Computer driven pitch control

[Murlin]

This post will gather the variables for generating a computer program to control a servo system on my wind turbine.

I was thinking about a simple program that will work off the RPM of the turbine shaft and not wind speed.

When the program starts, the servo/encoder refrences itself to zero.

The blades will already be at their highest attainable TSR via a spring preload.

In order to keep the servo from overworking, I must come up with a solution.

So having the blades update themselves on intervals of every 5, 10 or 15 min to an average turbine RPM.

The servo would not turn on until the RPM of the turbine became greater than a predetermined RPM for a given time.

Then the counter would start and keep up with average RPM of the drive axis for a set amount of time.

After the set amount of time had been reached, the servo would make an adjustment and lower or raise the TSR of the blades a predetermined distance based on the RPM of the machine and not the wind speed.

At the console one could input offsets to the blade pitch algorythm on the fly to increase or decrease the the TSR based on what read outs you were seeing on the computer screen.

With this manual offset you could fine tune the system on the fly to take into acount for weather and temperature conditions that might not remain constant throughout the year in terms of output.

This all that comes to mine at this time.

Feel free to add input as it is always welcome.

Murlin teh fun fun....

[vawtman]

Why overcomplicate things

 Just design blades of a certain number and constant pitch that will self furl like the old days?

 The big guys need this but we dont with smaller swept areas.

 Me no understand.

[Flux]

I don't follow this either.

In normal operation I can see no reason to change pitch with a low drag alternator that starts below cut in speed. It may be nice to be able to choose the best angle.

When you reach alternator maximum output you need to do as a normal weight controlled pitch hub does and change pitch instantly, not wait 15 minutes.

with servo you have the option to do as the weight controlled thing does and sit there tracking pitch to hold constant speed or you can hold the new angle and let power fall or even shut it down.

If you get a good gust you had better get things moving virtually instantly or devise some other method of sorting the mess out.

For a servo to work you need good bearings, if you use good bearings you might as well use a brick on a string and throw the unreliable computer away unless you want it to do something that passive (so called) pitch control can't do.

With something over 30ft diameter it may be necessary to use a servo to obtain the control force, especially as blade speed falls and weights get bigger, but for normal pitch control, just keep it simple.

flux

[SamoaPower]

Flux,

Can't say I totally agree on this one. Active pitch control offers a number of advantages over passive (yes, so called) or no pitch control.

"In normal operation I can see no reason to change pitch with a low drag alternator that starts below cut in speed."

It depends on how you define cut-in. If one uses your often recommended scheme of a higher cut-in speed plus a boost converter, starting as soon as some power, albeit low, is available from the wind, is an asset. I suspect the optimum start pitch favors high drag, perhaps 45 degrees or so. Most passive systems I've seen don't offer this option.

Most of the passive systems are designed for high wind protection and offer little improvement in the wind speed region of prime interest, where most of the energy is. By being able to maximize rotor speed for any wind velocity below cut-out, we can optimize energy harvest. Fixed or passive pitch can't do this.

Passive systems typically work against a spring, which unless quite long, tend to be non-linear. Adjustment of such a system has to be quite tedious.

Being able to feather a rotor at will is a definate asset. It's a matter of being able to get total control of your machine.

Do you really think computers are unreliable? I've had one computer failure (a hard drive) in 25 years. My microwave micro-controller still does it thing after 14 years. My car is still running fine with its computer after 17 years. 'nuff said.

I don't think the only reason the large machines use servo controlled pitch is to save on actuator weight.

[Murlin]

Ya you are right, the pitch needs to be adjusted by the wind speed and the program needs to try to achieve peak RPM the fastest way possible.

 "If you get a good gust you had better get things moving virtually instantly or devise some other method of sorting the mess out."

I dont think the blades need to be that sensitive, not react instantly or even every few seconds, there would be too much wear.

Gusty wind below overspeed can be ignored.  No need to chase the gusts as long as the harvested energy average, works out at the end of the day, right?

To deal with gusts that could cause over heat, would require faster pitch response time, that could be incorporated into the computer program. Outstanding observation...

But for slow lazy days where the wind is 5-10, every 15 min would be fine on pitch update.

