings

[Flux]

Not sure if this was aimed at Chris or me. I am not sure I follow the question either but I will try to answer.

For direct connection you need a significant loss to get some sort of blade match and so it is not a bad idea to keep the stator resistance low and have a bit if extra resistance in the line ( better to heat that than the stator).

Now with mppt there is no reason to have any excess resistance anywhere as you can keep a very high electrical efficiency and not have to worry about blade matching.

That said, the line resistance is far less critical with mppt as the volts rise with increasing wind speed so the line will carry more power at high voltage with less loss.  If you have a 24v battery it is tricky to avoid a lot of line loss with direct connection unless very short, the cable cost gets crazy, but if the input at full load is over 100v then the same cable will give quite small loss.

Machine resistance needs to be as low as possible to get the best results, but even if it is higher than desireable you still have far less stator heating. it is all the heating losses that you can convert to extra useful power.

Unless you have done something similar to Chris you will see a dramatic improvement from blade matching even if you don't make the best use of the electrical efficiency. If you have got good blade matching in the first place then you can gain a fair bit if you go for the best electrical efficiency.

A point that Chris mentioned earlier is not quite so obvious, it is generally assumed that the conventional machines work well in light winds and only really loose out in high wind so the mppt may not be expected to give much improvement in the 12 mph region.

I have found that there is in fact quite a worthwhile gain in this region and I believe it to be due to the fact that the prop power curve is so flat in the region above cut in that if you don't have mppt you have to run the prop too fast at cut in, it comes about on design tsr at near 10 mph and by 12 to 15mph you are below peak and on the way to stall.  It is in this region that the converter is phasing back rapidly and the benefit is a fair improvement where you might not expect it. I have seen no figures for the classic but I don't see why it should be any different from my crude scheme.  For direct connection any attempt to cut in below 7mph will murder the performance in the main operating region but that limitation is removed with mppt. There is not much in wind below 6mph but you might as well have it when it doesn't cost you anything in the higher winds.

Not sure if this answers your question but i think I covered most things.

Flux

[kitestrings]

Thanks Flux.  This is very helpful.

Yes, as I've understood it with direct-tied applications we are intentionally building in some alternator inefficiency (thru less than ideal magnet flux, air-gap and/or higher resistance windings) and/or adding line resistance to strike a balance on load matching particularly in the lower rpm bandwith.

If, on the flip-side, you're going into the design with MPPT available - as Chris is here - as apposed to retrofitting to an existing design, it seems the design parameters change (and the potential gain is greater).  In this case a high-flux, high-resistance, higher voltage, close gap alternator with low line loss system is ideal, correct, though as you say the latter becomes less critaclal?

I think what I missed earlier was that you were talking about line resistance, controller to trubine, and I misinterpreted this to be mean winding resistance.  Asked a bit differently, if Chris has his desired coil/cut-in voltage through proper number of turns, and could fit another strand of wire, all else being equal there would be any benefit to more copper?

Thanks, and again Chris, look forward to seeing your results.

~kitestrings

[Flux]

"In this case a high-flux, high-resistance, higher voltage, close gap alternator with low line loss system is ideal, correct, though as you say the latter becomes less critaclal?"

No you don't want high resistance, the best case would have none.

Don't get confused here as with a higher voltage winding you can tolerate a higher resistance for the same loss, but don't deliberately try to add more than necessary.

About adding turns to the stator then that is a different question. Adding more turns as such will get you more volts and that is not a disadvantage if you don't have to worry about prop matching.

If there is more room for copper then it probably isn't an optimum design and most likely it would benefit. The old problem still occurs, the volts you gain from extra copper is beneficial but the extra copper will add resistance. In most cases the net will be a gain, but if the added copper adds more resistance than you gain from the extra volts then it is bad. good design is all about having the maximum voltage with minimum resistance. That is one point where overlapped coils seems to fail, the end connections are long and add too much resistance.  adding turns in the centre of the coil increases resistance very little and if you can still gain volts then it usually works.  Wide spaced magnets gives you more winding space but leads to longer wire lengths and it is a tricky trade off between thicker wire ( less resistance) and longer turns ( more resistance).

Doesn't help much in that there are too many variables to give a definite answer, but in general you should use all the space available for copper and coils should touch. If you don't you have to use thinner wire than ideal to gut the turns you want.

Fortunately as long as you get things basically right there is more tolerance with mppt, but the ideal solution will use more magnet and copper than the cheap match for direct connection. The mppt version will not be much limited by stator heating but the cheap direct connected version will as all the necessary matching losses will be in the stator. If you wound a direct stator with too low a resistance and matched it with line resistance than you would have an alternator that is a far better starting point for mppt, but again it will have more magnet and copper.

