Blade comparison with data

[opo]

The graph looks reasonable for the GOE222, based on my experience flying them, even with furling.  The power just continues to increase the harder the wind blows, even with the machine running furled.  I have a tachometer on one of my machines and even as TSR drops as wind speed increases, there is never a drop in operating rpm due to furling.
Chris

Recall that a furled rotor only sees a slower wind speed, that is, a 3m diameter rotor furled is still a 3m dia rotor experiencing a slower wind speed. So, according the the graph and your experience, your blades need to furl early and be fully furled to be tamed. A lower pitch, less than 10 deg, moves the graph to the left (essentially), meaning that you may be able to move the saddle in the power curve closer to 20-30 ms.

opo

[opo]

Interesting.  Then I would have to assume the software calculates a drop in operating efficiency at about 45 mph for whatever reason.  How hard would it be to graph the S809 for this same scenario?  Those I have run on a non-furling machine at very high wind speeds and the particular blades I was using topped out and never went over their peak even with no furling.

It would be very interesting to see if qblade comes up with the same general idea of what I have seen when running S809's.

Chris

I'll give it a try tonight to the S809's.

[Flux]

If the simulation is for the speeds that seekscore has measured then I suspect the flattening over on the Clark graph is as it hits stall. This is probably confirmed by the other case with 10deg where the stall will hold off to higher wind speed.

If you use a loading scheme that doesn't let the tsr fall to stall then the power keeps rising as you would expect.

I really don't know enough about the simulation to be sure about any of this but it is very much in line with what I would expect. To keep the power rising you need to keep on top of the power curve and in theory that means running at constant tsr. I actually find that with my blades the optimum tsr becomes lower as the wind speed increases, but it doesn't have to go much too low before the power drops right off.

I am fairly sure that having taper and twist makes this issue worse, we all seem to slavishly follow this ideal that the calculators throw at us without question but I have found out from experiment that the excess twist and chord at the root is not beneficial. I am not totally convinced that twist and taper are of benefit either unless you can load to maintain the constant tsr assumed by the calculators. It may not only be the high lift profiles that work without twist and taper.

I have always tried to match the alternator to the blades but with Mike's experiments matching blades to the alternator you may get some surprises.

The only thing that may not work out is that by matching the blades to the alternator you may not in fact be able to do anything to the alternator to improve it afterwards. This is fine if it gives a better energy capture over a wider wind speed range with simple direct loading.

I wouldn't use the same blade approach for use with the classic or other mppt scheme without doing a lot more checks, each blade type would need a run with different programming. It may in fact work out that the swept area is the main factor and if you can get the matching right the blade profile doesn't make much difference.  if you can't do the matching by mppt then there could be a lot to be gained from changes to blades but you would never sort this all out without accurate data logging. keep up the good work Mike.

Flux

[Seekscore]

I have thought about what performance I may be able to get after I find the best blades for how my alternator is loaded now. I wondered if there would be much to gain after I open the air gap a bit. I have thought that once I open the gap, that the blades that performed best before air gap adjustment, may not be the best match after I tweak the alternator.

At least when I get to that point, I'll know...

Like I said, when I set this up, it was adjusted to cut in at 120 rpm. Up on the tower it seems to cut in below that so I am sure that the air gap needs to be opened up a bit. I know that is going to change the alternator load somewhat so I am hoping the best blades are still the best blades.

Mike

[birdhouse]

this is great testing!  it's really cool information to look at with pretty solid data tracking to prove what is happening every time you change a parameter. 

seeings how you are looking at finding the best blade to suit your alternator, have you thought about fudging the lengths of the blades to suit things better as well?  maybe this is just throwing a wrench into the mix, but it could prove to match the load well.

you mention the cut in may be too low.  maybe a slightly larger blade set that will naturally spin slower, yet have the umph to match the stiffness of the alternator because of its larger swept area. 

keep up the stellar work!   

adam

[opo]

I added a rotor with same dimensions, no twist no tapper with the S809 profile and 10 deg pitch.



