Author Topic: 3.2 meter ferrite magnet MPPT wind turbine  (Read 61324 times)

0 Members and 1 Guest are viewing this topic.

Offline ChrisOlson

  • Moderator
  • Hero Member
  • *****
  • Posts: 510
  • Karma: +29/-5
  • just trying to survive
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #30 on: January 19, 2012, 10:01:24 am »
Is it practical to just "jumper" part of the coil? Might be quicker, easier and more versatile if you can bring in as much (or as little) resistance as you need? At least while you're fine-tuning it.

I certainly could do that.  The resistor I built is made from some 6,500 watt 240 volt heating elements that I found, that came out of an electric furnace.  Each element had a spring about 10 feet long and it was arranged in a weave pattern, strung thru some ceramic insulators.  I measured along the springs with my ohm meter until I found a point where the resistance was 3 ohm, then cut the springs off to that length.

When I built the resistor I wired it wye so my resistance is 6 ohm.  At 100 volts that only draws 17 amps for a 1.7 kW load.  I actually need it closer to 3 kW load, so I have to cut the springs just about in half again.

I got one leg of the turbine generator going directly to one terminal on the resistor.  The other two terminals are powered up by a two pole SSR driven with PWM.  I think it's actually OK to have too much load because the PWM will regulate the voltage by varying the pulse width to the SSR.  So I'm pretty sure if I cut the springs in half I will have 3 kW @ 100 volts (or maybe even slightly more).  This should handle the full output of the turbine OK with it disconnected from the DC load.

In theory, it should never have to use the full capacity of the clipper unless the breaker between the Classic and battery should trip.  During "normal" operation when there's a DC load the clipper only comes into play when the Current Limit in the Classic is reached, which I have set at 84 amps now.  The clipper then only has to handle the excess power over the 84 amp DC load to hold the voltage at 100 volts.

So I think I have two options - either rewire the springs delta, or leave them wye and cut them about in half.  Rewiring them delta would lower the resistance to about 2 ohm, while cutting them in half would lower it to about 3 ohm.  I think the 3 ohm load is closer to what I need.
--
Chris

Offline Jimf

  • Newbie
  • *
  • Posts: 4
  • Karma: +1/-0
  • No Personal Text Set by User
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #31 on: January 19, 2012, 02:23:26 pm »
Quote
So I think I have two options - either rewire the springs delta, or leave them wye and cut them about in half.  Rewiring them delta would lower the resistance to about 2 ohm, while cutting them in half would lower it to about 3 ohm.  I think the 3 ohm load is closer to what I need.

Another option would be to make another set of resistors like you have, wire the two sets in parallel, and you will have double the load without stressing the resistor any more than you are now.

Keep up the posts, this is the kind of project I like to see!

Jim

Offline Wolvenar

  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 1474
  • Karma: +40/-0
  • Mr. Murphys pawn
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #32 on: January 20, 2012, 04:10:35 am »
I agree, I would like to see a litlte of how your control systems, and inverters are setup ( with some images)    ;D
Trying to make power from alternative energy any which way I can.
Just to abuse what I make. (and run this site)

Offline Dale S

  • Jr. Member
  • **
  • Posts: 42
  • Karma: +5/-0
  • My smoke got out again
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #33 on: January 20, 2012, 05:54:33 am »
Be careful what you ask for. ;) I'ts cool as hell but pretty elaborate, I can almost guarantee you've never seen anything quite like it.
I aint skeerd of nuthin....WTF was that?

Offline ChrisOlson

  • Moderator
  • Hero Member
  • *****
  • Posts: 510
  • Karma: +29/-5
  • just trying to survive
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #34 on: January 20, 2012, 09:37:37 am »
I'll have to do a walkthru of that sometime.  System and load control is done with two Morningstar RD-1's and a Xantrex ALM.  One RD-1 has temp input on Channel One and turns on generator pre-heat (block and oil pan heaters) when the ambient temp drops below zero F, plus handles water heating loads.  We have two water heaters.  Only the bottom element on the primary heater is power-on all the time with the stat set at 125 degrees, so we always got at least 55 gallons of 125 degree water.  I replaced the standard 4,500 watt element with a 2,500 so it doesn't take over half my inverter capacity to run it.