Flux

[ChrisOlson]

If, on the flip-side, you're going into the design with MPPT available - as Chris is here - as apposed to retrofitting to an existing design, it seems the design parameters change (and the potential gain is greater).  In this case a high-flux, high-resistance, higher voltage, close gap alternator with low line loss system is ideal, correct, though as you say the latter becomes less critaclal?

kitestrings,

What I've come up with here is that a MPPT specific design still needs the generator efficiency as high as possible, i.e. low internal resistance in the generator winding.  At the same time it also needs much higher voltage (which will also help with higher generator efficiency).  The line loss issue becomes less of a problem, especially on lower voltage systems like 12 or 24, because the MPPT curve allows you to use smaller conductors to keep the voltage drop on the line closer to <3%.  Therefore I believe MPPT power matching will provide greater gains (as a percentage) on 12 or 24 than you can achieve on 48 or 120 volt DC systems.

In generator design, using neos will yield a lower resistance winding than using ferrite magnets.  The challenge for me is to build a low resistance, high voltage generator with ferrite magnets for MPPT.  At the low end it's easy to get generator efficiency >90% and the only gains available with MPPT is to run the voltage as high as possible right at cut-in to reduce losses in the line and rectifier at these lower current flow rates.  I have determined that is possible to do.  Testing the Classic with a stock generator showed me that if I can run the voltage up to 40-50 VDC at cut-in, I can get more power from my turbine starting right at cut-in simply by reducing the transmission and rectifier losses.  It's not a lot on the low end because the controller itself eats about 5 watts right off the top doing nothing.  At low amp output rates the controller appears to be pretty efficient - greater than 97%.  As the amp output of the controller goes up it becomes increasingly less efficient, but the huge gains in generator efficiency and power transmission at the higher output rates greatly outweigh the losses in the controller.

With neos this would be a snap.  With ferrites it's going to more of a challenge to come up with a design that can take advantage of it.  But I'm "hooked" on building generators with ferrite magnets so I'm going to take a stab at it and see how close I can come to what all my figuring says I need    
--
Chris

[halfcrazy]

Chris
Another cool part to this is the one Stator design you are shooting for 40 ish volts cut in would actually charge a 48vdc bank although it may cut in a touch late. this means with one Standard Turbine you can charge a 12,24,36 or 48 vdc Battery less Stator combos to have on hand. Of course the way that machine puts out power you may need 3 Classics on a 12vdc battery

[ChrisOlson]

Another cool part to this is the one Stator design you are shooting for 40 ish volts cut in would actually charge a 48vdc bank although it may cut in a touch late. this means with one Standard Turbine you can charge a 12,24,36 or 48 vdc Battery less Stator combos to have on hand

I know.  And believe me, I already got that all scoped out.  I'm going to shoot for 50 volts at cut-in, then the same stator will work on all except a 120 volt system, and those aren't that common.   
--
Chris

[kitestrings]

Thanks for the clarifcations, and follow up.  To be clear, I was not suggesting increasing the number of coil turns, rather the effect of reducing the coil resistance by adding another strand (2 in-hand say), or increasing the wire size if it fit.

Best of luck with it.  With the recent trend in neo pricing your efforts should be of interest to a lot of homebuilders I'd expect.

reagrds, ~ks

[ChrisOlson]

Hi guys,

I'm still working on this new turbine but haven't had much time to post details on it.  Mostly doing a lot of testing and figuring instead of building.  I've reduced the rotor diameter to 3.2 meters because I feel that size blades is better matched to the power output capacity of the Classic controller on a 24 volt system.  I've also decided that I will not use a dual stator gen on this one because I can get the higher voltage that I want for this project more efficiently with a 16 pole three-phase.

I built a different gearcase for the larger gen that has an offset PTO shaft:



The type of hub for the blades I'm using has a tapered bore so I built a new input assembly for it:





This is the basic layout for the generator - it's a conventional 16 pole 12 coil flat three phase



For the smaller blades and .400:1 gear ratio I need to get the polar moment of inertia as low as possible in the rotating assembly, so I went on a weight reduction program for the rotors.  The rotors are 355 mm diameter and I got this rotor down to less weight (without the magnets) than the neo rotor laying alongside it.  I machined it to a thickness of 5.2 mm.