There could be an exponential equasion in there somewhere that increases the response time the faster the wind blows.

My intention is to wind the stator so I can get the biggest bang for the buck for my copper in terms of KW's produced.

This means the cut in speed is what 14 MPH on a 20 footer?  About 60 RPM I think...not sure about that but lets use it as a basis of comparison.

What governs that?

TSR of blades, voltage of coils right?

Now if you could spin the same alternator, 60 RPM in a 5 MPH wind, would it not make 52 volts in that wind? Sure, if once it got to a certain speed it would fry the copper, but we arent going to let it.  We just want to use those low winds to charge the batteries(not float them).

Wouldnt changing the pitch of the blades do this?

Sure it would.

If the alternator was so powerful that it could not get any torque off the blades to start the thing spinning, then you could just turn the blades a little to give it some, then slowly increase the RPM to achieve cut in speed at the earliest convience at the slowest windspeed possible and reach max output as fast as possible.

The way I see it, it would be kinda like overclocking your CPU...

[Murlin]

Yep, somehow I think you could insulate it from lightning strikes and interference.

[commanda]

There could be an exponential equasion in there somewhere that increases the response time the faster the wind blows.

Murlin,

Some study of PID control algorithms would be in order.

Amanda

[Flux]

I thought you would not agree, but my comments were not directly aimed at you or anyone with good knowledge of control systems.

I have said previously that in the right hands, virtually any scheme can be made to work reliably.

I don't follow the need for coarse pitch in low winds using the boost converter.

With an average size machine you can extract nothing below 6 mph, at 20ft you may get down to 5 mph. With normal running pitch an air gap alternator will start at speeds below this. I think it would be pointless to bring the converter in below the point where the prop is up out of stall. I strongly suspect that changing pitch at 6 mph would help very little.

With a machine with significant iron loss then you may well be right.

If the alternator tracks peak power fairly well then again the same pitch should be ok over the operating range. If you can't track the speed then it would be an advantage to increase pitch to gain a bit of torque in higher winds.

Changing pitch does change tsr, but for a true change of tsr to good advantage the blade chord needs to change as well so the gain will not be that great and if you can't track the speed you don't benefit from the gain in alternator efficiency.

Reliability depends on the constructor. I don't think the average person will make a reliable and effective passive pitch hub. The same crap engineering may operate with servo drive but again will not be reliable with poor engineering.

The computer has proved satisfactory on cars ( with what cost on development) On tractors and industrial things in bad environments it has not been so reliable, again the average person will not get things to run for long without trouble but it is not impossible.

I doubt that the average person who can't stabilise a passive mechanism will stabilise a servo under active conditions. It may be ok for them to attempt adjustments of pitch to suit conditions but it would need back up for speed limiting.

Your machine is quite large and with increase in size the extra complexity becomes less of an issue for someone who understands it and can fix it. If you need to call someone in to fix it then it is not attractive.

I ran a machine without furling for several years, using an anemometer and a control motor on the mechanical shut down. It never failed but I was glad to discover why my furling didn't work and I would never go back to such a thing.

Flux

[Murlin]

"I don't follow the need for coarse pitch in low winds using the boost converter."

You would only need to adjust to a course pitch in low winds to get some torque on the blades to start it spinning.

I am thinking ahead about the many different types of alernators that you might use the variable pitched hub on.

You may want to use it on an induction type motor that cogs real bad. In that case it will definately need some help starting the blades to spinning.

Adjusting to a coarser pitch will do this for you.  Then once it is spinning, you can go finer.

Using a boost is just going to get you even more power in the low winds than you would get if you were not using a converter. I will be using a boost as well.

"ith an average size machine you can extract nothing below 6 mph"

Yepp I can understand that.

But if my stator cuts in at 14 MPH, then from 6 to 14 MPH winds are lost.

In my area the winds blow 5 - 10 half of the year.

So on those days I would be producing ZERO power.

I plan on rectifiying this...

If I wire the stator to cut in around 7 MPH, then I loos 2/3 of my output on the other end.

So I thought and thought about it and decided that I was wrong about agreeing with Danb's strategy for harvesting energy.

If I was going to build a machine with that size,  with so much potential, it would be a shame not to use it.

