Guess-o-matic Airfoil

[wdyasq]

This is a continuation of johnlm comments on Dan's airfoil and angles.  First I'll state the TSR of the mill is unknown.  There has never been posted a windspeed and RPM number at the same time I can find. The best I can find is --This one seems to be starting to furl at about 1.5KW which is happening at 150 rpm.  If thats 25 mph.. then it's stalling badly,---

Next, we would need some airfoil data to "backplot" and hazzard a guess as to the -zero lift angle- of the airfoil.  In addition, the contures of the airfoil would need to be followed closely and checked with templates. Being done by hand and eye, I seriously doubt this was done.  The data would need to be for the proper Reynolds numbers also.  From there it sill still be a guess.  The Tip Speed Ratio is wind speed and tip speed ratio, not what we would like it to be.  If we could wish it so, it would eliminate a lot of extra work.

Research has been done on airfoils of far greater range of angles of attack before stall than an eyeball knockoff of the Clark Y.  Would they be better?  Would the blade just overspeed? Would one get more power? Would one get better lowspeed performance?

The answers are no one knows.  There have been studies of airfoils that claim 20-30% more production, better strength and better lowspeed performance.  The blade profiles are difficult to cut by hand and have cusps in the airfoils.

Ron

[johnlm]

Ron,

From previous posts Ive seen of yours, it looks like you have worked alot with airfoils. So please bear with me as I try to pick your brain to figgure this all out as I do not pretend to know alot about this. And Im not trying to second guess or be critical of peoples designs here , only trying to understand what is really going on, on my props as well as others.

For people just trying to make a blade from scratch with a desired (at least approximately close) tsr, without following an established airfoil, similiar to the simple drawing I posted on Dan's post how is it done?  In the blades Ive carved the chord (longest line from leading edge to trailing edge) is very close to the same length as the front face of the blade.  So I have always calculated the face angle  ( angle relative to the wind flow coming in at a 90 degree angle to the face) in the manner shown in my drawing, and thought the approximate TSR was this slope at the tip ( making the assumption the face angle and the chord angle are nearly the same).  As I mentioned in my post, The actual RPM I get (no loading) is always close to what I was trying to cut.  As best I can measure things -measuring the frequency of the AC output (Hewlett Packard multimeter with frequency setting or using an Oscilloscope)) and adjusting for # of magnet pole pairs I get revs per second which gives me a pretty accurate reading of RPM.  Measuring the wind speed is a bit less precise but Ive stood off to the side of a prop that was not too high off the ground with a store bought wind speed meter and also used as many intuitive factors as possible to judge if the meter seems to be correct.  Even if I assume the wind speed measurement is +/- 20 % I am still in the ball park when I compare what the RPM should be if I cut the TSR at some number and then compare it to the real world RPM's for a given wind speed.  I just finished a 3 blade 4.3 ft prop that I tried to cut 10:1 TSR and when measuring wind speed VS RPM (multiple samples taken then averaged) I am really getting 8.75 :1. Carving error? Some other factor I havent taken into consideration such as how flat do the bladed mount to the rotor hub?  All the measurements were made with very little loading on the prop.  I have not taken exact measurements to substantiate the following statement; but I notice that as I add loading to the prop it slows down slightly as load is increased up to a point where the loading gets high enough (say equal to about 35 to 40 % of the theroetical max for the given prop size and given wind speed) then it starts slowing down (stalling?) at a much faster rate. Ive always assumed the stalling prop resulted in slower RPM resulting in lower electrical output until the RPM began to stabilize at a lower RPM and a lower power output, and no longer funning at the intended TSR but at something lower for the given wind speed at that instant - essentially slipping relative to what it should be running.

So what errors in approach or calculation or reasoning am I making?  Is actual TSR measured at relatively light prop loading or at something greater?  

I agree with you that not enough data was given to determine the actual TSR in Dan's report.  There was another piece of data given that said he saw 1.5KW at 300 RPM.  If I understand how to calculate it correctly, assuming the blade is really a TSR of 6, then the wind speed needed to achieve 300 RPM on a 17 ft prop would be around 30 mph.  Was the wind really blowing at 30 MPH?  Im not sure, except in all of his report, he mentioned only light winds, and no mention of winspeed guess at this data point, but there was no mention of high wind either and I consider 30 MPH pretty high.  As a double check on this reasoning, if he was getting a typical power output from the system of 1.5 KW as calculated using P=0.0012 X Dia^2 X windspeed^3 (mph and ft. as the units) then he could have gotten this amount of power at slightly over 16 mph wind speed, but if the TSR is actually 6:1 this would only be giving him about 160 RPM, yet he stated it was at 300 RPM.  Thats one of the things that leads me to think the real TSR is much higher than 6:1, or 7:1.  Note to Dan, I realize your annimoter was broken at the time, but what would you guess the windspeed to be when you made this measurement.

Im totally open to comment / correction.  Thanks

Johnlm

[johnlm]

My mistake, To many numbers to remember without double checking the post. I said that he (Dan) saw 1.5KW at 300 RPM.  What Dan actually said was, while using a series resistance:

"the best I saw was about 3.5 KW at around 300 rpm" .

The double check example I gave in the last paragraph in the prior post should say.

