Looking for ideas for wind charge controllers

[Rob Beckers]

Does anyone have ideas for a wind charge controller that will work with a 48 Volt battery bank and that can handle 100 Amp?

What I have in mind is something along the lines of the TriStar TS-60. This is primarily a solar charge controller (60 Amp max), but it can do wind/hydro charging as well. It also has a diversion-load mode, though I don't need that for my particular case. If it would do 100 Amp it could work for my purposes, as it is now it's too small.

If anyone is using a solar charger for wind/hydro (the TriStar or anything else) please let me know what you're using and how it's working out.

Along those same lines: Does anyone know what Bergey uses as a charge controller for their Excel-R? That would be about the right size for my purposes.

Thanks!

-Rob-

[Wilco Vercoelen]

What about Magnetek MWI-5200? or Maybe the NC 100 at http://www.eirbyte.com/flexy25.html
What is the Voltage input range for your controller?

[Rob Beckers]

Thanks Wilco!
Does anyone have experience with the NCHC100 and a wind genny?

Actually what I'm looking for is a replacement of the MWI-5200. Magnetek has been bought by Power-One recently, and Power-One just announced they are discontinuing the MWI-5200.

That aside, the MWI-5200 always was (expensive) overkill for battery charge duties. It contains an inverter, gen-set controller, gen-set charger, and PV charger as well, none needed for 'just' battery charging.

So, I'm using the opportunity to see if anyone knows of a way to do simpler and cheaper battery charging (compared to the MWI). The source is 3-phase AC, 0 ... 240 volt, 0 ... 6.5 kW. That can be modified to any voltage using a 3-phase transformer, and it can be rectified to DC using a simple 3-phase bridge.

-Rob-

[Rob Beckers]

What inverters can do islanding, and then backfeed/charge the batteries from the grid (or rather, the island in this case)? If I remember right there are Xantrex inverters that do this, unfortunately I can't find them when I look at their site.

The idea is that another way to charge the batteries through wind is by using an inverter from the wind turbine, to backfeed an islanding inverter. Matt Tritt came up with that actually, now I'm trying to find inverters that will cooperate for such a scheme.

Of course, if anyone is already doing just that I'd love to hear about it!

-Rob-

[Rob Beckers]

To answer my own question: The islanding inverter is the Sunny Island from SMA.

The other part still stands: Did anyone try backfeeding over an islanding inverter?

-Rob-

[Logan Bryce]

Hi Rob

Because I am limited to 10KW on my net-metering, My Eoltec is "downstream" from my OutBack power panel. I have been using a Micro-grid configuration from day one. The Power-One inverters sync with the OutBacks and then the power from the Eoltec is used to supply my loads and any excess is then used by the OutBacks to charge the battery bank. It works seamless. The drawback is that you must have a dump-load the will be able to dump the fully load from the turbine or you WILL overcharge your battery bank.

In the next 6 months MidnightSolar (www.midnightsolar.com)will have the MPPT Classic Wind controller that will be able to be used with the Eoltec. The turbine will be listed in the controllers menu. You will have a slight power loss as we must use a transformer to reduce the high end DC voltage.

Logan

[Rob Beckers]

That is interesting info Logan!
One question though: Are you on-grid or off-grid? I can see this working on-grid. To be off-grid the Outback would have to do islanding (Does it? I've not studied their docs close enough to find that out). Regular inverters are made to do anti-islanding, so when it sees no load on the grid side it shuts down the grid-tie (in fact that's pretty much at the basis of UL1741, so linemen don't get electrocuted) and consequently the Aurora of the Scirocco would shut down too. If the inverter could be told to do islanding, like that Sunny Island I mentioned, then one could use the Aurora also for off-grid.

By the way, another interesting product that could be used as an inverter for grid-tie, or to do the same trick as you're doing, is the WindyBoy. They make a 6kW version that accepts rectified 240 AC (ie. the input can handle the resulting DC voltage), and it's MPPT, UL1741/CSA listed. Those things go for about $3.5K retail, so it might be an alternative to the Aurora (and having multiple sources for inverters would be a good thing too).

The guy that founded MidnightSolar comes originally from Outback if I remember right, so a charge controller coming out from him would be very interesting. Having looked at this now for some time I really thing the market is overdue for an MPPT wind/hydro charge controller. Pity if they can't build the input stage to directly handle higher voltages, that would make it the perfect product.

-Rob-

[Logan Bryce]

Works just as well off grid as it does on grid. It sees the OutBack power as Utility power. Would still need Tristars as dumploads to keep from over-charging the battery system. Works on 24 volt AND 48 volt systems.

The Classic will handel voltage up to 200V DC.

The WindyBpy is proven with the Scirocco also but not as good as it only has 2 MPPT setpoints.

Logan

[Rob Beckers]

Works just as well off grid as it does on grid. It sees the OutBack power as Utility power. Would still need Tristars as dumploads to keep from over-charging the battery system. Works on 24 volt AND 48 volt systems.

Hmm... I've been browsing through the Outback inverter manual a bit, and as close as I can tell their inverters will not do islanding. That means you can't use their inverters off-grid and expect to charge the batteries down-stream through a wind gennie plus inverter (like the Aurora).

When grid power goes out the grid side of the inverter shuts down. It seems it'll only charge the batteries from the grid side of the inverter, not the backup/off-grid side of the inverter. So in an off-grid situation you can't hang a turbine on the grid side of the inverter and expect it to charge batteries. Correct me if I'm wrong of course.