But again I have never owned a wind turbine or even seen one work up close.

This is all theory in my head, and I do respet the experience you guys have that have actually used them.

But the way the data all computes out in my head, pushes me towards computer control.

I have worked with computers and servo motors for 25 years and make my dailey bread with them.

The key to making it work, is to protect the key components from the rain.  If you can do that, I have no doubt it will work reliably.

Murlin

[SamoaPower]

"I don't follow the need for coarse pitch in low winds using the boost converter."

It depends somewhat on rotor blade design. Some don't want to start very well yet continue to run down to low speeds. I have a small one just like that. It won't come out of stall and accelerate until about 10 mph but continues well down to <5 mph. Lots of hysteresis. Other designs and materials don't allow for very much twist which impedes starting. Twist at the root induces drag, which in addition to lift, aids starting torque. My 16' rotor blades of aluminum monocoque construction are quite limited in twist because of the skin wrinkling limit. Consequently, they benefit from a large starting pitch, about 45 degrees.

On a larger machine, the rotor inertia plus frictional losses make starting an issue if one wants to start collecting power as early as possible. I'd rather have the rotor turning early to allow faster acceleration as the wind picks up to start the boost converter doing its thing. As soon as rotor speed reaches 5-10 rpm, pitch should be reduced to maximize rpm. Easily done with active pitch control.

"I think it would be pointless to bring the converter in below the point where the prop is up out of stall."

That's just the point. The prop can be brought out of stall earlier with active pitch. Sure, there's not much power there, but in a low wind area, we want all we can get even if it's only 10W.

"If the alternator tracks peak power fairly well then again the same pitch should be ok over the operating range."

It's a matching problem again, only this time, we're trying to match the rotor to the wind. A given blade design will perform best at a certain TSR, albeit the curve is broadly peaked. You, yourself, have commented on the observation that your rotors run at a higher TSR in low winds and vice versa. This shows the imperfections in blade design and alternator matching. Active pitch can compensate for these to a degree.

"Changing pitch does change tsr, but for a true change of tsr to good advantage the blade chord needs to change as well so the gain will not be that great and if you can't track the speed you don't benefit from the gain in alternator efficiency."

True enough, particularly for high aspect ratio blades. That's why I tend to favor using lower aspect ratio blades running at a lower TSR like 4.5. For a low speed (rotor) machine, this works out pretty well.

"Reliability depends on the constructor. I don't think the average person will make a reliable and effective passive pitch hub. The same crap engineering may operate with servo drive but again will not be reliable with poor engineering."

Amen.

[scottsAI]

Hello Murlin,

You are building a larger PM generator than is usual here.

Great time to show the difference between induction gen and PM gen.

The example is from a spread sheet for my 10kw design, 24 foot blades.

Assumptions: 36% energy recovery = 0.6 * 75% blade * 80% gen efficiency.

At 12mph (1431w), at 15mph (2795w). If done right real numbers should be bigger.

Let's start with the PM generator, say it's generating 1431w at 12mph with a fixed TSR = 5 spinning at 70RPM. Now the wind goes to 15MPH the RPM goes up to 87.5RPM. The load is fixed voltage, so the current can only go up based on the generators internal impedance with the increase in voltage. So power would only go up by 87.5/70 = 125% or 1788w. The power should have gone to 2795w, other 1,000w missing from the load is in the generator as heat.

The induction generator has a speed up gear, 70RPM would equal 1810 RPM at the generator.

The induction generator as above; 1431W at 70RPM. With wind speed up to 15mph (2795) the Pitch control will change TSR = 4 to stay at 70RPM, the generator will actually need to be running a few as in 10 RPM faster, this allows the rotor field to generate several more watts allowing the generator to output the additional 1000w as useful power, generator output at 15mph will be 2795w, the voltage is controlled by the grid.

Induction motor using Stall control things are nearly the same:

With the blade pitch set for 12mph (1431w), when the wind goes to 15mph (2795w), the power more than doubles:-) yes nice! The efficiency will decrease by using stall control, amount is unknown, lets say power drops 10%. Then it would follow the power would drop to 2516w, or 280w loss.