That's all I know for now.
--
Chris

[bj]

   Nice as always Chris.  Like the tapered shaft.  I always come to some grief doing those.  (make some scrap).

[ChrisOlson]

  Nice as always Chris.  Like the tapered shaft.  I always come to some grief doing those.  (make some scrap).

It's a 24 mm x 30 mm 1:10 taper on the nose of the input shaft.  I machined the shaft from 4140 chromemoly steel.

I should add that I don't particularly care for those tapered hubs.  Once you seat that hub on the shaft you can take one blade off the turbine, take the bolt out of the end of the input shaft, and run the turbine in a hurricane with two blades on it - it'll tear the tower down but that rotor won't come off the shaft.

I got a special puller I made to pull those PowerMax tapered hubs.  You crank down the puller with a breaker bar with a 4 foot cheater on it and when it finally comes loose it lets go with a big BANG!  If you got the blades still bolted to the hub you're suddenly hanging on to the whole rotor because when it comes off it shoots right off the end of the shaft.

--
Chris

[scoraigwind]

Hi Chris

Here's another way to do the coils.  Leg width is 30 mm.  This style of 'concentrated' coil makes sense to me where the magnets are jammed in like that. Inner turns are very short, but still produce some voltage.  The triangular hole could probably be a bit shorter than I have drawn it.

[ChrisOlson]

Here's another way to do the coils.  Leg width is 30 mm.  This style of 'concentrated' coil makes sense to me where the magnets are jammed in like that. Inner turns are very short, but still produce some voltage.  The triangular hole could probably be a bit shorter than I have drawn it.
(Attachment Link)

Hi Hugh,

The coils shown in the photo are wound with 45 turns of 13 AWG wire.  The pin dimensions are 47 mm top to bottom, 46 mm wide at the top, 13 mm wide at the inside point.  I wound these coils for a 10/10 dual stator, then decided at the last minute to lay it all on my CAD (Cardboard Assisted Drawing) to see how it would fit for a three phase 16/12.  Upon realization that I don't need the ampacity for this gen but I could use more voltage.

This is the second three phase 16/12 ferrite generator I have built.  The first one had bigger rotors with a gap between the inside corners of the magnets and the coils were more rectangular with a wider pin spacing at the inside.  I was not happy with its performance compared to the dual stator ones.  It was pretty weak.  I decided that having the corners of the magnets touching is not that big of a deal.  I get about the same flux in the air gap as I would have if I used wedge shaped magnets.  And it lets me get the diameter down to 355 mm.

Now, if I used the layout shown in your very nice drawing, how would that affect the resistance vs voltage?  Using those "fat" coil legs, I'm afraid, would add up wire length too fast on the outside turns and negate the advantage of the shorter length/turn on the inside.  It might reach a point of no return where I would get more voltage by having more space to wind turns, but the increased resistance might eat up any voltage advantage?  I don't know that - it would have to be tried.

According to what I have calculated, this thing as it is shown should pretty easily reach 100 volts with 45 turns of wire at 10 m/s wind speed.  I didn't measure the width of the legs of the coils shown in the photo but I think they are about 20 mm after they get smashed down to 10 mm thick.  I have shortened the top to bottom pin spacing from the first ones I built and have gained by less resistance with no apparent loss of voltage by doing that.  I'm pretty sure it could be shortened even more.  But the limiting factor in the diameter of the generator becomes the magnets, not the room for coils,  because these ferrite blocks are so huge.  355 mm is as small as you can build a 16 pole using these particular magnets.  Since there's not really any lack of room for wire at that diameter when you're building a generator with a 25-30 amp continuous capacity,  it would be fun to play with some more coil shapes and variations to find what the optimum is   
--
Chris

[kitestrings]

Nice project, Chris.  a few quick questions, just exploring your approach/thinking:

I'm curious about your size selection for the magnets.  They look to be pretty much square.  Was this based on economy/availability, or was there a design consideration that prompted you to choose them?  Otherwise, I wondered why you wouldn't have picked something more rectangular?

It appears that you have a little bit of verticle angle (I assume for blade/tower clearance) at the point of attachment to the yaw-tube assy.  My basic question here is how much of an angle you typically use, and whether you prefer this to moving the rotor forward more (and the effect that has on furling)?

Lastly, if you're not using the dual stator, how are you planning to "clip"?

Regards, ~kitestrings

[ChrisOlson]

I got those magnets because of availability.  I would like wedges for this three phase but I would have to have them custom built.  I got the 2 x 2 x 1" thick blocks for $1.84 each.  I'm basically "wasting" the corners of the magnets by arranging them with the corners touching.  But I determined I get the same flux in the air gap this way as I would have with the less surface area of custom built wedges, so it's no big deal.

