Methods for machining gas turbine wheels

[DeusExMachina]

This is my first post here.

I had an idea the other day and after some research found that it already exists (always does), anyways the idea was to create a parabolic trough with some sort of heat transfer fluid (synthetic oil or something along those lines) that would collect sunlight, the oil gets hot, hot makes water get much much bigger, and suddenly the compressor end of a jet engine is taken care of.

In other words, using sunlight to boil water to run a steam turbine.
Ie This: http://en.wikipedia.org/wiki/Parabolic_trough

The excess steam after being run through the turbine could either go through more stages to help extract just a little bit more of the energy stored in it, or it could pass through a condenser of sorts.

Originally the idea was to shove a bunch of troughs on the roof of a house and use this to provide some of the electricity for the home. Further more, a condenser could allow the excess heat left in the steam to heat hot water for the home, as well as the home itself.

After that massive brain storm... I started thinking about building this thing (or atleast a model of it just to prove to myself it was feasible). The parabolic trough isn't terribly hard to build, the boiler isn't either -- Water Tube boiler, replacing flames with hot oil from the trough.

But the turbine, that's a whole other story....

Impulse turbines such as a Pelton turbine (while designed for water) are relatively simple, bunch of buckets strapped to a wheel.

Reaction turbines are much less so. But the same problems remains. How on earth with manual tools (such as a sieg x1, and a taig lathe) does one go about building a turbine, impulse or no? Obviously we're talking small here, say wheel diameters with blades not exceeding 3".

I had thought about turning a blank cylinder down and then using a slitting saw and an indexer to cut out slots in the blank. Then blades could be formed using a press/die setup (arbor press, or even bench top vise). But how to keep them attached to the wheel? Or even in the case of a Pelton turbine, keep the buckets on the wheel?

Furthermore, reaction turbines are built in two parts, the outer half, the Cask, houses stationary blades which curve the incoming fluid stream in to the next set of blades. I'm not even sure where to begin here. Complementary turbine wheels with one somehow held stationary could work, such as a collar that attaches to the housing. However again, how are the blades attached to the wheel? And how are they attached so that their centrifugal force doesn't have them flying off in to the casing?

With all these base machining questions, there are still questions as far as material requirements. Steam isn't going to be ungodly hot as opposed to the innards of a jet engine, but, is it too hot for aluminum? Most likely creep and the high rotational energy will be enough to fling aluminum apart, so steel then becomes required - perhaps.

Steel is pretty much outside my patience as far as milling goes, the poor x1's rigidity is dreadfully apparent trying to mill steel, it can do it, but the feeds are tiny as to minimize stress on the machine and thus any serious milling work is going to take hours - days - weeks to finish in steel. (Hence the forming idea further up)

Anyways after reading that wall of text, I'd like to ask for help to try and invent feasible ways of producing turbine blades and wheels at home. Preferably without access to a CNC machine - The Turbinia didn't need one... And hopefuly without casting anything, steel is outside most home foundries anyways, but even aluminum investment casting would require some finishing...

The blades don't need to be absolutely perfect nor 3d airfoils of exact optimal shape, to be honest just curved chunks of metal that are all curved the same way to the same length

Certainly more issues will pile up and in regards to my original idea above, the system itself has more issues. However, lets try to come up with some ways someone at home could build a steam turbine.

For those who made it through the massive wall above, don't worry I'm never this long winded, just been thinking about this for awhile....

[Gizmot]

why dont you form sheet metal into shape using a jig (like aircraft panels) then weld parts to a core?

[Geof]

Have you bothered Googling 'Parson turbine'.

The first steam turbine was built back in 1890 or thereabouts long before CNC. You should do a bit of research and find out how it was made.

[DeusExMachina]

Originally Posted by Geof Have you bothered Googling 'Parson turbine'. The first steam turbine was built back in 1890 or thereabouts long before CNC. You should do a bit of research and find out how it was made.

Actually I have looked it up, actual construction techniques are a little sketchy (Parsons is who built the Turbinia's turbine). I found a rather interesting article written by Parson: http://www.history.rochester.edu/ste...ons/part1.html

That articles also contains some images and some hints on how it was constructed primarily:

Originally Posted by Parson The turbine teeth or blades were like a cog wheel, set at an angle and sharpened at the front edges, and the guide blades were similar. These are shown in Fig. 6 on the next page. Gradually the form of the blades was improved as a result of experiments and some of these are shown on page 10. Curved blades with thickened backs were introduced. The blades were cut off to length from brass, hard rolled and drown to the required section, and inserted into a groove with distance pieces between and caulked up tightly.

- Would like to know what he meant by caulked

@gizmot, that's a good idea, I don't know if the weld would cause any damage to the rotor or anything but its simple and would get get the job done.

[RICHARD ZASTROW]

You might also look into the "Christmas Tree" blade root and the much simpler dovetail roots. They attach the blades to either the rotor or stator matching slot/groove.

Dick Z

[Geof]

Originally Posted by DeusExMachina ..... Would like to know what he meant by caulked....

Swaged, peened, riveted (?).

The edges of the groove where bashed over to retain the blades.

I think it is a case of machining grooves in the face of a disk of and having blades that had something of a dovetail on the end. The dovetail was stood in the groove and then the edges of the groove swaged over to retain the dovetail.

Parson's turbine was Radial flow not Axial flow and was a low speed (for turbines) low pressure turbine.

Actually the turbine side of your idea would be practical in terms of DIY; after all Parson was a DIYer when he invented and built it, he didn't have a bunch of CNC Jobbing shops available to make all the parts, just highly skilled craftsmen.

The aspect where your idea falls down is the parabolic collector trough using oil as a heat tramsfer medium. Semi-focusing collectors do not generate high enough temperatures to get acceptable steam temperatures and the maximum temperature availble at the water will be even lower than at the collector because of losses in the heat transfer via the oil.

You should find some old copies of Mother Earth News a lot of things like this where looked at back in the 1970s.

Large scale focussing solar collectors can generate electricity with an overall efficiency greater than photovoltaic panels. But they have to be large as in hundreds of square yards of collector surface and they are expensive. I doubt that they are at all practical for a home based system. You need to consider that the solar energy available at noon on a clear day is about 1000watts per square yard so if you can get an overall efficiency of 20% you can generate 200watts. Figure out the size of focussing collector you would need bolted to the roof of your house and imagine a strong wind coming by.