The top element, and both elements in the secondary (pre) heater are used as "dump" for wind systems.  They're still powered by 240 VAC, but the RD-1 and Xantrex ALM turn them on in stages to control bank voltage when the wind is blowing good.  I do not have any DC "dump loads".

The second RD-1 shuts down the wind systems, one at at time, when both water heaters get up to 170 degrees.  And it puts the turbines back to work one at a time as power is needed again.  On a good day with both water heaters at 170 none of the turbines will be running and the solar will be floating the bank and carrying the loads.  On a bad day when the sun only shines in Argentina and the wind only blows at South Pole Station, the generator will be running.

Like Dale said, you've probably never seen anything like it.  I spent a LOT of time (and money) getting it to work the way I wanted it to, so I can just walk away and do other things and it takes care of itself.  The only real PITA is that those RD-1's are in the utility room and when I want to change some operating parameters of the system I have to lug my laptop out there to program them.  They only have RS232 serial connection.  One of these days I'm going to pick up a StarTeck RS232 - > IP ethernet server and connect them to  my network router.  But I'm saving that for when I really start getting lazy   :)
--
Chris

Offline rossw

  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 879
  • Karma: +35/-0
  • Grumpy-old-Unix-Admin
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #35 on: January 20, 2012, 02:43:27 pm »
The only real PITA is that those RD-1's are in the utility room and when I want to change some operating parameters of the system I have to lug my laptop out there to program them.  They only have RS232 serial connection.  One of these days I'm going to pick up a StarTeck RS232 - > IP ethernet server and connect them to  my network router.  But I'm saving that for when I really start getting lazy   :)

Watch ebay for a used Lantronix ETS8P. I have a couple of them here, and they're great for this purpose.
Each device has 8 RS232 ports, individually addressable.
I have one, for example, that connects my house automation system, weather station, security system, touchscreen, PABX, lightning sensor and a couple of 4-channel RS232 temperature sensors. The other one is out in a different equipment room for other stuff (inverter, charge controller etc). They don't take much to run, are reliable and stable. Perfect for constant logging and occasional control/reprogramming. Leave everything connected, just run your software as required.
The only downside with the ETS8P is the RJ45 modular connectors they use, TX and RX each come out on 2 conductors, so you need to muck about with cables for D25 or D9 - they're not just straight through or one-wire-per-pin. I think I picked up my spare one for $30 including delivery. (Oh, and they are made with the provision for a fan but I've never seen one with the fan installed. I usually put one in - if they get over about 50 deg C, I've had them lock up and need a power-cycle. But then, you probably never see those sorts of temperatures!)

Just went looking for any photos I might have of them - here's one:




Here's one modified to add in the fan. (They have two boards, stacked in the case. This is the power-supply board, obviously)


It's a bit of a mess, but here's my homes central network/phone patch panel. The ETS8P is on the top, with its lid open - this was before the fan modification.

Offline Dale S

  • Jr. Member
  • **
  • Posts: 42
  • Karma: +5/-0
  • My smoke got out again
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #36 on: January 20, 2012, 03:11:52 pm »
Jumpin Jehosephat!
I aint skeerd of nuthin....WTF was that?

Offline ChrisOlson

  • Moderator
  • Hero Member
  • *****
  • Posts: 510
  • Karma: +29/-5
  • just trying to survive
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #37 on: January 20, 2012, 04:05:57 pm »
Watch ebay for a used Lantronix ETS8P. I have a couple of them here, and they're great for this purpose.
Each device has 8 RS232 ports, individually addressable.

That's a good idea.  I never thought about looking for one on eBay.  But I would like to have one, and I could use at least a four port because my solar MPPT controllers got the same setup and use the same software.

Sometimes I might tweak something by just a couple tenths of a volt, or change a minimum-time-on timer on a water heater element or something.  It sure would be nice to not have to take my laptop out to the utility room and find a place for it to set within reach of the serial cable so I can program something really simple in a controller.