The Classic will handel voltage up to 200V DC.


That's a pity... If they changed the design to handle 600V we could do without the transformer. That's one big hunk of expensive iron and copper. By the way, what's the power rating on their charger?


The WindyBpy is proven with the Scirocco also but not as good as it only has 2 MPPT setpoints.

Logan

Yeah, I saw that. Just a straight line for MPPT. Sometimes you have to wonder who specs the design of these things. There's already a microcontroller in there, so to do interpolation on a 16 point (or so) table would be trivial.

-Rob-

[Logan Bryce]

I can assure you that it works well in this configuration. The output from the Eoltec is on the supply (inverter) side of the inverter. So, when the inverter shuts down the grid side the turbine inverter is still connected and push power to the loads and any that is left is used by the inverters charger.

Logan

[Wilco Vercoelen]

Hi Rob,

The Outback G-series (GTFX and GFX are tested by ETL (not UL) in compliance with UL-1741. CSA 22.2 possible with a CSA Sticker which can be purchased for Powerboards. With the (internal) Automatic or some cases extra external Manual (center-off) or Automatic Transfer Switch this can be accomplished. In grid-tie mode (charging mode) the grid feeds the loads and in inverter mode (Grid absence) the inverter will feed the loads the other way. Both Xantrex and Outback can do this. Xantrex SW series with GTI (Grid-Tie Interface). Outback has a better sinewave than Xantrex. Xantrex is actually UL and CSA approved. Logan's configuration is possible using the Outback as Virtual Grid in charging mode. When the wind turbine is not producing (grid-absence) the loads are covered by the Outback Inverter in Inverter mode (internal transfer switch) and therefore Islanding.

[Rob Beckers]

Just to add to the little list here: Xantrex also makes charge controllers that can be used in diversion load mode. In particular the C60. So two of those could also handle the load I need. Not sure if there are any advantages (or not) of Xantrex over TriStar, haven't looked at their price either.

-Rob-

[Rob Beckers]

Well, this saga is still continuing. One option that's being examined closer is to use an NCHC-100 from FlexCharge (http://www.flexcharge.com/flexcharge_usa/products/nchc/nchc.htm). The heavy lifting in this controller is being done by a relay, making it less likely that electronics get damaged in case no diversion load is used (with an alternator as the source). Relays can also be put parallel for increased capacity. Basically it's just a switch that's set to close at 54.8 Volt, and open at 57.6 Volt, as measured from the batteries. Here is a picture of it:

http://www.flexcharge.com/flexcharge_usa/products/nchc/nchc_24_photo.jpg

Since there is no 'set' voltage that the controller maintains this begs the question how it is going to get along with other charge controllers and the sell-back engine in an inverter. Those normally do use a specific 'set' voltage to start/stop charging or selling to the grid.

For example, say one has some photovoltaics in addition to wind, with an MX-60. What would one set as the voltage for the MX-60 to stop charging? Below 54.8 and there won't be any solar input if there's wind, above 57.6 and wind won't be used if the solar panels are producing. Between the two values and the solar charger will be pulsing rather than continuously producing. Is there a way to make this type of controller to 'play nice' with the rest of a system?

-Rob-

[Matt Tritt]

[QUOTE=Rob Beckers;660]Well, this saga is still continuing. One option that's being examined closer is to use an NCHC-100 from FlexCharge (http://www.flexcharge.com/flexcharge_usa/products/nchc/nchc.htm). The heavy lifting in this controller is being done by a relay, making it less likely that electronics get damaged in case no diversion load is used (with an alternator as the source). Relays can also be put parallel for increased capacity. Basically it's just a switch that's set to close at 54.8 Volt, and open at 57.6 Volt, as measured from the batteries. Here is a picture of it:

http://www.flexcharge.com/flexcharge_usa/products/nchc/nchc_24_photo.jpg

Since there is no 'set' voltage that the controller maintains this begs the load controller
>>>>

OR, you can wait a little while and choose the new SunnyIsland 5200, which can also be stacked to create a large system. I've been working recently with an off-grid engineer at SMA (Germany) to adapt their diversion load controller for 60 Hz applications. In this case, it's for an upcoming Iskra installation going into the Northeast. This will create, in my opinion, the ideal off-grid hybrid system. The Iskra requires no voltage limiter on the output because of the variable-pitch speed control, and a dynamic brake system.

Oh boy!

Matt Tritt

[Rob Beckers]

So that would be a SunnyIsland plus a WindyBoy, right?
Matt, I'm not disagreeing that it would be a beautiful solution. My problem is with those that just want a charge controller. Because they already have an inverter, or because they want to go with some other brand/type inverter. The price tag involved with a full inverter plus MPPT charge controller can be a bit much for those cases.

I will keep the SunnyIsland/WindyBoy in mind, to recommend it for those cases that warrant the solution. Still, I need a simple, working, and hopefully cheap charge controller. That's where my FlexCharge questions came from...

-Rob-

[Rob Beckers]

Alright, some more "talking to myself" here :rolleyes: . Just in case someone else is interested.

