Generator for microhydroelectric power generation station?

[tsgrue]

.

People That Know Electricity,

I am working on the design for a new microhydroelectric power generation station. The station will be located at an existing pond in a small, urban watershed. The pond has an existing dam with 9+/- feet of available head. The base streamflow is 3+/- cfs and the maximum flow that will be utilized is 23+/- cfs. This translates to roughly 1.5kW base power potential and 14kW maximum power potential.

The proposed system schematic is as follows:

Reservoir --> Waterwheel --> Generator --> Inverter --> Grid

The most suitable power converter in terms of overall performance (cost, efficiency, maintenance) is a breastshot waterwheel (18+/- ft diameter). Breastshot waterwheels are most efficient when converting potential energy into 'mechanical rotational energy' (as opposed to kinetic energy to 'mechanical rotational energy') – in practical terms, this means the waterwheel is most efficient at slow rotational speeds. Ideally, I'd like to have a base waterwheel rotation of 1 RPM, which would roughly result in a maximum rotation speed of 7 RPM. Assuming I could achieve a 40:1 gear ratio, this would be a base generator shaft speed of 40RPM and a maximum generator shaft speed of 280 RPM. As this is an urban watershed, the flow (and thus power output) is highly variable. (I could go beyond the 23+/- cfs input, but I am limiting it to this due to various civil/mechanical reasons).

With all that, I have the following questions:

1) Is the "[DC] Generator --> [DC-to-AC] Inverter" portion of the system schematic what I should be looking at or is there a better system? I've seen various systems, but this seems to be the most straightforward, efficient, and (assuming I can find the generator) least expensive and most viable option.

2) Does anyone have a suggestion for the generator I should use? I've looked around, but haven't found much. I am guessing that I will need to find a high torque / low speed (multi-pole) DC motor used for mixing or grinding operations and run it as a generator.

I've searched the posts and haven't found anything covering this...

Any help is greatly appreciated!

.



tsgrue: site engineering, stormwater
management, landscape design, ecosystem
rehabilitation, mathematical simulation

http://hhwq.blogspot.com

 

[tsgrue]

.

Electric Wizards,

A few notes:

1) Properly designed and installed overshot, backshot, and breastshot waterwheels are very efficient, with 70% to 95% able to be realized over a wide power band. These are not worse or better than turbines, each appropriate on a site-by-site case-by-case basis. I am estimating an average of 75% efficiency at the shaft for system design purposes. As these are 'potential energy converters', the key to the better efficiencies is keeping the water input velocity low and minimizing fluid losses (especially turbulence losses at the entrance and exit).

2) I am fairly familiar with the crossflow and most other turbine types. The crossflow is an option, but for a variety of reasons - including aesthetics, visibility, and education - the breastshot waterwheel was chosen. The breastshot is also expected to be somewhat less expensive and more robust than the crossflow. Interestingly, the crossflow works partially in the same manner as the breastshot waterwheel, but with much more of the power conversion from kinetic and not potential energy in the crossflow.

3) The potential for pulsing is fairly minimal. With 60 buckets likely for the final design - which will be determined in part by the generator selected - the flow effect is relatively continuous.

4) The buckets are sized to accommodate the median and peak flows. It is possible to partition the buckets with dividers (similar to crossflow turbines) to maintain constant speed while torque increases, but - as mentioned before - that level of effort and cost is probably not feasible for this project.

5) I am familiar with the primary current German (around 3) and American (around 1-1/2, the 1/2 being wooden) manufacturers of waterwheels, as well as the 'historical' American manufacturers (mainly Fitz and Campbell) and have reviewed information available from them - especially the German. It's the generator for this site that has me stumped. There used to be various generators produced that would do the job for this scenario - especially from 1900 to 1950 - but these don't seem to be around anymore. It may be no coincidence that power converter speed increased significantly during that time along with mechanical system advances - also the era during which most waterwheels were phased out.

6) The easy answer seems to be speed the thing up, but I need higher efficiencies to make the combination of aesthetics, visibility, education, and life cycle cost work out. You know - everything!

As before – thanks very much!

.



tsgrue: site engineering, stormwater
management, landscape design, ecosystem
rehabilitation, mathematical simulation

http://hhwq.blogspot.com

 

[MikeHalloran]

Typical urban sites suffer a fair amount of vandalism.

Where copper is available, drug-fueled thieves will arrive. They will bring tools, and persistence, and LN2 if needed.
Plan on a total enclosure around whatever electrics you choose, and plan on welding it shut. Ordinary locks will not suffice.





