Centrifugal pumping?

[TomG]

This may well be old hat (hence posted in "Newbies"), but has anyone tried hooking a wind turbine to a centrifugal pump? I know that wind-powered pumping is common, but all the versions I've been able to find use positive displacement pumps. Their power curve is a horrible match for wind power, which is why they mostly require an American Multiblade type. Centrifugal pumps have a cubic power curve, very low starting torque, and are mechanically simple. You can also put multiple stages onto a single shaft to increase the compression.

They seem like an obvious match to me, maybe as a heat pump?

Most centrifugal pumps are designed for high RPM use, because they're powered by electric or fossil fuel motors, but low RPM designs aren't impossible, if we wanted them.

Just a thought; wondering if this has been done, or if there's a good reason why not

[Flux]

Yes you can use a centrifugal pump and there are benefits from the cube law match over the upper part of the range.

There are also issues that make it less than ideal under other conditions.

Firstly there is a minimum speed at which a centrifugal pump will overcome the head and any speed below this it will deliver nothing. This probably prevents you making effective use of the very light winds.

Secondly the efficiency is well below that of a positive displacement pump and again this impacts badly on the low wind performance.

They are best suited to high flow at low head and if you are only moving water a few feet the match would be very good.

For higher head things may still be ok on good windy days but the prospects of doing much in low winds are not so good.

Each individual case would need to be treated on its own merits.

It looks as though you are not proposing to use an intermediate electrical stage so you have a few interesting mechanical problems to tackle.

Flux

[TomG]

Re: overcoming the head:

Yes, much as you have to overcome the battery bank voltage before you can start charging it. I imagine the same is true of a positive-displacement pump: if you don't have the wind to move the piston against the force of the head, you can't complete a single stroke. But the length (or rather shortness) of the cam or crank is used to trade volume for force, so even a light wind can keep pumping.

I guess the equivalent for a centrifugal pump would be using a really small one, designed for high pressure, low volume. It would never shift huge volumes, even in a high wind, but the bladeset could keep turning it even in low winds, raising a small volume to the required pressure.

But a continuous-use application like a heat pump doesn't really have a head.

It looks as though you are not proposing to use an intermediate electrical stage so you have a few interesting mechanical problems to tackle.

Heh! I've just had a vision of a 'mill with two hosepipes running up to it, twisted around the tower

I was imagining it used with a VAWT, so I hadn't even considered the rotating linkage to connect hoses to a HAWT.

And yes, no intermediate electrical stage. Much cheaper with no magnets or copper, and it would surely be reliable - heating and/or cooling, with only one moving part (the bladeset right through to the compressor stages, all on one shaft).

[joestue]

I'm not saying this isn't a bad idea, but unless you have your own research department, its not going to work very well, without some kind of control over the system.

You are essentially forced to cast your own pump in this situation, i don't see too many 200 rpm centrifugal pumps, under a few thousand horsepower.

IIRC some dam had a direct connection between a pump and a Kaplan turbine, but i don't remember which one.

in any case, you have a pump that can't prime itself either, but if you could add a small positive displacement pump and a check valve, this could work.

[TomG]

Yeah, I doubt you could get one off the shelf. But then that's true of a lot of the stuff discussed on this board

I've played with a couple of pump design calculators, and it does seem possible to get super-low-RPM pumps without them getting too huge. Multiple small stages seems to do it. But as you say, it would need research, because it's not exactly a commonly used corner of the design envelope.

(Also, not being self-priming isn't a problem for heat pumps)

[joestue]

It isn't hard, we would just need to change the relative sizes of the inlet and outlet ports, and maybe use multi stage.

if for instance i take an 8 inch diameter impeller, but cast it with a 1 inch diameter opening, as opposed to a 3 inch diameter inlet port, things could get interesting,

IMO the operational issues are bigger, unless this is used to push water up grade, as opposed to pump it out of the ground, in which case the pump is on the ground, not up on the tower.

[wooferhound]

Here are some other ways to do it . . .

wind powered water pump works with compressed air

http://www.fieldlines.com/story/2005/11/19/195317/48

discussion about different ways to pump water

http://www.fieldlines.com/story/2007/1/13/54932/4786

You could let the water pump itself like in these videos

http://www.fieldlines.com/story/2007/9/11/23258/3210

related to these stories

http://www.fieldlines.com/story/2007/1/20/15324/6627

http://www.fieldlines.com/story/2007/9/2/173915/8438

http://www.fieldlines.com/story/2007/9/11/235538/617