I got kind of spoiled by the Classic controller because it has built-in ethernet and I hooked it into my home network.  I can log on that thing any time I want over my wireless and see what the turbine is doing.  It's got a bunch of chrome bezel gauges that pop up on my laptop screen that show input voltage, watts, and bunch of other stuff.  It even shows what the FET temp is in the controller.  It's pretty cool   :)
--
Chris

Offline ChrisOlson

  • Moderator
  • Hero Member
  • *****
  • Posts: 510
  • Karma: +29/-5
  • just trying to survive
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #38 on: January 21, 2012, 03:23:50 pm »
Kind of a summary to this project;

My first clipper got burnt to a crisp a few days back because I didn't understand that the Classic functions as a charge controller on wind power.  When it reaches absorb voltage (or float, depending on the SOC of the bank) it starts unloading the turbine by letting the voltage (and rpm) climb to maintain the absorb (or float) voltage.  My first clipper was not powerful enough to handle the excess power from the turbine when the Classic did this.  It burnt my heating coil springs into black carbon.

It does the same thing when it hits the Current Limit you have set in the Charging Menu.  I have mine set at 94 amps, but the Classic will de-rate itself, depending on temperature.  When it hits the Current Limit it will start unloading the turbine and let the voltage (and rpm) climb.

So I had to re-think my whole concept on this because on any good day with both wind and sun the turbine will be running up against the clipper continuous when the bank gets charged up.  I don't want to "waste" that power.

So, I re-configured my water heating system.  Instead of using the RD-1 to turn on three elements, I am now driving them with an Omron two-pole SSR and the Classic's Aux 2 output providing the PWM control signal to the SSR.  I ran one leg from the turbine directly to the delta-wired elements, and turn on the other two legs with the SSR.

I ran to Menards this morning and bought three 2200 watt 120 volt water heater elements and screwed 'em into the heaters.  I put a 2200 in the primary heater for a bottom element, and in both the top and bottom in the pre-heater.  Then switched the 2,500 from the bottom of the primary heater to the top.  That 2,500 watt is the one that's always on (with the thermostat), powered off the inverters.  The three elements, wired delta, came in at 4.2 ohm so it'll pull over 3.4 kW @ 120 VAC.

If the Juice Cop or Building Cop, or whatever Cop it is that inspects wiring, ever comes to our place to inspect stuff, and sees those two water heaters with jumpers running between them to wire three elements delta three-phase, he'll probably have a conniption and tip over in a coma.  But the frickin' thing works, and it works good.

It got tested before we lost our sun for the day, but now we got a stiff breeze from the SE.  I "cheated" by setting the absorb voltage on the solar MPPT controllers a couple tenths of a volt above the Classic.  This makes the Classic start diverting the turbine three-phase AC to the water heater (with PWM) just early enough so the solar controllers don't "throttle back" the panels at the same time when bulk is done.  So the solar is still working at full power, and as the bank comes up and requires less amps to hold the absorb voltage, the turbine with the Classic on it proportionately gets switched over to just water heating.  At that point its full power is no longer required for battery charging, letting the rest of the system do its job and get the bank thru absorb into float.

That array of three 2200 watt 120 volt elements seems to track the shaft power of the turbine pretty nice with changing wind speed.

So I'm happy with the thing now.  With a matched aux heating load available for it I'll never have to worry about a runaway again (unless an element burns out).

I bought another Classic controller after I figured out that I like the first one really well.  It came in the UPS truck a couple days ago.  I'm going to build another turbine just like this first one because it's beating the bigger 3.8 meter machines hands down in kWh every single day.  I thought about putting the new Classic on one of my 3.8 meter machines, but then decided against it.  The type of blades on those turbines (GOE222) are simply not suited to MPPT, and they will not perform well with it.

This one has been fun.  And a HUGE learning curve.  But now that I got it "tuned" I like it.  I REALLY like it.  Time to get to work on another one.
--
Chris

Offline Volvo farmer

  • Forum Advisors
  • Jr. Member
  • ****
  • Posts: 40
  • Karma: +5/-0
  • milliwatt
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #39 on: January 22, 2012, 08:08:30 am »
I'm curious why you believe different blade profiles are more or less suited to MPPT.  I'm thinking of doing some modifications to my 10' Otherpower axial flux turbine, since I am also running a Classic.  Maybe dumb it down for me a little. I sorta understand how pitch affects torque, and that some blades really start to pull once they are up to speed due to the aerodynamics kicking in.

