planning upgrade to 10 footer

[Boss]

For the first time in a while I am planning to upgrade our 10 footer AND  I am looking for input on a idea I have.
I have lots of info and pics on my site http://outfitnm.com

Tower lowering video can be seen at http://www.youtube.com/watch?v=oca6DTmTIWM
 
Though-out this Summer I have watched the turbine and it is often resting in a position with the tail extended all the way to the stop. It rarely furls since we made 6 inch longer blades. From reading here it is apparent that the turbine off-set is lacking.  Also this Summer we doubled the quantity of batteries more about that can be found at http://outfitnm.com/2010/04/22/bmn-battery-bank-wiring-or-how-i-learned  as well as here on this site where I documented all of this wonderful project.

So I'm thinking if I have to pull down the turbine and cut off the turbine housing and add an inch or whatever to the off-set then this may be the time to build new blades and turn this into a 12 foot turbine and hopefully increase the power going to the larger battery back.

First this is engineering new blades using Solidworks  for use on our neighbor's  CNC router. I am thinking the blades need to wider at the base as well as longer. My designer (Kevin) is working on the aerodynamics issue and already told me we need to get a wind data logger up so we can see what the old blades are doing compared to the new prototypes.

We're looking for any information about blade designs that are working well for others.
This is what we have now and sorry until Kevin returns from his Summer motor bike travels I don't remember any of the terms

From http://outfitnm.com/2009/11/03/bmn-new-cnc-wind-turbine-blade-set-together
Previous blades



Maybe something like these found on this site


http://outfitnm.com/2010/05/07/bmn-airfoil-designs-grammatically-challenged


--
Brian Rodgers
Biodieseler, axial flux wind turbine, food gardener, on the side.
Part time  job: I'm a rural (mountain) & urban wireless Internet installer.
All Internet traffic relayed through remote solar powered WiFi tower. http://outfitnm.com/

[Janne]

Hi,

Not sure how your systemi is wired up, but if you have the turbine charging up the battery bank straight through the rectifiers (the usual way), then trying to engineer a set of blades is going to be hard. The problem with the straight connection in the battery bank is the load matching, the rotor will be perfectly matched to the load only at a certaing part of the power curve, which can be adjusted by adding resistance to the line, cut in speed etc, but it's always a compromise. For this application the best blades might be some crude "back of the envelope" blades, even with no taper / no twist, because even if they're running out of their power curve, there are always some portions of the blade that are providing lift. If you put an engineered set of blades into a machine loaded like this, the whole blade will stall at once.

If the load matching is taken care of, for example with an MPPT controller, then the engineering on the blades will start to pay off. There are spreadsheets available for calculating an optimum blade shape for a certain Tip Speed Ratio (TSR), I think Hugh Piggot's site is also hosting one.

That's my 2c worth on this matter.. Nice looking system BTW, I liked the cushion you used when taking down the tower =)

[Boss]

I am thinking about building a PWM for testing looking at http://en.wikipedia.org/wiki/Pulse-width_modulation
Has anyone done this? If so where would the best place to put the PWM, prior to the rectifiers?
Perhaps on the three phase AC side

The goal is to attempt to keep the TSR and RPMs higher

[joestue]

depends on the freq, topology and purpose.

you want to keep rpm's higher.. means you want to reduce the voltage?

the issue with doing that on the ac side is it increases I^2R heating in the alternator significantly, or you have to replace all 6 diodes with bidirectional switches, and that's only effective for high voltages due to conduction losses in the switches, and you will need about 30 mips or so to control it. Essentially what you end up with is a 4 quadrant synchronous motor controller.
Replacing the 6 diodes with scr's is an option, but you wiill have to install sufficient capacitors line to line, and inductors between the caps and the scr's

buck regulators are cheap, but the losses in the diode get excessive due to the increased current, and with 24 volt systems you can't really get away with using the 30 amp 45 volt, .3Vd schottky diodes that are so common in psu's. but fets are so cheap now that there is no reason not to use them for the low side too.

As far as buck regulators go, on every modern computer motherboard there is a chip that will interleave 2 to 8 phases. you have to interleave more than one phase (build two or more and run them in parallel)... or you have to spend money on good quality capacitors for both the input and output, both to keep the emi off the lines running up the tower, and to decrease weird stuff from happening when there is too much leakage inductance in front of the switch..  stuff blows up.

You can even go with a high frequency, 150Khz - 300Khz, and use air core inductors wound with litz wire extracted from CRT's
core costs are then zero but switching losses higher, and you don't have any magnetics to mess with.

boosting the 20-80 volts from the turbine to 60-400vdc then convertering to whatever your batteries are at now is also an option.