First, backfeeding Outback inverters is entirely undocumented. However, when prodding Outback about it their answer is "yes, it is possible". Digging a little deeper shows it to be possible indeed (and Logan has this working reliably after all), but it has a number of potential issues related with it. See the replies on Outback's forum (http://www.outbackpower.com/forum/viewtopic.php?t=1900&sid=4b0a9d50ef2d3afa73b54fb14616fef1). At this point I believe this is solution of last resort, not something I'd recommend in general.

Asking around about the TriStar as a charge controller reveals that indeed this won't work well together with other charge controllers or devices that work with a 'set' voltage. It also seems to be hard on batteries to do 'hard' on/off switching of the full charge current and accompanying voltage swings.

So, it looks like it's back to using two TriStars plus transformer/rectifier. The new MidniteSolar MPPT controller would be nice, then again it will take a transformer to make that one work, and no one knows when it will be available. It's surprising how few wind charge controllers there are out there...

-Rob-

[Paul Bailey]

Are you looking for dump load or charge charge control of the excess wind power from your mill?? I guess I should read your whole post to see whats what!!! Why not dump the power to heat ..especially water.

[Rob Beckers]

Are you looking for dump load or charge charge control of the excess wind power from your mill?? I guess I should read your whole post to see whats what!!! Why not dump the power to heat ..especially water.

Hi Paul,

I'm primarily looking for something that will do actual charge controlling from wind to battery. This is for the Eoltec Scirocco wind turbine (http://www.solacity.com/Scirocco.htm), it's a 6kW turbine that produces 240V AC, or after rectifying that makes around 340V DC. When charging a 48V battery bank that works out to around 110 Amp at full power.

This doesn't have to be a diversion controller. I think that part is quite easily solved by putting two TriStars in parallel, directly on the batteries, so they don't deal with the turbine and possible voltage spikes of switching an inductive load. The turbine doesn't need a diversion load, it'll quite happily run completely unloaded and never exceed 245 RPM.

Unfortunately I have not found anything better than using those same TriStars in parallel for charging (with a big 3-phase transformer, rectifier, and some way to keep transients in check), or charge directly from the rectifier and then put the TriStars as diversion controllers on the batteries. Not everyone has a need or the space for diversion loads. An MPPT controller that can handle the input voltage and current directly would be very nice. Know any? :cool:

-Rob-

[Paul Bailey]

I understand that you need design for max capability (6kw) But at what percentage of time are you at max ??? 10 % maybe. Have you done wind study /monitoring for you area. ??? What is you avg. wind spd.? How big is this Battery bank that your charging(Ahr)?? what is you daily usage in Kwh or Ahr VS what you may need to Dump off as excess???? Exelltech used to be big on 240 V inverters and Control Balance of system stuff. Oh and how are you controlling all this now.?? Seen enough of the bigger guys just heating the house with the Mill since the powers coming in anyway!! Paul

[Rob Beckers]

I understand that you need design for max capability (6kw) But at what percentage of time are you at max ??? 10 % maybe. Have you done wind study /monitoring for you area. ??? What is you avg. wind spd.? How big is this Battery bank that your charging(Ahr)?? what is you daily usage in Kwh or Ahr VS what you may need to Dump off as excess???? Exelltech used to be big on 240 V inverters and Control Balance of system stuff. Oh and how are you controlling all this now.?? Seen enough of the bigger guys just heating the house with the Mill since the powers coming in anyway!! Paul

You are of course correct that maximum power contributes very little to total produced energy over time. Taking a Weibull wind distribution and converting this to energy would show that. I would be quite happy with anything that can charge at less than full power, within reason.

This is not so much for me (though I actually do want to put one in my backyard), but more in general. I import/sell these turbines in Canada. We used to have Magnetek's MWI-5200 inverter/charger to do MPPT charging of a 48 Volt battery bank (and it has an inverter as well, though one does not have to use that part). Magnetek was bought by Power-One, and they decided out-of-the-blue to discontinue the MWI-5200. There is a replacement, sort of, the MWI-5000. Identical except for that 200 Watt less inverter power. Unfortunately it is 50% more expensive at wholesale pricing than that MWI-5200, making it just too expensive as a charge controller. With around 50% of potential customers in Canada (and probably the US as well) looking for off-grid use I need a charge controller.

Hope this explains it.

-Rob-

[Mark Parsons]

Rob,

Why can't you just rectify the wild AC and connect direct to the 48V battery bank? Why are you trying to transform the wild AC down?

Cut in RPM will be low since it is a 240VAC @ 245 RPM alternator which is about 1.4Vp per RPM. For a 48 V battery bank the cut-in RPM will be about 35RPM. The excess RPM creates the amperage to charge the batteries. You need a diversion controller and dump load that exceeds the power rating of the Scirrocco to keep the batteries from getting over-charged.

Mark

[Paul Bailey]

Look in Europe as everything is 220-240v ? Paul

[Mark Parsons]

Rob,

I'm no expert in this area, but with what I've read MPPT is used to gain battery charging amps from solar panels that produce a fixed amp output at a voltage higher than the battery terminal.

How is this an advantage for a Wind Turbine wild AC Alternator? Each RPM above cut-in produces extra amps for the battery. I would think a MPPT in the middle would just lose efficiency due to being only 95% efficent.

Please educate me.

Thanks.
Mark

[Rob Beckers]

Rob,

I'm no expert in this area, but with what I've read MPPT is used to gain battery charging amps from solar panels that produce a fixed amp output at a voltage higher than the battery terminal.