Is it worth changing the pitch to recover the 280w? Why not!

If the grid is missing the ouput can be excited with CAP.

Your shop must have 3 phase power?

Battery for more than a couple Kw generator becomes costly.

I am sure some battery is desirable for backup.

So why not store power in the grid? Do you have Netmetering in your state?

Then why go through all the power conversions to back feed into the grid?

The electronics to back feed becomes near zero with induction generator.

Pitch control seems to be required, which your doing plus some with electronics.

Using pitch control, other furling systems can be eliminated. Turn blades into wind.

Back to pitch control.

I came across some large wind turbine pitch systems.

Slop = backlash; Covers wider group of precision issues.

Pitch motor has a huge reduction to the blades, 1000:1 Slop did not seem to be a concern.

Goal on the few I could find: rotate 12 deg per sec.

(3600 RPM / 1000 ) * (360 / 60) = 21.6 dec per sec almost needs another divide by 2.

By what ever means, lever, gear we have a target of 20ish deg/sec.

At one time I too was thinking the pitch would be moving frequently.

I came across a report on using a combination of stall and pitch control (can't find it).

Claimed the power generated would drop by a couple percent if the pitch remained fixed while the wind varied. Do not remember how much variance was allowed other than it was more than 1 MPH.

Report more or less stated, the overall impact was not worth chasing.

Another report has the blades moving in real time as it rotates each revolution. This is where the 12 deg/sec came from, they were worried about wind gusting causing damage to system. (have this one)

Claimed the movements are small not causing much ware on the system.

So go figure!

Consider a helicopter, blade pitch changes couple times per revolution! Seem to last a while.

Have fun,

Scott.

[Murlin]

Howdy scott,

Yes I have a 3-phase converter for my CNC.

I am a green horn on all this electrical stuff, my field is mechanical engineering.

Matching the Load was what I was going to try to do to keep the heat out of the stator.

I do not know about all that grid tie stuff, heres what I wanted to do with my turbine:

Charge batteries.

Pump water.

Heat water.

In that order.

Some sort of a cascade dumpload at the very end to suck out the last of the heat losses.

In a good windy day I may wish to switch over to run a welder some how but I haven't got any of that figured out yet.

I have to build the thing first.

I wasnt going to try to store all of the energy, or sell the excess to the power company.

They won't pay me enough for it to compensate the wear and tear on my machine.

But I reserrve the right to change meh mind on that one

20 degrees per sec eh?

A perfectly attainable goal me thinks

If I lived in an area that was a tad more breezy, I would have went with induction.

But half the time I would not be able to make very much power because of a 5-10 MPH wind.  All my research suggests that induction is not good in those wind conditions.

So compromises have to be made....

"The power should have gone to 2795w, other 1,000w missing from the load is in the generator as heat."

One should be able to bleed this heat off using a dump load.

Yes it's wasting power but that is one of the compromises.

But look at it this way, you don't have to worry about turning every little light out

Murlin

[Murlin]

Hey Sammo, one thing I was thinking about was the fact that with a computer controlling the pitch, it could also consider the load and try to match it?

Just a thought

Murlin

[BigBreaker]

The exponential response you refer to is simple to implement and solves the "searching" or "chasing" phenomenon you are trying to avoid.  Exponential sounds hard but it's the result of an easy (the easiest) differential equation.  Namely you make your pitch adjustment proportional to the difference between the current state and the "goal".

So if your computer says that your are at 15 degrees and should be at 5 degrees than tell the servo to go to 10 degrees.  That would be an example of a 50% proptional scheme.  Other fractions are useful too, depending on how quickly the feedback information becomes available and how rapidly the turbine reacts.  See my comment below on ringing.  Over the course of several cycles the adjustments get indisquishably close to optimal.

This control scheme is dynamically stable but tends to oscillate if you set the proportion too high.  It automatically "averages" over time but weights more recent data higher than older data.  You can update it continuously if you like, but check the proportion constant to avoid ringing, obviously 100% is too high.  Halving the update time will require halving the feedback proportion as well.

