Grab's original post linked two different pages. I did look at the first one immediately to get an idea of where he was going, but only glanced at the second. I just got around to taking a closer look at the second (Tim Hunkin) link.

In Hunkin's first picture it looks like he's attempted to use an automotive "radiator fan." I can't tell from later pictures whether he got something much better for later trials, but the auto fan is one of the worst possible "randomly selected" propellers one might choose for a wind turbine project. It's designed to push a lot of air at low speed (low rpm), but doesn't derive much torque from air flowing through it. The thin flexible blades, in fact, are intentionally designed to "bend out of the way" when ram air due to vehicle motion is sufficient for engine cooling, and to "flatten" and pump nothing at high engine rpm (where it's assumed the vehicle's moving and the fan isn't needed).

If one must use a fan of this kind, he's got it mounted backward to what is needed to get best (which won't be much) result from this kind of fan.

A fixed, stiff-bladed fan from an older automobile (pre-1950s emission regulations and the push for efficiency?) would be a slightly better choice. A blade salvaged from a "shop fan" would also be a possibility, and you might find one up to about 30" diameter if you look around a bit. But for an effective "propeller" generator, I'd suggest visiting the nearest model airplane shop and asking for "the biggest ya' got." That probably won't be quite as big as you'd want for a practical generator, but would give you a starting point for trying out things.

An "airplane propeller" actually is "built backwards" for use as a generator driver, but the efficiency with a true airfoil will be very much higher even "running in reverse." For an airplane prop, the "angle of attack" is off the plane of rotation of the prop, so that rotation produces lift in the airplane forward direction. For a turbine-generator prop, the angle of attack needs to be off the perpendicular to that (from the plane of the blade and direction of flow) to produce lift in the direction of rotation of the prop. Approximately, the prop blade needs to be rotated 90 degrees on the hub, and the propeller ¹allowed to rotate in the opposite direction - not an exact description but close enough.

¹ If you want to keep the same direction of rotation, you have to reverse the twist of the blades.

For a starting point for developing your propeller turbine, a full airfoil blade section probably isn't necessary. You'll need it when you start to work up the efficiency, but a proper "lifting line" in a flat (metal?) blade would give reasonable performance for a start.

The classic reference is Abbott and Doenhoff, Theory of Wing Sections, from NACA wind tunnel research ca. 1949, republished 1958 by Dover and probably still easily available since it's a "classic." My copy was $4.50 but that was in 1970, so it's probably a bit more now.

I won't attempt to estimate how many draughts you'll need to buy for your neighborhood aerodynamicist for assistance with the calculations, but they're not really all that difficult once you get the definitions straight. (Although getting the "twist" right, so the angle of attack varies with distance from the hub can be tricky.) Just be sure to design so that tip speed doesn't exceet about 0.7M at your max design wind speed, for the airfoils in the book.

If you prefer to work with the "vertical" turbine, considerable information on the design was published when one of the ocean research vessels tried using one as a sail. The generic search term "Darius sail" or "Darius turbine" should get some info, although the sail application may be a bit old to find much of the technical stuff via Google.

I will suggest that although technology has advanced somewhat, and the requirements are better known, it's not likely that you'll get much better results with a "home built" than was available when an uncle tried a commercial wind generator ca. 1950.

Results were:

In Kansas wind with avg wind velocity 27 mph (approximately),

on a 40' windmill tower on top of a hill,

with 2-bladed prop direct coupled via a step-up gearbox (unknown ratio, probably 4:1 but maybe 7:1?),

running 24 hours per day, 7 days per week,

he was able to keep a bank of 6 or 8 auto batteries sufficiently charged to listen to the radio for about 2 hours at a stretch - usually on at least 3 days per week.

You can do much better with commercially available hardware now; but location and weather are critical to whether you'll get useful results, relative to what you hope for.

John