How is this an advantage for a Wind Turbine wild AC Alternator? Each RPM above cut-in produces extra amps for the battery. I would think a MPPT in the middle would just lose efficiency due to being only 95% efficent.

Please educate me.

Thanks.
Mark

Hi Mark,

Actually MPPT is quite the cat's meow when it comes to wind or hydro turbines! It is a DC-DC converter, or some start with AC, and converter that to DC, they can use frequency for MPPT lookup as well, see below. The input voltage can be just about anything, while the output voltage is the battery voltage. When the input voltage is low (because there's not much wind for example), it'll boost it to battery voltage. When the wind is howling it'll buck down to battery voltage.

An MPPT controller has a lookup table in it, with the optimal load for either voltage or frequency for the particular turbine. So for each turbine RPM it will get the maximum possible energy out of it, and all that at over 90% efficiency.

Compare that to charging batteries directly (from a transformer). At low wind the turbine would spin at low RPM if you could load it optimally. However, until the transformer output reaches 48 - 57 Volt (depending on state of charge) there is no load, so the turbine will simply speed up, even at low wind. As soon as it reaches the battery voltage it'll get loaded, but it's now spinning at entirely the wrong RPM for that particular wind, and the load on it is haphazard, whatever the turbine will allow so the voltage stays at battery voltage. There's a little bit you can do in matching by changing the transformer ratio, but it'll only optimally match a very small part of the turbine's power curve, and only at a specific battery voltage, at best. Keep in mind that turbine RPM and wind speed translates to angle-of-attack of the blades because of the effective angle of the wind vs. the blades (for fixed-pitch turbines, and even for almost all variable-pitch machines since they only change pitch at full power to limit max. RPM). The wrong RPM means the wrong angle-of-attack, means higher drag if nothing else, and therefore lower efficiency.

From what I've heard, a 'regular' charge controller (non-MPPT) for wind/hydro will at best get 50% to 75% efficiency. Those numbers are at full power though!! I don't have numbers for it, but have a feeling that efficiency at low wind will be piss-poor, (much) less than 50%. Unfortunately, it is the low wind regime that contributes the majority of energy for most people. Places that are good for wind turbines are not the places you'd want to live, so most of us have to settle for low average wind speeds.

There's also an issue with using a transformer in combination with wind/hydro alternators. They are variable RPM, and transformer efficiency drops off as frequency goes down (as does power handling, though that's not a big deal since power output of the alternator also goes down fast with RPM). At zero RPM the transformer presents essentially a short-circuit to the turbine, or rather, just the DC resistance of its primary windings, which should be low if you want the transformer to be efficient. This makes start-up of the turbine hard.

I hope that explains a bit why a good MPPT charge controller is such a big deal. Now, having said that, I'd settle for a conventional solution with transformer and old fashioned charge controller, until something better comes along.

By the way, I would love to see a real-life comparison of MPPT vs. non-MPPT charge controllers. I have not found one on the Web. As I said, I don't have comparison numbers, just a feeling that for average/low wind sites the difference is large. Something that shows if I'm right or not would be very interesting. I would also welcome anyone that has a better theoretical explanation of conventional charge controller and their effect on wind turbines. Interesting stuff!

-Rob-

[Rob Beckers]

Rob,

Why can't you just rectify the wild AC and connect direct to the 48V battery bank? Why are you trying to transform the wild AC down?

Cut in RPM will be low since it is a 240VAC @ 245 RPM alternator which is about 1.4Vp per RPM. For a 48 V battery bank the cut-in RPM will be about 35RPM. The excess RPM creates the amperage to charge the batteries. You need a diversion controller and dump load that exceeds the power rating of the Scirrocco to keep the batteries from getting over-charged.

Mark

I've not tried this (too expensive to just experiment I fear), but suspect it'll burn up the alternator if hooked up directly to the batteries. Batteries present a very low impedance.

What I've been trying to do is find alternatives to the missing charge controller so I can present them to Eoltec, the manufacturer. Hopefully I can convince them to try a few.

-Rob-

[Paul Bailey]

What is your battery Ahr rating?? And as Mark stated output will be pulled to the Battery voltage anyway. If theres too much available current dump some (somewhere...HEAT) as the Battery is still taking the charge.. Just an Idea.Paul

[Rob Beckers]

What is your battery Ahr rating?? And as Mark stated output will be pulled to the Battery voltage anyway. If theres too much available current dump some (somewhere...HEAT) as the Battery is still taking the charge.. Just an Idea.Paul

This discussion about finding a charge controller is really meant to be independent of the size of the battery bank. Either the charge controller needs to do exactly what its name implies, control the charging of those batteries so they don't get overcharged, or there needs to be a diversion controller that takes care of excess energy. Either method would work.

-Rob-

[Mark Parsons]

Rob,

Wild AC by definition indicates a PM alternator with phase coils brought out direct to terminals. By rectifying this output no current will flow until the generated voltage created by RPM's exceeds the battery terminal voltage. The alternator will only produce the amount of current into the batteries as determined by the power harvested from the wind divided by the voltage. By feeding this power into a voltage sink (battery) the current is increased to satisfy Ohm's law. Is this higher current your fear? 6kW / 50VDC = 120Amps.

