Project PMA wind turbine model

[scottsAI]

Goal is to take local weather data, define a system in detail and predict system performance over years of simulated run time, hour by hour.

Minimum data needed:

Cut in speed and current, battery voltage and Ahr.

Diameter of blades, number of blades, any info on profile, twist etc.

Generator resistance, line wire size, length, single, DC, three phase etc.

Must have actual wind speeds, If wind speed are based on RPM please say so.

Would like measurements at twice cut in speed, and again at three times. Or any 3-4 points.

Guess on how accurate your data is.

Extra data:

If you wound the generators coil turns, wire gage, length anything else your willing to tell me.

Rectifier type, magnets, inverter model or power range.

Load dump? How often used?

Extra extra data:

ZIP code, guess at power use, daily, weekly, monthly etc. Off grid, on grid.

Anything I forgot to ask!

I found some data here, but a key piece always seems to be missing so figured to ask.

3-4 sets of data would be sweet. Thanks.

If you don't want the data posted please email it to me. Will keep private.

Yes you have to pass an IQ test to send!

Have fun,

Scott.

[finnsawyer]

Seems to me all you need is the cut-in wind speed, the output voltage varying linearly with wind speed, and an efficiency of 35%.  Then given the wind speed history at a given site you can determine the power output as a function of time, even for years, for any particular size mill you wish to consider.  Why are you making this so complicated?

[scottsAI]

Hello finsawyer,

Well because it is.

If you want to know if diodes, Schottky's, synchronous rectifiers, Y, delta, Jerry rigged, Blade profile changed to another, Blade pitch changes, wire 000 AWG, 0000 AWG, DC or single phase or three phase or Jerry rigged, or 70' tower or 120' tower, adding series resistance for heating, AWG of wire in generator, turns of wire, strength of magnets, air gap, Larger/smaller battery, change system voltage...

I like to write complete models, just like I do when simulating the entire electrical system in a vehicle.

Need data to confirm math of the interaction of the blade to the generator and to battery.

I found a good blade profile simulator but it cost $3k, so need something else.

Found simulator for a wing profile, do not have the concept working as a rotating blade.

Any help with the above would be appreciated.

Have fun,

Scott.

[Flux]

I don't understand what you are trying to do. If you are hoping to collect data from various people on performance of different systems then you have no chance. There is virtually no one with the ability to collect accurate data .

Measuring wind speed is a major issue and even if you can measure it at the machine height you will need to analyse the data very carefully. I still wouldn't expect to find much correlation between different observers and different sites.

No problems with the electrical side, we can all measure alternator performance to a reasonable degree of accuracy. To be of any use I think you will need to do all the wind measurements on the same site with the same equipment.

Even doing a run on a fairly low wind day  and a high wind day will give quite different results even though you have a decent number of samples in the wind speed bins.

Probably the higher the tower and the better the wind quality the more consistent the results.

Perhaps I have misunderstood what you are trying to do.

Flux

[finnsawyer]

I'll take a pass on the first part, the diodes and all.  As far the power output from the blades, if you can calculate the lift as a function of the blade width you've got it knocked.  Then all you need is equate the power in a small section of the blade at a given radius to the power in the air stream for the small annular ring at that radius.  I've given the necessary equation in a recent comment.  Maybe I should put it in a diary.  You can use the standard equations for lift and drag for an air foil, relating the lift to the total width of the blades.  I don't know what the nature of your simulator is, but you could write your own.  The only place where calculus would come into this process would be when you need to integrate the power curve (power from lift - power lost to drag) over the entire radius.  Some might take issue with this, as they always seem to do.  If anyone has a better Idea, they certainly are invited to let us know.

[scottsAI]

Hello Flux,

I figure if I can simulate a car, then the wind turbine is possible.

Not easy but, possible.

I have collected some data from posts over the last few months, searching did not get me anything.

Used this data to create the first model.

With the data I have managed to collect, the generator and the blades interact with each other in surprising ways, often people make (incorrect) assumptions, then test without the expected results.

I would like to make a simulator so you can at least give it a go first then consider building it.

I am used to getting loose data, as long as all data is collected the same way I can "fix" it. Physical laws are hard to break!-) The math should describe the system behavior if correct.

