16' turbine blades finished, some glassing tests for stiffness

[jacquesm]

So, after almost 6 weeks of continuous work we have a set of blades and a 'recipe' on how to manufacture them fairly easily varying parameters like 'tsr' and 'taper' to create a cnc cutter program that then automatically produces the desired blade.

I know a lot of you must think these guys are nuts, I could have done that in a few days, and you'd be essentially right about that, making three blades should not normally take 6 weeks. But our side of the story is that we also developed the software that models these blades and that simulates their manufacture. That software works for any airfoil (in fact for any complex surface) and will produce a new shape in a matter of hours, without any marking or manual labour.

As long as you don't mess up your filenames (which I managed to do the day before yesterday, ruining one perfect blank) everything goes by itself and the shape will come out exactly as specified.

So, the upshot of that is, if you you just want a set of blades then it is probably  faster to whip our your chisel/grinder/axe/chainsaw/insert favourite chip producing tool here, but if you want to be able to make a lot of blades or many small variations on the same set of blades then the extra work that we have done will pay off. In order to produce these three blades we have also made - and discarded - about a dozen test blades, each of them subtly different from the previous one, homing in on the final shape.

As I have said long ago, I would like to make these blades in foam, because it's cheap and very easy to work (no roughcut, you can immediately cut your final shape), but the problem with foam is that you have to attach it to the rest of the machine somehow and that is not a trivial problem.

So, for now these wooden blades will be what we will be using.

Fibreglassing or not was a question for a while, but we ran a small test to see what the stiffness change in the blade would be after fibreglassing it, and coated a small piece of wood with two layers of fabric on each side and kept a similar piece uncoated.

Then we hung identical weights to the ends of the pieces of wood and measured the deflection.

The thin fibreglass coated piece is more than twice as stiff as the untreated piece !

That's a lot of extra stiffness gained for a minimal weight penalty, so we will be going that route in the next few days, stay tuned !

Pictures follow:

The three blades side by side. They are still in their 'webbing', left over from the blank that they were cut out of, to protect them from damage.

Closeup of the 'root' portion of the back of each blade. Notice the improved section at the root, this is much stronger than before now.

Closeup view of the two test pieces for the fibreglassing test. The left one is the 'glassed' one, the right one is plain. The weights are suspended from the wires. In the 'neutral' position (no weights) the pieces of wood are level with the top of the measuring tape.

An overview picture. (hope your shop is not as messy as mine

[monte350c]

Hi Jacques,

Where's the order form?

Seriously though, very nice work. Will you be letting out any intimate details on the table and software?

If not, I'd be interested in a bladeset -

Ted

monte350sms at yahoo.com

[3Phase]

>As I have said long ago, I would like to make these blades in foam, because it's cheap and very easy to work (no roughcut, you can immediately cut your final shape), but the problem with foam is that you have to attach it to the rest of the machine somehow and that is not a trivial problem.<

This may sound nieave(or howerver you spell the word I am thinking of), but have you considerd using a simple woden bone structure centerd in the block of foam?

I was thinking you could use a long thin piece of wood running the full length of your blade,laminated progresively mainly towards the root where it is full root width and hight. The wooden "bone" could be simple un refigned laminations simply made.  For the flesh of the blade use two pieces of foam half the thickness you need.  The cutting table seems quite capable of cutting half of the reverse of the "bone" in to each piece of foam.  Make a sandwich with the wood in the midle using a little construction glue.  Only a small roughcut is nessasary at the root to feather the wood into the foam and then you are off to the races, with a lot less wood and a posibly lighter blade but still quite strong at the root.  If a small portion of the wood is cut into by the router closer to the tip it would likely be within tolarances of the post rough cut and cause little problems for the router.  Add a littel fiberglass for the stiffnes and you have a hybrid wood, foam, fiberglass blade that will have no competators, In our world.

This may be more trouble than it is worth, but I thought that someone that just spent 6 weeks making a set of blades might be crazy enough to try it.  :-)

Sure would be neet to see.

By the way your work looks increadable to me as is.

Just my 3 cents worth.

3Phase

[gibsonfvse]

Or perhaps you could use a carbon fiber spar or a wooden spar wrapped in carbon fiber or kevlar and then vacuum-bagged.  It should be extremely strong in bending; I'm not sure about the torsion loads, though.  Here's what I'm thinking of: the Allegro R/C glider.  Take a look at the "Building the winch-proof spar" section for some ideas.  In fact, the R/C discus-launch glider field may have quite a bit of knowledge to offer to such an endeavor as this.  I think what you've done is excellent, and well-worth the time invested!  While I'm thinking about it, may I ask what airfoil sections you used for your first set of blades?  And is the twist from root to tip linear?  

My thoughts on attaching foam blades to the rest of the turbine is to do it by the spars.  If you make a foam blade with a spar, you should be able to embed the spar into the hub.  Even better, use two spars in the blade to alleviate the torsion loads.  Good luck!

[JF]

Dear Jacquesm and other friends

Many congratulations to you and your friends with your long and trying work. This is truely a good development for all.

I think that by now you must have deserved a good professional Bosch GOF-model router or perhaps a DeWalt.

I sincerely hope that you will make your program and also a more detailed description of the whole frame and control-system etc setup - freely available, so that others may also have the possibility to replicate your good work.

With best wishes to all - JF

[jacquesm]

wooden spar wrapped in fibre ? that's a neat one, had not thought of that. We'll definitely look into that as soon as the current set of blades & the machine are up and running. We plan to use it as a testbed for various configurations.

The airfoil is 'regular' 4412, about 10% height to width ratio.

the twist from root to tip is fairly aggressive at the root, then less and less towards the tip.

Taper is limited in width near the root to about a 10" near the root, artificially increased to a minimum of 6" near the tip end.

the blades have a 'design' TSR of 8, but I realise that is not very realistic (saves a lot of wood though .

I very much like the central spar idea, someone else was also thinking in that direction and we will definitely do something with that over the winter.

regards,

 Jacques Mattheij

[jacquesm]

Hehe, we'll see if Bosch is amendable

Yes, we will be making all the data available. That takes a lot of time though, but trust me, we'll do it. Both the CNC machine and the windmill.

best regards,

   Jacques