As with every alternator (motor) - it will get hot under load conditions. I^2T protection is designed to thermally protect every generator / motor under continuous load and overload conditions. The advantage of the variable pitch system is no runaway under no load condition if circuit protection trips. If concerned about excess I^2R losses creating excessive heat in the alternator stator then reduce the I^2T current setting - i.e. use a smaller circuit breaker or motor overload element. I assume Class 10 protection is required. Class 10 defines overload trip upon 6 times rated current for 10 seconds.

The Eoltec brochure indicates battery charging with a PWM charge controller. A PWM controller on 48V batteries could load the alternator with 120amps under full power conditions and depleted batteries.

Am I misunderstanding your concern? :confused:

Mark

[Rob Beckers]

Hi Mark,

The Scirocco alternator is meant to deliver 6.5 kW at 240 Volt AC, or 340 Volt DC after rectifying it. It's maximum current rating is around 16.6 Amp per phase (it's 3-phase). Translated to DC that is 19.1 Amp. Hooking that up directly to a 48 Volt battery bank would surely drive the current far past what's allowed, burning up the alternator.

The "power harvested from the wind" is also somewhat of a moving target: The blades and efficiency of a turbine may well exceed the capacity of the alternator that's hooked up to it. I don't know if that's the case for the Scirocco (though I have a feeling it is to some extend, since it's optimized for low-wind performance), but what I'm trying to say is that just loading up the alternator 'as much as it'll deliver' may get you into trouble in a hurry.

So, it would seem to me that to hook up a high-voltage alternator (ie. meant to deliver its power at a voltage well above that of the battery bank, so its current is meant to be much smaller as it would be for the same power at battery bank voltage) directly to a lower voltage battery bank is a quick way to do several thousand dollars damage to a very expensive permanent magnet alternator (I can tell you from the lingering pain in my back that a 6 kW alternator of this type is massive! ;) ).

-Rob-

[Ethan Brush]

Rob,

I am in a similar situation as you in that I am trying to figure out what to put between my turbine and batteries and am frustrated by the lack of wind charge control products. I have a whisper 200 HV (240 vac out). The ‘normal’ thing to do would be to buy the $1250 transformer from southwest and their $700 controller, but this is big money and of course does not give me mppt. If I do stick with the transformer route, I can build my own three phase bank for probably half the cost southwest wants (even considering the derating required due to the low frequency), and their controller seems like a waste because all it is is a rectifier and a dump load – again for less than half the cost, I build my own rectifier bridge and get a trace c60 for diversion control. I have been looking for dc-dc converters thinking it would be nice to avoid all the transformer problems – rectified three phase 240 gives about 325vdc and V infinity makes such a dc-dc converter, but it is pricey and efficiency is only in the mid 80’s. Of course the dc-dc converter route does not get me mppt either. Then I thought about using an MX 60, not for its solar mppt, but for its dc-dc converting capabilities. However my voltage would be too high, and there may be other issues anyway. Using a windy boy sounds like an intriguing idea, then I could just use my iota to charge the batteries and dump the extra. Needless to say I am eagerly awaiting Midnight Solar’s controller. In case you haven’t seen this, here is December’s announcement from midnight with a preview of the classic

http://www.fieldlines.com/story/2006/12/13/154529/53

[Rob Beckers]

Hi Ethan,

I've read the same press release about the MidNite solar controller. Unfortunately it won't work all that well for you or me, the maximum input voltage (DC) is 150V initially, with a 200V version coming out later. So, a transformer would still be needed. That's not only expensive, but also brings with it all the issues of driving a transformer with a variable frequency source.

Like you, I've also looked at the various DC-DC converters available. I don't think they are going to help. They are not charge controllers. Besides, once you get up to 6kW they are expensive too.

Lately I've been spending serious thought on having my own MPPT charge controller developed. I'm an electrical engineer, and have built many electronics projects, though my power electronics knowledge is limited. A buddy of mine, who's also an electrical engineer, does know power electronics well. So the past two weeks we've been going back and forth on what would be needed. While it is not your average toaster, these things are not exactly rocket science either. If I go this route I will probably try to find someone locally with hands-on power electronics experience to make a design and develop a prototype (anyone on this forum that wants to apply? :cool: ).

The other side of a project like this is the business case. There's a reason why there's nothing available, the market for it is just plain small. I would not build just a single one, the idea would be to make at least some money from having these things produced in small batches. I have several thoughts on this, ultimately it comes down to how much the unit costs to produce and support.

So, there's nothing firm yet in terms of go or no-go, but it is being considered seriously. To that end I've raided the Carleton University library yesterday and took home half a dozen power electronics books, so I can get myself up to speed on circuitry and components. We'll see where it goes, if nothing else it's a fun project to work on.

-Rob-

[Mark Parsons]

Rob and Ethan,

My $0.02 worth.;)

My background in Industrial Electronics has me thinking about your quandry. There are relatively inexpensive devices on the market that will take a 230VAC input and convert to a variable AC voltage and frequency. They don't do MPPT. They don't do voltage / current ratio transformation like a transformer or a DC - DC converter with a torroid. Maybe with some tinkering as they are cheap enough these days they can be used as front end driver to a cheap autotransformer (7.5kVA $500) between your alternator and a MPPT controller. I can buy a 3HP 230VAC 3 phase brand name unit for about $800. Off brand units are less. 3 phase units will also accept single phase inputs.