The input shaft of the turbine is inclined 6° from horizontal.  This machine has the gearbox mounted forward of the yaw to provide better balance on the yaw shaft with the heavier ferrite generator.  The rear of the blades at the hub are 200 mm from the yaw centerline and with the 6° tilt on the input I should get better 30 cm clearance to the tower at the blade tips.  I'm not as concerned about clearance with this one because the rotor I'm using is a right hand turner and the head is built with left yaw.  The other ones like this are left hand turners with left yaw so I moved the rotor out further on those.

As far as furling, I don't want this one to furl as aggressively because I'm going to push it to better than 2,500 watt output with a 3.2 meter (10.58 foot) rotor.  I'm also not using spring loaded furling on this one - it will have a angled tail hinge with the tail working against gravity so it aggressively faces the wind except in the most severe high wind speeds.

Since I've decided to go three phase on this I'll build a three phase resistor wired wye, and clip on the AC side of the rectifier based on input voltage to the controller, and using the controller's Aux 1 output to drive the clipper relay.  It will be basically the same as "Ryan's Clipper" shown in the controller manual.
--
Chris

[kitestrings]

Wow, that's a huge savings over the NdFeb's.  We bought ours quite awhile a ago (.75" x 1.50" x 3.00"), but I'm thinking they were ~$20-22/ea.

Since I've decided to go three phase on this I'll build a three phase resistor wired wye, and clip on the AC side of the rectifier based on input voltage to the controller, and using the controller's Aux 1 output to drive the clipper relay.  It will be basically the same as "Ryan's Clipper" shown in the controller manual.

That's our plan as well.  I'm going to move our diversion load over to the Aux 2 (PWW relay).  It is currently on the solar CC; a trusty MX60, but it doesn't have the solid state relay/PWM choice that they later incorporated.  The clipper will operate off the Aux 2, but we also have an Omron voltage sensing relay (like Rob Becker's advocated), so if life-blood to the Classic, or component fails we have a "hi-limit" relay on the thing.  My latest thinking is to close on the load bank at the same time if/when this occurs.

I never used to rely on electronics; it still makes me nervous.  We've come a long way.

Thanks for the reponses, and keep us posted on the progress.

~ks

[ChrisOlson]

Wow, that's a huge savings over the NdFeb's.  We bought ours quite awhile a ago (.75" x 1.50" x 3.00"), but I'm thinking they were ~$20-22/ea.

Oh, for sure.  It does take more mags with the ferrites because, for example, I'm using a 16 pole gen compared to normally using a 12 pole gen with a ~10 foot turbine.  Plus I'm using gearing to get the voltage, where with neos it wouldn't be really worth putting gearing on it.  So there are some basic design differences, but the ferrites are more fun and more challenging and I can buy enough magnets to build two turbines for only ~$120.

That's our plan as well.  I'm going to move our diversion load over to the Aux 2 (PWW relay)

I'm not totally sold on going to the extent of using PWM on Aux 2.  I think I'm going to prefer to use a regular old two pole contactor and just slam it shut with the output from Aux 1 to "clip".  I look at the "clipping" as an emergency thing to keep the voltage from going over the 150 volt limit, and not necessarily trying to regulate it at the clip point with PWM.  I'm going to run the gen at around 100 volts at full output on this turbine, and that will only require about 27-28 amps at the top limit of what the controller will handle for output amps on a 24 volt system (not figuring losses in the controller and so on).

Since I'm well under the voltage limit with the turbine loaded, if it gets to ~140 volt then that means I got a problem someplace - the machine didn't furl, the breaker on the Controller blew and unloaded the turbine, or something similar.  This should never happen.  And if it does, at that point, I'm considering just tripping a latching contactor or relay to throw a 3 kW load on the turbine, which won't shut it down but it'll severely brake it even in strong winds.  Once it trips, it requires manual intervention from me to fix the problem and let it go again.

That's why I said it's similar to Ryan's setup, but not exactly the same.  Ryan's setup with the Teledyne Triac attempts to regulate the voltage at the upper or "clip" limit.  I think his setup is better if you design your turbine to run right at the raw edge of the over-voltage limit of the controller.  I'm not doing that, but instead "clamping" the voltage down on the turbine when it reaches around 2 kW output to start dropping the TSR of the rotor, reduce the power efficiency of the generator a bit, and limit the top speed of the rotor to about 400 rpm.