Mike Halloran
Pembroke Pines, FL, USA

 

[Warpspeed]

I would probably use an automotive rear end with locked differential to support the weight of the water wheel, and provide the first increment of speed increase. That would be robust, simple low cost and pretty well sealed.

Compound sprocket chain drives would be practical up to perhaps a hundred rpm, they could run fairly loose in an oil bath with very little frictional loss. Fairly easy to make too.

The ever experimenting do it yourself wind power guys are now using home constructed alternators with neodymium magnets.
These can easily be fabricated as a large diameter non magnetic disc with the neodymium magnets (facing axially) studded around the periphery, then have a very large numbers of stationary field windings arranged on either side.

Many windings running in phase, are connected in series right around the disc, so that the turns count per pole need not be very high, but even at low rpm, the output voltage and frequency becomes very usable.
The wind guys find that doing it that way requires no gearing, and with air cored coils (!) there are no eddy current core losses, and it is far easier and cheaper to make.
All rather surprising if you have never seen this approach before, but it seems to work rather well.

 

[tsgrue]

.

Thanks to all who posted. The information has been very helpful. It looks like the system will employ an asynchronous generator.

Thanks again!

.



tsgrue: site engineering, stormwater
management, landscape design, ecosystem
rehabilitation, mathematical simulation

http://hhwq.blogspot.com

 

[starkopete]

This is for potteryshard and rasevskii, it took me a little while but I found it:

http://files.engineering.com/getfil...724-c30a2323f515&file=Inductor_Alternator.pdf

 

[potteryshard]

Yes... Exactly the book and page I was thinking of. The illustation on the left (if a permanent magnet is substituted for the field electromagnet) shows quite well the basic magnetic structure of what I had envisioned as a caliper-type alternator with a slow-speed direct drive non-magnetized rotor.

Although not an off-the-shelf device, it would seem to be mechically and conceptually simple enough to home build.

 

[Warpspeed]

This is pretty much what the wind guys are building.

They have the particular problem of "cogging" which may be much less troublesome with a water wheel.

What happens is, the permanent magnets align themselves with the iron pole pieces of the electromagnets, and the whole thing strongly resists initial turning.

The result is the alternator locks up stationary, and the initial break away torque can be high enough to prevent the wind turbine from starting up at all at very gentle wind speeds.

The drive torque also fluctuates massively as each iron pole piece and magnet pass each other when it does start up. That is the evil "cogging" effect.

The solution is to use AIR CORED COILS in your calipers. There is then zero cogging effect, and the alternator turns completely freely where there is no load current.

Now it might be thought that this massively reduces efficiency, and it certainly does from the size, packaging, and power density aspect.
But amazingly it has almost no effect on the efficiency of mechanical power in versus electrical power out.

In fact, the complete elimination of hysteresis and eddy current loss that always occurs in an iron magnetic path
is actually a bonus.

So what you end up with is an inconveniently large machine, of excellent overall energy efficiency. This is real fringe engineering, but it sure is interesting.

Here is a link to how a home constructor in Italy went about building a 1Kw machine. The air cored alternator pictures begin half way down page two:

http://www.thebackshed.com/windmill/articles/SpitsFire.asp

 

[potteryshard]

One way I thought of to reduce cogging was to build the rotor teeth and the poles at different pitches. The pole coils couldn't be series, but a full wave bridge rectifier is cheap and easy for each individual coil. Rather than one preferred position there would then seem to be a series of slight preferences, making it easier to start and less prone to vibration.

 

[LionelHutz]

The inductor alternator is an interesting idea. You'd want a relatively large air gap that gets closed to a relatively complete magnetic path so the flux density varies as much as possible. I don't think this would work very well with a large air gap.


Here is a good site showing different wind power construction teqniques. You can see the general evolution over time if you go through the links in the boxes from the bottom to top, with the newest projects being 20', 5kW turbines.

http://www.otherpower.com/otherpower_wind.shtml

Gearing up to a few 100 rpm would be a lot easier than gearing up to 1800rpm or 8000rpm.

 

[Warpspeed]

Yup,
There have been a great many amateurs all over the world working on these low rpm axial flux disc alternators, and even the crudest built examples seem to work amazingly well.
This definitely seems to be the way to go.

An internet search will give you plenty of ideas of how to do this, and your goal of 1.5Kw is right in the ball park.