What are these devices, you ask? AC Drive controllers. Normally used to control the speed of industrial AC motors. They are just as happy feeding a transformer load.

Educational info - http://www.kilowattclassroom.com/Archive/VFDarticle.pdf
Product info - http://www.ab.com/drives/powerflex/4/index.html

Imagine the wind turbine alternator is feeding the input 3 phase bridge rectifier. The DC buss is chopped by PWM carrier into AC waveform. Set the frequency at 60 HZ and feed your cheap autotransformer to get the voltage / current ratio you need to protect your wind alternator current limit. Rectify the output and feed to your lower input voltage MPPT controller.

OR design your own MPPT unit with 400VDC input capability and create a company to market and sell the result. OR wait until one of the existing MPPT controller companies make a unit for your turbine.

Good luck in your quest.

Mark

[Ethan Brush]

Rob,

I noticed the voltage limitation of the midnite controller - quite strange they chose to limit the number of turbines it will work with. I guess they are assuming 12- 48 v nominal are the majority. Can you give me a brief rundown on how a windy boy might work for my situation? I'm a little vauge on it. Will it work without a grid "signal"? Is the output 120vac 60 hz or does it have a battery charging output? If it is the former, would just plugging an iota into it nfor battery charging work? What would happens if the iota tries to draw more than the windy boy is getting from the turbine? What is the purpose of the other inverter (sunny island) that would be used with the windy boy in this propsed setup?

Mark,

I read some literature and I'm a bit unclear on how the drive would be fed. Would one have to take the rectified turbine power and "tap in" to the section of the drive after where the ac is rectified but before it is PWM'd into AC? Or could an off the shelf drive just be fed variable frequency since (if?) it is rectified immediatly anyway?

Regards,
Ethan

[Rob Beckers]

Rob,
I noticed the voltage limitation of the midnite controller - quite strange they chose to limit the number of turbines it will work with. I guess they are assuming 12- 48 v nominal are the majority. Can you give me a brief rundown on how a windy boy might work for my situation? I'm a little vauge on it. Will it work without a grid "signal"? Is the output 120vac 60 hz or does it have a battery charging output? If it is the former, would just plugging an iota into it nfor battery charging work? What would happens if the iota tries to draw more than the windy boy is getting from the turbine? What is the purpose of the other inverter (sunny island) that would be used with the windy boy in this propsed setup?

The reason for the 150V or 200V limit is that beyond that, given the currents involved, one has to change the design from using cheap MOSFets to very expensive IGBTs to do the switching. The difference is about 90 dollars per MOSFet/IGBT, so depending on the number used in there (which depends on their design) this can add tremendously to the price. To make matters worse, MOSFets can handle higher switching frequencies, so when one switches to IGBTs the frequency has to come down and that makes for a larger and more expensive ferrite inductor in the design.

You are correct that the Windy Boy must have a grid 'signal' to work. It's a traditional inverter, that has anti-islanding build into it, as required by UL1741. So when the grid that it 'sees' goes out, the Windy Boy stops producing as well.

The Sunny Island comes in because it has a mode that allows for islanding (as opposed to the anti-islanding in the Windy Boy). What that means is that it will produce 120/240V even without a grid, to create an 'island' of power. It can also charge batteries from its grid output. So, the Sunny Island creates your local grid, the Windy Boy is happy by seeing a grid and converts wind power to grid power, the Windy Boy power is passed back through the Sunny Island to the batteries.

I've talked to SMA about the Windy Boy design. The downfall is that it is a solar charge controller under the hood. While they claim MPPT for wind, it is limited to a lookup table of just two points. There's a cut-in point, and a maximum power point (each point has a voltage and power associated with it). Everything else is interpolated on a straight line between those two points. Especially for low winds this creates a big mismatch between the wind turbine and the load on it, causing it to overload the turbine to some extend. So, the whole solution seems better than a straight transformer/rectifier type charge controller, but it's expensive and far from optimal.

-Rob-

[Mark Parsons]

Greetings Ethan,

These devices are normally connected to 208 - 240 VAC (or higher) 50/60HZ. There are also versions with no front end rectifiers that are made to utilize a 'common DC bus' for regenerative purposes between multiple drives.

I have not connected a AC input version to a variable frequency source. As far as I know they do not monitor incoming frequency. They monitor the internal DC bus voltage closely. A low or high bus voltage alarm will turn the output section off until manually or automatically reset. A minimum voltage sensor will be required from wind alternator before feeding to the drive to trap this fault out. The drive should accept the variable frequency without trouble since all it does is rectify immediately.

Attached link shows the specifications for AB Powerflex 4 product family.
http://www.ab.com/support/abdrives/powerflex4/datasheet.html

See the line for the 230V 3 phase 5hp(3.7kW) model 22A-B017N104. Input voltage range is 180 - 265 VAC. The section below shows allowable DC bus range of 210 - 405 VDC. You can see it is robust enough to handle 60 seconds of 150%, 3 seconds at 200% and trips at 300% of its power rating. There are about 140 programmable parameters in the drive. This 'Parameter Listing' can be downloaded as an Excel file from http://www.ab.com/support/abdrives/powerflex4/index.html
It can be configured to roll back output voltage and frequency when bus undervoltage approaches due to loading. This will help to maintain a connect between input power available and battery charging current (output power).