If the system behaves differently then something is wrong with the model and must be fixed.

I like to understand something before I build it, considering I have at least a year before I will move and have the land to put a wind turbine on it.

Designing my next home: a Zero Energy Home, coming along surprisingly well.

Does not seem to be much interest here for it, so need to find where it is.

Thanks,

Have fun,

Scott.

[scottsAI]

Hello finsawyer,

Yes, please do!

I have the open source wing profile program.

I can see how to break the blade into sections of a wing.

I have not wrapped my brain around how to convert the sections into one rotating blade(s) with an active load called a generator into power.

The interaction of the generator to battery is easy, and blades to generator is easy...

Thanks,

Have fun,

Scott.

[SparWeb]

Just some friendly advice, Scott:

I went down the road you're on, as a "winter project" once.  It doesn't lead anywhere.  This fact wasn't obvious until I started actually building what I figured was a "good starting point".  With an alternator built, blades cut, and a tower under construction, now, the MOST satisfying part of my hobby is seeing things put together and working together.  Ahead, I have a tower to raise, and the joy of watching my creation feel its first wind.  No computer will give that satisfaction.

Next winter, when my garage is inhospitable, I will be back at the computer calculating the next design, but only to reduce the summer & fall's data collected, and forecasting what can be built next year.

Don't fret if you don't agree, I'm not telling you what to do.  Just sharing experience that I think might be helpful.

Good luck, in all your endeavours.

[finnsawyer]

Well, you can measure the power out accurately.  The next thing you need is the wind speed.  As flux points out, this is a major problem.  The solution would be to mount the mill on the front of a vehicle as far forward of it as you can and as much above it as you can to avoid effects due to the air flowing around the vehicle.  Then you must measure the vehicle speed vary accurately.  Finally, you need the RPMs, since the mill may not actually be operating at the design TSR..  You can get the RPMs from the frequency of the output voltage.  

Next you need to model the blades.  You need to know how they were constructed.  Given that, you find an air foil profile that closely matches the one used.  Then you can use the standard equations for lift and drag to find the component of lift force and drag force in the direction of rotation as a function of radius.  You get the power out at that radius from the equation:

      P(R)xdr (Fl - Fd)xV(R)xdr

I've added the dr in the equation to emphasize that we are talking about the power from a small section of the blade.  V(R) is the magnitude of the blade velocity at R.  P(R) is actually the power per unit length at radius R.  The forces are also force per unit length.  To get the total power you integrate over R.  Problems can also arise with this modeling.  P(R) as determined from summing the effect of all the blades may exceed that allowed by the Betz' Limit.  How do you deal with it?  You also may find that parts of the blades exceed stall.  How do you deal with that, when the operation is outside of the usual range of angles for the air foil operation?  Well, you could forget about modeling the blades and just model the overall wind mill behavior.

Anyway that's my take on the issue for what it's worth.  I think you would need better data than you are likely to get.    

[scottsAI]

Hello SparWeb,

Always interested in friendly advice.

Yes, it does sound like you have gone down that road. The fact that you went makes you sound like me. Did the wind turbine follow your model?

I lost interest in building things 10 years ago. I try to find people interested in what I'm doing, work with them to get it built. Have several books full of stuff that will never get built.

Need a whole room freezer?

How about air conditioning using 100w of power?

Near zero power refrigeration (34'F system)

Whole roof solar collector, cost about the same as normal shingle roof.

-want to warm up your garage in the winter? Needs south facing roof.

10kw grid tied wind turbine. ~$2.3K

Etc. Whole bunch more.

Sounds like you have data? Not willing to share?

Yes, I do not give up easy. More data will leak through and I will use it. Good or bad.

Most of my work life I have worked on things that were impossible. Just my style.

Need help with any interesting designs? Next system?

Have fun,

Scott.

[SparWeb]

All the data I have, that is worth sharing, I have already posted.  Lathe tests of my generator.  My mill has yet to rise out of my garage, but the tower is steadily going together.

I threw the "from basic principles" model in the garbage (recycle bin icon, actually).

Nowadays I curve-fit to test data.

Just my style.   :^)

Good luck to you.