Hope this is helpful Chris, (and not too off-topic). I do believe this type of simulations may help in the understanding of what is going on. Already I realized that pitch control could be a very effective mean of controling power output. By changeing the pitch angle in the simulations the power graph essentially moves right or left. This way, in higher winds, one could controll the blade stall in a wide range of wind speeds by keeping the rotor near the saddle of the power curve. Too bad it is difficult to implement.

Cheers,

opo

[fabricator]

Mike,
I wish I had more time than others to write on the forum.

Then I could answer your questions in much more detail, instead of letting others who don't know what they are talking about mislead you.
I have tried to explain and illustrate it to them before but to no avail.
Airfoils are easy to understand, unfortunately there is so much BS out there that the novice can't filter out the good from the bad.
You say you want to learn most of all, I'm in it for the same thing.  Others are out to prove their own points, and you've just walked into their pet project domain.

I'm pissing off the usual PO'd crowd but right now I don't have the time to do anything more polite.  If you keep listening to them (because they are louder than me) then you could be making junk thinking it's gold.  The best you'll get is making so-so blades without understanding why.  You have a fighting chance of figuring out for your own self what's better since you've got a fine datalogger, which is miles ahead of those that keep telling you what to do.

As long as he is experimenting and it's relatively easy for him to make blades what possible negative aspect could trying a wide range of air foils have? With his excellent data logging set up I can't see anything negative here, I guess I don't understand the negativity here.

[ChrisOlson]

I added a rotor with same dimensions, no twist no tapper with the S809 profile and 10 deg pitch.

Thanks for posting that.  I was curious to see if the S809 had that "saddle" where I found it to be when I was experimenting with a non-furling turbine.
--
Chris

[ChrisOlson]

I am not totally convinced that twist and taper are of benefit either unless you can load to maintain the constant tsr assumed by the calculators. It may not only be the high lift profiles that work without twist and taper.

The Bergey XL.1 uses an airfoil that looks like a GOE222, but I don't think it is.  It has no twist or taper.  They cut a notch out of the trailing edge of the airfoil at the tip (I'm not sure why) and they also have a sort of ridge on the backside of the airfoil at it's fattest point.  The purpose of that ridge is to make the blade run quiet.

There has been many successful commercial turbine designs in the past and present that use blades with no twist or taper.
--
Chris

[opo]

Can somebody provide me with Dan's blade dimensions? Is there a calculator for Dan's design somewhere?

Specially the ones beeing discussed.

What I need is the following data

station    angle(pitch)     chord
1                   
2
3
.
.
.

Thanks

Octavio

[Seekscore]

I don't have the exact layout for their blades for a 12 foot machine but as far as I know, they are just a scaled up version of the 10 foot machine blades.

The width at the root is 8.75" and the tip is close to 5.5" The 10 foot blades are listed in the Google book starting on page 197. I'm not sure what this link is going to do but you can select the contents at the top. I think there is only three stations on the blade.

http://books.google.com/books?id=0kKoyE7EZ9cC&lpg=PP1&pg=PA197#v=onepage&q&f=false

Mike

[kevbo]

You can see in the image that the twist stops at a certain point only due to the width of the root. I get as steep of pitch as I can and then carry that to the root.

If you made the tip "negative" pitch (with hub flat, as you showed it) and then mounted the root between some wedges, (to restore the tip to correct positive pitch)  you could get a lot more pitch at the root for not much more lumber (just the wedges).  

But maybe you thought of that and have a good reason for not doing it?  Ain't much power in the middle, but having steep pitch somewhere will help a stopped blade get moving.

[fabricator]

The root aint worth the trouble, the root provides pretty much nothing worth worrying about.

[Seekscore]

The blade doesn't have a negative pitch. The angle of attack at the tip is 5.5 degrees. It follows hugh's spreadsheet from the tip until the width of the blade prevents any steeper pitch. Probably 60% from the tip to the root.

Mike