In theory, this should work to change the variable frequency and voltage of your alternator between the spec ranges of the drive to a fixed frequency and voltage to feed a transformer with very good efficiency. The drive is rated 97.5% efficient. An autotransformer operating at 60Hz is about 96% efficient.

I state again, I have not done this. However, we integrate these products everyday around here in some of our custom engineered automated equipment solutions. If I had access to a wind turbine alternator in question I could perform some shop tests to verify this solution. Rob - have a Scirroco alternator you want me to test? There, the offer is out.:eek: I am near you when at my cottage.

Regards,
Mark

[Ethan Brush]

Hmm, this is the second time Ive posted something and it has disappeared. Well here is about what I said...


Thanks for the great info guys. I agree the windy boy and sunny island is far too expensive of a solution to be practical (for me anyway). I might look into the VFD idea a bit more. It seems to me the main thing is to figure how much you gain in transformer efficiency by feeding it 60 hz. The 60 Hz would probably knock off around $300 extra in transformer costs, so the $800(?) for the VFD minus that is about $500 for the increased efficency. Of course a further analysis of how the VFD would function, i.e. under/over voltage, cut in, etc, would be in order. There doesnt seem to be an elegant solution for us folks who have a killer wind spot 1300 feet away.


Ethan

[Ethan Brush]

Rob (or anyone else of course),

I have been reinvestigating the dc-dc converter route and was wondering how extensively you researched this angle, what your thoughts are, and how you think this method compares with a tranformer. I was recently informed of another brand, vicor, that makes some units that look like they could work. The voltage output can be adjusted too , which might allow different charge rates and a degree of mppt. The particular unit I was looking at can handle 425vdc in and is 600 watts, so two would be needed in parallel for my application. I have yet to call the company for prices. The major potential problem I see with the dc-dc converter route is that a fixed voltage output would mean that the battery would want the same amount of current regardless of what the turbine is doing and Im not sure what the dc-dc converter would think of that, i.e. not enough current available on the input side at low speeds. However perhpas this vicor unit could be set up with a little controller to vary the voltage output based on what the turbine has available.

Regards,
Ethan

[Rob Beckers]

Hi Ethan,

You raise a couple of different issues: First about using a DC-DC converter. While an MPPT controller uses a DC-DC inverter inside, a DC-DC converter by itself does not make a charge controller. They are voltage controllers, not current, which is what you need for charging. Now, if they were designed by someone with half a brain, there's going to be some form of current-limiting in there before self-destructing. Depending on the implementation it can back off the current completely when it senses a short-circuit, making it useless for charging. By the way, it only takes a very small discrepancy in the converter's output voltage and the battery voltage for it to see in essence a short-circuit. Batteries (and their associated thick cables) are very, very low impedance. The other part of the equation is that if we assume there's a current limit, the DC-DC converter is going to use that same limit when charging a set of half-empty batteries while there's little wind. Essentially putting the brakes on your turbine. So, a DC-DC converter by itself is not going to be all that useful for charging batteries, IMO. On top of that, a DC-DC converter that can handle the 6kW I'm looking for is not cheap.

The other issue you raise is about comparing an MPPT controller with a simple transformer plus rectifier. The more I look into this the more I get the idea that the transformer solution is pretty bad. Most of us live in areas with an average wind speed that's not of the 'gale' type, so any wind turbine will be producing the vast majority of its energy in the low to mid strength winds. It's at those winds that the transformer is a pretty bad match, the alternator frequency tends to be very low (=high transformer losses), the turbine has to spin faster than optimal (=poor airfoil efficiency). Then there are issues with protecting the alternator, meaning that if the blades are relatively large (so the alternator is a bit undersized) it has a good chance of burning up the alternator on a windy day with relatively empty batteries. My feeling is that one looses much more than the 25% claimed by for example the Bergey Excel-R (7.5kW) vs. Excel-S (10kW), because those numbers are based on maximum power output and not on the energy delivered. Remember most of that energy is going to come from the lower wind speeds, not rated wind speed, where that transformer is a much worse match.

Now, this is really just my 2 cents. Maybe my gloom-and-doom vision is completely off, and the transformer solution works just fine. After all, people have been using this for decades. Then again, maybe I'm right. I don't have a wind turbine here (yet), so I can't test a transformer vs. MPPT controller. In fact, this is really something that only the manufacturer can do with some accuracy, and none of them are forthcoming.

As I mentioned before I've been studying up on power electronics the past few weeks. I think I have a pretty good understanding now of how things work, and how they should be designed. In fact I have a design in my head that could do the job. Right now I'm (re)learning SPICE, the electronics simulation program, last time I used it is two decades ago, so I can realize the design on paper and simulate it. This doesn't mean I'll actually produce anything, that is still a long time and many decisions off.

-Rob-

[Ethan Brush]

I am also of the school that a tranformer is an archaic crude solution, which is why I am grasping for other alternatives. What gets me is southwest sells a step down transformer to get my 240 down to 12 nominal, but their attitude is that you just plug and play - no mention about inefficiencies, sizing it differently for your average wind speed, and most importantly, the potential problem of the transfomer not allowing the turbine to start due to the (essentially) dc voltages at start up.

Those are definitely valid concerns for the dc-dc converter. Essentially what it boils down IMO is that the output current needs to be coordinated with the available input current. Since the vicor unit has adjustable output voltage, this could (presumably) be used to accomplish this if properly interfaced. Of course for you and your 6kw turbine, this is unlikey to be a cost effective solution even if the other issues were resolved.