[BigBreaker]

I just posted a quicky on a PID control scheme.  You are the real deal Amanada.  I played around with a feedback controled robot arm in school.  With just a cap, an op amp and two pots that arm had incredible responsiveness and holding strength.  Do not discount the PID!  Seriously PID basically runs the world and makes otherwise crappy stuff seem magically good.  Most cruise controls use it, for instance.

[Murlin]

BigBreaker and Amanda

Outstanding!!!

PDI.... yepper looks like the ticket to me...

Murlin teh "I love it when a plan comes together"

[Murlin]

Just replying to myself again....

The thought had occured to me that you could  "match the load" Somewhat, by using variable pitch.

The next question, would there be a reason to need to do this?

Murlin

[scottsAI]

Hello Murlin,

All my research suggests that induction is not good in those wind conditions.



I agree, but whether it's PM gen or induction gen, does not matter:

A watt is a watt is a gen is a gen.

Large blades and as the tower documents say go higher for more wind.

For a 24 foot blade power vs wind speed: (by cal and 36% eff)

MPH m/s watts HP

5. 2.24 103.55 0.14
   
6. 2.68 178.93 0.24
   
7. 3.13 284.13 0.38
   
8. 3.58 424.12 0.57
   
9. 4.02 603.88 0.81
   
10. 4.47 828.36 1.1
    
11. 4.92 1102.55 1.47
    
12. 5.36 1431.41 1.91
    
13. 5.81 1819.92 2.43
    
14. 6.26 2273.03 3.03
    
15. 6.71 2795.73 3.73
    
16. 7.15 3392.98 4.52
    
17. 7.60 4069.75 5.43
    
18. 8.05 4831.02 6.44
    
19. 8.49 5681.75 7.58
    
20. 8.94 6626.91 8.84
    
21. 9.39 7671.48 10.23
    
22. 9.83 8820.42 11.76
    
23. 10.28 10078.71 13.44
    
24. 10.73 11451.31 15.27
    
25. 11.18 12943.19 17.26
    
26. 11.62 14559.33 19.41
    
27. 12.07 16304.70 21.74
    
28. 12.52 18184.25 24.25
    
29. 12.96 20202.98 26.94
    
30. 13.41 22365.84 29.82
    
31. 13.86 24677.80 32.9
    
32. 14.31 27143.84 36.19
    
    

Spread sheet has RPM for TSR 1 to 15 for each MPH. Anybody interested in it?

Motor size will be based on how much power and how high winds you want to deal with.

I plan to use a very high L/D blade, TSR = 14, to start producing at 5-6 mph.

Cut in speed will depend on the gear ratio I can find, forcing the blade RPM to match the motor. Using 1800 RPM 3ph motor. Would like 1200 RPM seem to be hard to find.

Most induction motor designs forgo pitch control, giving up the lower speed power.

Thus, the reason for not doing well in low wind speeds.

Since your doing pitch control, I see no reason you can't get power at low wind speeds, just as well as you could with PM gen. Without it's limitations.

The control circuitry to extract all the power over conditions is going to be more electrically complicated and costly vs than the induction motor.

No inverter $$ would be needed to power your equipment or to back feed the grid.

Have fun,

Scott.

[SamoaPower]

Sure, as long as you know the alternator characteristics well enough, which is why I want to comprehensively test mine. You might consider adding yaw control to this.

I could see an integrated controller taking care of it all. I think I'll do separate analog controllers initially to sort out the issues since there are some interactions. Once satisfied, the algorithms could be implemented in a micro-controller, although there is something to say for minimizing single-point failure potentials.

On the pitch algorithm:

It seems there are a few modes to consider.

Stowing:

The default pitch would be the feathered (90 deg) position with output shorted (brake applied) and, perhaps, turned 90 deg to the prevailing winds with yaw control. Controller self tests would determine if it can be brought out of this mode. Error detection would also put it into this mode.

Start-up:

When average (5 minute period) wind reaches 3-4 mph, and self tests pass, yaw position is turned into the wind, the brake released and pitch is set to to start-up. I'm guessing about 45 deg for my blade type (low twist). You do need positional feedback from the pitch servo, by the way. When rotation is detected (5-10 rpm), mode is switched to 'operation'.

Operation:

There are a few options here. I tend to favor using something akin to MPPT to search for and maintain maximum RPM. In any case, pitch would be initially brought to a nominal operating position, perhaps 5 deg.