It would probably take me years to fully understand this so maybe you can just give a recommendation. If you had a turbine like mine, with the big round magnets, and stator wired for 24V, what modifications do you think I could make to make it work better with MPPT? Right now I'm flying it exactly as it was when it was direct tied to batteries, and it worked pretty good that way.  It runs a little faster through the Classic now, but still furls well.

I'd prefer not to make a new stator, though I wouldn't mind fooling with the air gap (I could go a lot smaller) and trying a new set of blades.  I'd also like it to work reasonably well in low wind (as that's mostly what we have here)  and not run so fast as to sound like a helicopter when a storm blows through.

Any ideas?  ;D

 

Offline kenl

  • Newbie
  • *
  • Posts: 5
  • Karma: +0/-0
  • No Personal Text Set by User
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #40 on: January 22, 2012, 08:19:32 am »
Chris,

 Turned out really nice,, you attention to detail shows. I don't think many people would have come up with 3 WH elements wired in delta, now that's thinking outside the box.

I have a question on your blades. How close to perfect balance are they and how did you go about balancing them? Also what size pipe is your tail boom?

 And again great job!

kenny

Offline ChrisOlson

  • Moderator
  • Hero Member
  • *****
  • Posts: 510
  • Karma: +29/-5
  • just trying to survive
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #41 on: January 22, 2012, 09:23:09 am »
I'm curious why you believe different blade profiles are more or less suited to MPPT.

Well, MPPT is not a magic thing.  It does is allow your turbine rotor to run at optimum efficiency, and to operate the turbine at very high voltages to improve power transmission efficiency.  A fixed pitch turbine rotor has a Cp curve.  If the blades are designed to run at 6.5 TSR so their efficiency is 35% (meaning they can capture 35% of the total kinetic energy flowing thru the swept area), both above and below that tip speed ratio, the Cp will drop and you will capture less than 35%.  So if you run the blades too fast you'll get less power.  If you run too slow you'll get less power.

So to get the most "bang for the buck" from MPPT it's important to have blades (and a blade profile) that will track the wind speed.  And this means from cut-in all the way to full rated power.  If you run your rotor excessively fast right at cut-in, as is the normal thing to do with a direct hooked machine, you will make less power in low winds than you will if you run the rotor at the proper (most efficient) tip speed ratio.

So let's say the rotor is running at optimum TSR at 8 m/s wind speed, and suddenly there's a gust to 10 m/s.  The TSR of the rotor drops immediately when that gust hits it and it must accelerate immediately (gain rpm) to get the TSR back to optimum.  Blades that have a lot of "meat" out near the tips and outer blade stations do not accelerate well.  They're generating too much drag.  Those blades might work well for low wind cut-in, and they might generate the torque required to lean into a "stiff" generator that "stalls" the rotor as wind speed picks up.  But they are not good for MPPT because it's impossible to make them track the wind speed.  Their tip drag eats up a good chunk their power at higher rotational speeds.  And if you can't turn the rpm, you can't get the volts required to make MPPT work.

One of the things that blew me right out of the water was the fact that I am getting better performance from this turbine from cut-in all the way to 12 m/s, and it beats my bigger direct hooked machines every single day in total energy production, even in lighter winds.  For a long time I argued with Ryan and boB, and whoever else, believing that was not possible.  Well, turns out I was WRONG, and it IS possible - if you tune the turbine for it.  The key to improved low wind performance is not running the rotor too fast.  Forget everything you thought you knew about how wind turbines are "supposed" to work, and throw it out the door.  Then look at how your blades really like to run.  All the homebrew book recipes turn these machines at very high TSR at cut-in  - dead WRONG with MPPT.  I'm cutting this little 3.2 meter rotor in at only 111 rpm.  The Otherpower 10 usually cuts in around 145-150.  So the Otherpower machine does not have enough copper in it to make the higher voltage required for peak performance on the top end, and it's running WAAAY to fast on the bottom end.

To make the Otherpower 10 work good on 24 volt with MPPT - put a 48 volt stator in it.