Maybe I just need to not worry so much about efficiency, My turbine is quite oversized for my loads and battery bank, especially since I should have 14 mph average winds. A more elegant solution would sure be nice however.

Ethan.

P.S. let me know when you start production of your controller ;)

[Dan Lenox]

Rob,

I recently got the Xantrex Xw-6048 inverter, bought it from Aten Solar http://www.atensolar.com shipped to me for just over $3300.

It is 48v compatible with max amperage of 60amp, so stacking two would easily solve your 100amp situation.

Just got inverter and equipment mounted and plumbed, but will not be till spring till I get my 17' dia. blades and tail components up onto turbine.

Dan Lenox

[Rob Beckers]

Thanks Dan!
Correct me if I'm wrong, but isn't the XW-6048 basically an inverter only? Meaning, that to charge from solar or wind you need a separate charge controller for each source. Looking at their wiring diagram they recommend a diversion controller in combination with direct charging from wind or hydro. That means no MPPT for wind (or hydro) and the XW wouldn't be doing the charge controlling either.

What I was, and still am, looking for is an MPPT charge controller for wind. The MWI-5000-48 from Power-One will actually do this, and do it well. The downside is that it's ridiculously expensive for charge controlling duty.

-RoB-

[Dan Lenox]

Rob,

the Xantrex 6048 is a 6kw grid tie inverter, they refer to it as a hybrid. it does contain a charge controller, but it is not mppt. for wind you don't need anything else but diversion dump, but solar requires their MPPT charge controller - I believe that it sells for another $500.

I think that you originally pointed me towards the PowerOne units, but for my grid-tie w/battery backup application they would not work.

Dan

[John Szegda]

Hi RoB,

I was reading this old thread you posted few years ago and was wondering, what your solution was for a Wind Controller.

John

[Rob Beckers]

Hi John,

To the best of my knowledge there is no solution available at this time.
What I was after is an MPPT type charge controller for turbines that can handle a decent voltage (600V DC for example) and power (6kW). The only wind MPPT charge controller I know about is MidNite Solar's "Classic" which falls far short of this in power and voltage.

-RoB-

[John Szegda]

RoB,

The Classic is a bit expensive for My needs. I have a 1KW Turbine that feed a small battery bank coupled to grid tied Outback 3648 Inverter The factory supplied Controller works best for battery charging. The excess power is dumped and little is feed to the grid.

John

[Rob Beckers]

RoB,

The Classic is a bit expensive for My needs. I have a 1KW Turbine that feed a small battery bank coupled to grid tied Outback 3648 Inverter The factory supplied Controller works best for battery charging. The excess power is dumped and little is feed to the grid.

John

There's a fundamental difference between your turbine, doing direct battery charging, and high-voltage turbines that use MPPT (Maximum Power Point Tracking). In your case, the turbine in essence is directly coupled to the batteries, and the battery voltage controls the turbine RPM. There's some electronics in a controller to prevent overcharging of the batteries by switching on a dump load, but that's it. It's very simple, and very inefficient.

An MPPT controller can handle wind turbines that run at a higher voltage, such as those meant for grid-tie. It takes care of converting the high voltage down to battery voltage, and it loads the turbine in a way that extracts the most power from it. The difference is at least an additional 25% energy from the same turbine, and the ability to use high-voltage turbines with batteries.

It's the latter that I was after, and that to the best of my knowledge still doesn't exist for high-voltage wind turbines. There are MPPT grid-inverters for wind turbines (such as Power-One's Aurora series) but no charge controllers.

-RoB-

[Robert Mann]

Hello I do have been working on this line a for long time.
I am the inventer of the Tree Top Wind Mill a VAWT I first used a PMA car kind with a Belt drive it worked well and sold the unit on Ebay.
Put I found I needed more power.
I learned of your F&P Smart Drive and statered to up grade it.
Here is it.
http://www.vawts.net/download.spark?ID=980354&aBID=125317
on the test brigeport.
http://www.vawts.net/download.spark?ID=980353&aBID=125317
Here is a test stator

In delta a make one volt for every RPM I can run it to 2000 RPM PLus very safely.

So I hads some help from Bob from Down Under helayed out this for me

http://www.vawts.net/download.spark?ID=980355&aBID=125317

Yes I have over come some of the supper High Voltage my unit makes 1000 Volts +

I worked with Bob from Down Under to come up with this set up for off Grid charging

12/24/48/ Volts it work like a dream. If Steve can fix the Grid Ties from smoking up we will be good to go with Grid Ties.

If I can be of help let me know.

Bob Mann

Tree Top Wind Mill
www.TreeTopWindMill.com

[Stewart Corman]

Bob,
if you were posting photos, they aren't showing ??

just read your last reply and looked at your article:
http://southshorexpress.com/pembroke/5733-high-st-inventor-hopes-to-market-treetop-wind-turbine.html


Why don't you open a new thread on your VAWT design, with some photos, design specs, and data that you took when driving it around the back of your pickup.

We have kicked VAWT here for many years, and NOBODY has ever shown data that confirms it as a viable wind turbine design.

Also a problem is the relatively low rpm for a direct drive generator ...need lots of poles at large diameter.

Interesting to mount it 80 feet up on the top of a tree.

thx,
Stew Corman