Power Limiting:

When output current (averaged) reaches a maximum desirable set point, control is switched to an RPM limiting mode. In my case, this would be about 150 rpm. Stator coil temperature could also be used as an input here.

Shut-down:

When the average wind speed reaches a predetermined limit (I'm thinking 25 mph), shut-down is initiated. Pitch is brought to feather and yaw turns it out of the wind. When rpm decreases to less than say, 30, the brake is applied and stow mode is entered. Start-up mode is locked out until average winds decrease to some lower set point.

These are my initial thoughts and are certainly open to modification.

[Murlin]

Great discussion Scott...

No inverter $$  hrmmm....ya that would save 4 grand...no batteries....that would save another 4 grand....

Hrmmm....most excellent points...

No wires or slip rings to the servo cause it's up front powered by PM power....

I would say you have done alot of brain working on this one.

I thank you for sharing your knowledge on this subject.

So you are saying I could just use an off the shelf ST generator heads like the 10 KW one they sell at generator depot for 395 bucks and gear it down to match the blades, ect, ect....

My ...my.....that is one HELL of an Idea.... and it already puts out the correct voltage for grid tie....

Wow, I wish I knew that before I spent 800 bucks on NEO's...

Outstanding man, just outstanding....

You could machine one of the gears out of a hi-density polly and just use cheap strait tooth design to keep the cost and noise down...

Smooth running bevel gears would most likely have to be custom made unless you lucked out and found some gears out of a tranny or something that would work.

I notice no one is comming back with any negative comments from what you have been saying......interesting....

Murlin

[Murlin]

Dude have you been checking out what Scot has neen telling me?

He has some pretty interesting thoughts on all this stuff...

Murlin

[scottsAI]

Hello Murlin,

Thank you.

I almost said something earlier, just was not sure of your goals. Looked like you were having fun building.

So you are saying I could just use an off the shelf ST generator heads like the 10 KW one they sell at generator depot for 395 bucks and gear it down to match the blades, ect, ect....

Yes, and no. Not sure of the durability of using ST generator heads. Secondly, it can NOT be placed directly across a live grid. But an induction motor can. An induction motor can be used as a generator, you run it a little faster it will produce power! For example a fully loaded 1800rpm motor will run at 1745rpm. So, as a fully loaded generator it must be run at 1855rpm. The 3% slip is needed to supply the rotors field. We can make use of this feature by monitoring the RPM, we can tell if the motor is loaded or not. No need to measure the power, just look at RPM to control pitch. The Pitch controller needs to attempt to speed up the blade, if it slows, then return it back. The measurement of the RPM is all that is needed to prevent from overloading the generator in high speed winds. When 1855rpm is reached, no longer attempt to speed up, now keep it fixed rpm. Generator can continue to work by reducing TSR up to 70mph or what ever you design it for. Puts much less stress on the blades than yaw furling more often used in smaller wind generators.

For gears, I am looking to vehicles. Gears seem to be quiet, handle much more HP than I need.

Rear end has a nice wheel to put on a blade, has brake, with ABS wheel speed sensor = RPM or get RPM from the little PM gen (pitch power), then goes to a differential, with a 4x speed up gear. Then going down need another 5x speedup gear, hope to get from a transmission. The induction motor is mounted on the tower, below the rotating wind gen platform. NO twisting wires. Tail is almost normal.

Motor needs a reverse power relay to disconnect it when the winds are below 5mph.

Been working on this design since April.

Click on my user name to get my email. Yes, is an IQ test. Have document on 10kw.

Last month or two, check my posts, have covered a lot of ground in this design.

10kw wind turbine cost: PM gen is $12k, induction generator $6k.

Which fits in your wallet! I may have low balled the PM gen cost.

Includes 24kw of back up UPS, without battery cost, person expects to get free.

$6k is one forth of the cost of the cheapest 10kw wind generator I can find.

Tower is additional cost!

Have fun,

Scott.

[Murlin]

Man I must study up on what you are doing, you have peaked my interest now.