Edit:  I figured out how to get this off my computer screen using my Picasa program.  For some reason, when the Classic resets its daily logs it sometimes disconnects from my Local Status Panel on my computer.  I noticed a bit ago that it disconnected and reset its daily kWh counter.  So I had to restart the Local Status Panel to get it come back up.  The turbine is putting out about 500 watts this morning.  But what I wanted you to see, VF, is that at only 500 watts output the input voltage is already at 66 on a 24 volt system.

[ Guests cannot view attachments ]

This is how you make MPPT work for you.  If you run it at only 40 volts or something, you're not going to get the most you can get from your MPPT turbine.  I could pull out the logs and take a look at the CSV, but I'll bet the generator is only putting out like 8 amps or so @ 500 watts.
--
Chris

Offline ChrisOlson

  • Moderator
  • Hero Member
  • *****
  • Posts: 510
  • Karma: +29/-5
  • just trying to survive
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #42 on: January 22, 2012, 09:35:24 am »
I have a question on your blades. How close to perfect balance are they and how did you go about balancing them? Also what size pipe is your tail boom?


They were close out of the box, but not perfect.  I static balanced the rotor and it runs pretty nice.  It has a slight problem with dynamic imbalance at around 150 rpm, but above that it smooths right out.  One day, when it's a hell of a lot warmer outside than it is now, I'll lower the tower and have a look at that dynamic imbalance problem (I adjust that with tip tracking).

I don't use pipe.  The tail boom is made from chromemoly tubing - the same thing they make race car roll cages and racing go-kart frames from.  It's 7/8" OD x .125 wall CRM tubing.
--
Chris

Offline ChrisOlson

  • Moderator
  • Hero Member
  • *****
  • Posts: 510
  • Karma: +29/-5
  • just trying to survive
Classic Data Logging
« Reply #43 on: January 22, 2012, 05:31:34 pm »
This is an example of what the Classic controller can do as far as data logging.  This is an excerpt from the logs over about three minutes worth of run time last night during light and variable wind when the turbine output was changing a lot.  The raw data set is included as an attachment in PDF format and also shows input voltage and current to the controller, logged every 2 seconds.

This graph shows turbine output in watts over the three minute run vs turbine rotor rpm.  It's interesting to see from the graph how fast the rotor is able to accelerate to track the wind speed when it changes and keep TSR at optimum.  During the gust where the power increased to 850 watts the rotor went from 245 to 335 rpm in 2 seconds.  That's why I'm using these blades on this turbine - they can accelerate at that rate, while GOE222's cannot.

Edit:
I don't have my APRS Wind Data Logger tied in with the Classic (yet).  That's on my To Do List.

So I looked back in the wind data from last night during that same 3 minute run and found the wind speed was at 6.3 m/s (about 14 mph) prior to the gust I was looking at to analyze how fast this turbine actually accelerates.  The gust was 8.4 m/s (about 19 mph) and lasted for 3.7 seconds.  The rotor was running at TSR 6.71, and it dropped to TSR 4.94 when the gust hit it.  It took two seconds to get the rotor rpm up to 345 and TSR 6.9 and it was able to run at peak efficiency for 1.7 seconds before the gust died down.

This all may not seem very important, but when you're harvesting energy from the wind you can't just let it slip thru the rotor like that.  Every time you lose out on a gust accelerating the rotor to take advantage of it, you're losing out on kWh production and it all adds up at the end of the day.

Looking at my data, I need to tweak the midrange power curve some more and get the rotor up to TSR 7.1 a little sooner so when gusts hit it, the TSR doesn't drop below 6 before it can accelerate.

That's what data logging, and tuning your turbine for peak performance, is all about.

[ Guests cannot view attachments ]

[ Guests cannot view attachments ]
--
Chris

Offline kenl

  • Newbie
  • *
  • Posts: 5
  • Karma: +0/-0
  • No Personal Text Set by User
Re: 3.2 meter ferrite magnet MPPT wind turbine
« Reply #44 on: January 22, 2012, 06:30:06 pm »
Chris,

 Do you feel that the GEO222 profile is not a good candidate for mppt or are you only saying that a profile that can accelerate quickly with the wind is preferred?

kenny