It would be easy enough to let it spin and run the voltage up to 140.  But I don't want a MPPT "Buzz Bomb" on the tower.  I'm not all that interested in getting big high wind numbers - more interested in tuning for the midrange  So my power curve for the controller cuts the turbine in at 100 rpm with the rotor running at ~6 TSR, then ramps the voltage up very quickly right off the bottom to keep the rotor at 6 TSR up to about 20 mph wind speed.  It might sound stupid to cut a 10 foot rotor in at only 100 rpm, but with the Classic you can do it with no danger of putting your turbine into "stall" - and you don't have to overspeed it to 8.5 TSR at cut-in to make the turbine work properly.

Yes, I have my distrust of electronics too.  But the more I tested and learned about how the Classic works, the more I like it.  It took me a long time to even convince myself to try it, and I came up with every excuse I could think of why it wouldn't be any good.  But, frankly, this is no cheap Chinese controller slammed out with substandard components in it.  This thing is built like a tank and it's pretty obvious boB didn't cut any corners on designing it.  So I'm learning to trust it because it's the coolest thing I've ever played with   
--
Chris

[boB]

Kitestrings wrote: That's our plan as well.  I'm going to move our diversion load over to the Aux 2 (PWW relay)

I'm not totally sold on going to the extent of using PWM on Aux 2.  I think I'm going to prefer to use a regular old two pole contactor and just slam it shut with the output from Aux 1 to "clip".  I look at the "clipping" as an emergency thing to keep the voltage from going over the 150 volt limit, and not necessarily trying to regulate it at the clip point with PWM.  I'm going to run the gen at around 100 volts at full output on this turbine, and that will only require about 27-28 amps at the top limit of what the controller will handle for output amps on a 24 volt system (not figuring losses in the controller and so on).

  
--
Chris

I'd say you're absolutely correct on using the NON-PWM  Aux 1 output to clip if using a relay.   Aux 2 and its PWM is suited more for SSR Solid State Relays and the like.  Even triacs.


And, I am happy that you are liking the Classic and are leaning towards MPPT.  After one plays with one for a while, it becomes more and more obvious how this works and
how it can help, either with the low or the high end of the wind speed spectrum, or both.

Look for improvements in the Classic's operation as well as time goes by.

boB

[kitestrings]

Chris, boB,

Aux 2 and its PWM is suited more for SSR Solid State Relays and the like.  Even triacs.

I don't think I was very clear.  We're currently using a water heater (with new port bushings and DC elements) to pre-heat when excess solar, and indirectly wind, is available, but this is all on the DC side via SS relays.  Described more here: http://www.fieldlines.com/board/index.php/topic,130053.html  I'm thinking to increase the diversion capacity and move the control signal to the Classic Aux 2.

On the wind side, we plan to clip off Aux 1 into a resistive load bank in much the way you've [Chris] described.  I do like the overall approach i.e automate diversion where applicable, then clip on the AC side for Vmax protection using a load other than the windings - that's my preference anyway.  I've never liked shorting the windings other than for parking.

boB, I think it is quite simailar to what Edward (in OR) is doing.  I coresponded some with him (& you) on the MS forum awhile back.

happy holidays, ~kitestrings

[ChrisOlson]

On the wind side, we plan to clip off Aux 1 into a resistive load bank in much the way you've [Chris] described.  I do like the overall approach i.e automate diversion where applicable, then clip on the AC side for Vmax protection using a load other than the windings - that's my preference anyway.  I've never liked shorting the windings other than for parking.

Yes indeed, I did misunderstand.  What are you going to use for a load on the AC clipper, and how are you going to match it to your turbine power?

I was thinking about a 3 kW load would be right for thiis 3.2 meter machine.  If it got to 140 volts, where I plan to "clip", it would be putting out about 21 amps into a 3 kW load.  When the load comes on (in the event something goes awry to cause the clipper to activate), the voltage will instantly drop and there will be an initial huge amp surge as the turbine comes back under control.  But if nothing happened to the furling and the machine just became unloaded because of a blown breaker, etc., a 3 kW load should provide enough thrust from the rotor to furl it and get it back under control.

I figured three 3 ohm resistors wired wye should provide an approximate 3 kW load @ 140 volts.  As the voltage drops, then my 3 kW load becomes only a 1600 watt load @ 100 volts and a 600 watt load at 60 volts.  But looking at my calculated power curve for the turbine, this matches it pretty close with the rotor at 6 TSR.

That's my thinking anyway.  It has to be tested to make sure I figured right.
--
Chris