I still don't see how it will make any power below winds of 15 MPH, it seems the friction losses would be too great by the time you shot all that kenetic energy through all the gears and stuff.  But I cannot argue with you because you know a heck of alot more about it than me.

I guess once the rotor diameter got past a certain size, losses would be manageable.

From what you are saying 25 ft is about where it begins.

That must be where small turbines end and large turbines begin.

Very interesting.

Thank you for your time man....

Murlin

[scottsAI]

Murlin,

Many think gears are bad, they can be. I have seen a lot of junk made. Thus their reputation.

Well designed gears should be 95 to 98% efficient. Once you over come the initial friction.

Not much power in low wind speeds so, need large blades. Going up higher helps MORE.

Did you read the homepower's paper here on Tower 101?

Take a look at a vehicle, entire power train has dozens of gears! Yet can handle hundred HP.

The big wind gen use 50:1 speed up gears and an induction generator.

I realized if it's good for them then this big wind gen can use them too, if careful.

Not done with the pitch, need to figure out which is better: lever or gears.

I did the math on a vehicles operating life. Assuming 15,000 miles a year that is 15000 / 60 = 250 hours, so have some city driving so lets go with 30mph ave, or 500 hours a year run time. Life expectancy is 100,000/30 = 3,333 hours of operation. Just little more than a third of a year! Each year your wind gen operates that is 2.5x the life of a vehicle. Something to think about.

Yep, want to build it out of good stuff.

Have fun,

Scott.

[Murlin]

Sammo, I think that if one is going to utilize pitch control, you would not need to yaw if you incorporated a brake.

Using pitch control, I can make the blades come to a dead stop in Gail force winds without ever yawing.

Heck I could even make the thing run backwards.

So the question is, which is better, yaw or a brake.

A brake is going to require slip rings, yaw is not.

Yaw is going to require advanced wind tracking algorythms, a brake is not.

Yaw requires another motor.

You can buy an electric trailer brake for the same cost as the yaw motor and all the electronics to run it.

A brake does not add any tower stress due to the gyro forces.

Wind tracking/yaw control.  I am thinking, why try to do this job when the wind will take care of it automaticly.

Just make the yaw axis as frictionless as possible with a thrust bearing on the top of the pole and nylon on the sides instead of steel.

I think the whole Program just needs to be trying to reach the max RPM in the earliest, fastest way possible.  Utilizing the PID of what has already been done and therefore not needing to actually re-invent anything.

If the logic revolves around that then the program would automaticly start going back the other way after it reached peak RPM.

It would already be predetermined what the MAX RPM would be to keep the heat out of the stator.

And in case of error, a temp switch in the stator would initiate shutdown of the whole system, hopefully feathering the blades before applying the brake.

But the brake, if tough enough, could just shut it down reguardless.

I think all the little gizmo's have already been invented to get 'er done.

The only thing left to do is to come up with a computer program.

You could do it with a micro processor, but the PC wound not only be cooler, it would allow manual control, operation, and observation, right at your finger tips.  

Wireless Electronics are steadily becoming more reliable and cheaper to utilize.

We are fortunate that we live in an age that we do.   It wasn't too long ago that an Idea such as this wouldn't even be worth considering....

Murlin

[Murlin]

When I said you would not need yaw, I was talking about controlled yaw to brake with, not to track the blades to the wind.

We need to be able to edit out posts for a short time.

Murlin teh itch button pusher syndrome...

[SamoaPower]

murlin,

There's more advantages to controlled yaw than just furling. You might want to reread:

http://www.fieldlines.com/story/2006/6/26/12214/2424

"Sammo, I think that if one is going to utilize pitch control, you would not need to yaw if you incorporated a brake."

I think the three are mutually exclusive. They do different jobs. I'd like to use them all.

"Using pitch control, I can make the blades come to a dead stop in Gail force winds without ever yawing."

It's hard to get an appreciation for the issues without having seen the gyrations a machine goes through in turbulant, gusty winds.

"So the question is, which is better, yaw or a brake."

I don't think this is the question at all.

"A brake is going to require slip rings, yaw is not."

Actually, if you use controlled yaw, slip rings are not needed for the brake or any other wiring to the machine. It can only rotate 360 deg.

"Wind tracking/yaw control.  I am thinking, why try to do this job when the wind will take care of it automaticly."