He could achieve this 3.5 KW at between 20 and 21 mph wind speed using the stated rule of thumb formula (as above), and if the prop was a TSR of 6 he would be getting about 200 RPM at 20-21 mph and if it was a TSR of 7 it would have been about 235 RPM at this wind speed.

Please understand this is not an exact check as many other variables (with associated large tolerances) were introduced in this double check.  Its just another rough check, given there is not enough data to determine more precisely what is really happening.  Working my (this revised one) double check example backward with the info given one comes up with a TSR in the neighborhood of 9.

In Dan's information he stated that adding series resistance to the output (which would load the mill less at a given RPM) made the prop spin faster (and I assume here allowing it to move eithr up?? or down?? on the lift vs angle of attack curve, thus getting into a better operating area.  This brings up a question to me again that Dan and I discussed once before.  If one knew exactly the calculated (designed) TSR and then tested the prop in the wind, does it run faster than the calculated TSR with very little loading and then under load slow down to the real TSR, at which time if loaded further it begins stalling and then slows down much faster;  or does it run at the calculated TSR with little load and slow down slightly as additional load is applied untill it reaches a point that it looses connectivety with the wind and starts stalling thus slowing down at a much faster rate with any additional load?  

johnlm

[Victor]

Most people are talking about a TSR under load,as that is the only way it can be of any use. If the turbine is unloaded it will overspeed usually 20-50% depending on many factors. When looking at angles of attack don't forget that a loaded turbine slows the wind which effects the relative wind angle.

Victor

[johnlm]

Victor,

Thanks for the info.  Any suggestion as to how much loading?  25% of theoretical max?

More or less?

John

[wdyasq]

Most people are talking about a TSR under load,as that is the only way it can be of any use. - thanks Victor, that is exactly correct.

----For people just trying to make a blade from scratch with a desired (at least approximately close) tsr, without following an established airfoil, similiar to the simple drawing I posted on Dan's post how is it done?-----

It isn't - unless one has some data, it does no good to calculate.  It's sort of like me driving to a town in Colorado without a map from Texas.  Sure, I can get to Colorado, but finding Opher, Colorado or Climax without further information?

Note the airfoil is flat but the AOA would be 2 degrees positive.  That would change all your numbers if you were using the bottom as -zero-.

About the only thing we can do is try to get accurate data and record that data.  It will help if everyone uses the same point for calculation.  It would be wonderful if we could place all of these various airfoils on the same mill at the exact same time.  Until then - if ever- all we will have is guesses and as long as there are guesses we will be guessing at what the TSR is.  BUT - once we know what the RPM is, the mill diameter and the speed, we can get the TSR.

Ron

[rotornuts]

I'll gladly tell you the tsr of a rotor! Right after it's up and running.

[richhagen]

I'm currently trying to plot an airfoil to carve out of a 1 foot 1 by 2 for a 'toy' mill with a stepper motor.  (3 blade 2 foot radius) I plan to cut it on my mini cnc mill.  I've reviewed a lot of posts on airfoils, and this, Hughe's, Ed's, and a bunch of other sites.  I'm a bit confused about some of the facts, and face the reality that I don't know the loading range.  I've basically guessed at the angle at the tip to approximate what I think should give me a TSR of about 6 as I'm more concerned with low wind conditions, and approximating an angle at the tip, smooting it back to the root which will be as steep as the small board allows, and then, what I still have to code is an approximation of an airfoil for the backside.  Then if it doesn't start easily enough, or if it stalls too badly, I'll make adjustments to the pitch on the next set.  I figure I'm just going to guess based upon one of the profiles I've seen.  I've made small wooden blade sets like this before, and while probably not optimized for efficiency, they seem to work OK.  This will be the first trial on a mini-mill. (plagerizing Jaques idea and method in miniature here by using a cnc mill)  This seems pretty similar to the theory that I've read here that Dan and others are doing and they always seem to get great results.  I'm all thumbs, so I may still be struggling on trial prop set #33, who knows.  Rich Hagen

[ghurd]

I for one, would like to see what you come up with!

I was considering 1x3" to get a little more width at the root. I know nothing about blade design.

G-

[richhagen]

Assuming it is not too embarassing, I will post a story on it once I have some results.  Rich

[ghurd]

Dude!

I have decided small is more diffucult than big.  Screw up 10% on a 1000w machine and it still works really good.  10% on a micro-mini and you get nothing.

You should see what I am doing!

Like 20" factory fan blades of every discription... central air units, garden windmills, window fans...  Then there are the 5 gallon bucket 16" (each) ZubWoofer blades...

And when I get one hung on "The Tower" (see 'garden windmill' above), I'll post it!

Shouldn't be long. I just got it to light a 12v LED bulb, in the wind, holding it, with many other bad conditions... About 2 hours ago.

I would love a good small 3 or 5 blade design about 25" dia. for high torque and low wind speed.  Problem is I just learned enough to be confident and get a good mini conversion done, now I am starting on blades.  I knew about watts.  The blade stuff is starting 90% from scratch.  I'd like to scale down a big blade- but know that does not work right.  But I know enough to know that I don't know what I need to do next.

I don't know enough to run the dang blade design programs.

Embarassed or not, works or not, a lot of people (me) will learn something!

G-

[richhagen]

If that town your looking for without the map in Colorado happens to be Fort Collins, just look for all the hombrew gennies on the tall towers and you know your close!  Rich