But, not in a very friendly fashion. It's hard on the machine. I'd like mine to last a while.

"You can buy an electric trailer brake for the same cost as the yaw motor and all the electronics to run it."

The brake will cost considerably less than controlled yaw but it only does one thing for you.

"A brake does not add any tower stress due to the gyro forces."

Neither does controlled yaw - one of the main advantages. It only rotates at one rpm.

"I think the whole Program just needs to be trying to reach the max RPM in the earliest, fastest way possible.  Utilizing the PID of what has already been done and therefore not needing to actually re-invent anything."

I think you may be overrating what using PID in a controller can do for you. PID is an algorithm that can be used in feedback control systems to advantage. There are others. By itself, PID won't maximize anything. Three basic elements to a control system. A desired set point element, a feedback element (how are we doing?) and a control element. In the case of pitch, we don't have a set point because we don't know what it is and it's a variable. Using PID isn't a magic cure-all.

You can buy PID controllers for various applications. I don't know of one you could buy for pitch control and have it drop right in and do what we want. It'll take development.

[Murlin]

You make some damn fine points buddy.

I had not considered controlled yaw eliminating the sliprings....most excellent.

On the other hand, if you have a 25k turbine in the air, a 3rd redundant backup would not be a bad idea at all.

But going back and forth from 360 to 0 degrees for an extended period of time could be kinda detramental me thinks...

But you have just given me an Idea on limiting the yaw friction the faster the wind blew.  A soleniod and some brake pads would do the trick just fine on the yaw axis....

PDI is just going to be used in conjunction with a computer program. It wouldnt do it all.

"It's hard to get an appreciation for the issues without having seen the gyrations a machine goes through in turbulant, gusty winds."

Point taken....I have no actual hands on....

Any little gusty conditions below 20 MPH would be easy to deal with, and from 20 - 30 MPH winds the system is on full alert and ready for bear and the rotors are still not spinning over a couple hundred RPM.

But wouldn't you have the turbine shut down anyways in violent wind conditions?

I know I would.  It would be time for me to kick in the deisel.  Weather reports would be monotored quite frequently and if a fast storm blew in, one log on to mypc.com, a couple keystrokes, and the deisil comes on and the turbine shuts down....

And I am miles away, that's the cool part.  A micro processor cannot do this without a ton of electronics.

Great discussion man.....btw...

You and Scott have got my head spinning for sure...

[Murlin]

Ok, after alot of great discussion and a little research, I have found there are a wide range of products available on the market today that incorporate PID Algorythms in their motion control systems.  So as far as the hardware is concerned there is all types of gizmos and gadgets out there and no need to re invent the wheel.

I will choose a single axis servo control unit of some kind, not sure which one to use so if anyone has any prefrences please let me know so I can check it out.

I am thinking that a simple magnetic reed switch will send out signals to a data logger that will collect the pulses and record them in real time and save them to a file.  This will work good enough for the tach.

If you plan on using the motion controller to hold the turbine to a steady RPM (as Scott intends to do) you will need to get a little fancier with the data logger and collect alot of pulses so the servo can update instantaneously like Flux and others have pointed out.

However for what I am doing there is no need to get this precise, but the same principle can be built upon after the initial design is developed.

So basicly all my program needs to do is to read from the Tach file, and move the motion control to a specific point based on RPM.  Updating itself at preset intervals, getting faster and faster as the wind speed increases.

So there will have to be an anonometer hooked up to the data logger as well.

Most of the ones I have seen have 8 channels anyway so there is plenty of room for a temp sending unit and volts, amps and anything else you want.

At the start of the program, the servo will reference itself, probably using a built in TPA in the motion control.

There will be an point offset, for console tuning as well as the pots on the control itself.

I am pretty sure I can utilize a wireless USB hub to collect the pulses from the tach and the other monotoring devices.

All in all, an easy task for my programmer Gurus to do. it will need to interphase with the motion controller somehow.  More research will probably uncover that the software that comes with most motion controllers will make allowances for that, we will see...

Thanks everyone for all the great and educational imput.

Now to build the thing.

Murlin the to grid or not to grid, that is the question.......