Using low pressure water flow to generate low flow enhanced pressure. 2

[asimpson]

I have a project where a water supply of less than 0.1 bar and flow rate 5-10 L/min. I would like to use this water flow to produce a 2 bar stream of water at 0.2 L/min. Effectively intensify pressure using the kinetic energy of the flowing water. I have seen systems like hydraulic ram pumps that use river flow to pump small amount of water well above level or river.

http://en.wikipedia.org/wiki/Hydraulic_ram_pump

However these are large and pulse at low frequency (0.5 or less Hz).

Anybody come across a miniature version that pulses at 10 Hz or higher to produce a smoother flow?

I have the system working with a miniature electric pump the performs well but I would like to eliminate the need for external power source.

Application is domestic water.

Thanks

 

[JStephen]

The water ram uses energy from a lot of water to get a little of it up higher. Should be do-able. Use an accumulator tank at the top end to even the flow out?

 

[asimpson]

Thanks everyone.

All the ramp pumps on the market seem a bit "agricultural", no offense intended. I need something a bit smaller. Guess I'll have to do some work and try to build one to my specification. Should be interesting.

 

[rb1957]

what's wrong with 77's suggestion of a Davey #2 ? seems to meet the flow and pressure req'd ??

 

[hydtools]

I like zdas04's water wheel driving a pump.

Ted

 

[btrueblood]

Have to agree, 77's link is a pretty close match. There are two questions - 1) will there be sufficient head truly available to make the ram pump cycle - it wants a few feet at minimum, and simpson's OP specified a head of only a couple feet; later he says 6 ft. That's about a 2:1 or more difference, which will matter. Same thing matters for any other scheme (turbine driving a pump, boost cylinder pump) - the higher the available supply head pressure, the more energy and higher the energy "quality" in the source flow. The second question will be whether the top pressure and flow delivery will be sufficient. Really it boils down to looking at the flow vs. head curves/tables given for the #2 Davy pump and seeing if it can be made to work. I'm confident that some flow can be made available with some scheme at the source pressure stated, but I'm not at all confident that the specified flow can be achieved; as somebody else noted the system efficiency would need to be about 40 to 50% to get there, which will call for some pretty sharp pencil work. I think the whole system has a better chance for every foot that can be added to the source head (which should be a "duh" statement, right?).

and Dave, I hate to nitpick especially somebody like you who is pretty sharp all around, but this statement is incomplete:

"You can get the velocity you want/need by picking the right pipe. "

- I would add: "but only if you have the source pressure (head) to drive that velocity", which was why I kept bugging asimpson to spec his source pressure. I know you stated you missed the low source pressure in a later post, but that comment just bugged me, seemed like you were saying "duh" to my earlier post.

 

[asimpson]

Sorry I cannot be specific on details in terms of pressures and flows. I am investigating the principal and will see how close I can get to desired performance with a product or system I can get off the shelf. It is an R&D project and I do not have a definite application.

I misread data on Davy #2 and I will look back at it. However so far it is the only system available close to what I require

 

[1gibson]

Waterwheel driving a centrifugal pump would probably need some sort of gear drive for speed increase. Then, it would be tricky to size. If flow varies between 5-10 l/min, then the waterwheel speed would vary also. So the pump speed varies, and the pressure varies to the ^2 of the pump speed difference.

I think most feasible is turbine (or waterwheel) driven generator used to charge a battery, which drives an electric motor centrifugal pump.

Low end of supply flow and not enough power? Add another turbine/waterwheel to the system, or other "free" power source (solar.) Or if it is a temporary reduction in flow, just ride it out on battery power.

Don't know how consistent you need the high pressure flow to be, but I think you will have to go electrical here.

 

[btrueblood]

Gibson, shame on you! This is the Mechy forum, take your sparky heresy elsewhere! Ram pump and a constant-pressure accumulator(*), no need for messy chemical batteries and dangerous electricity.

(*) A standard hydraulic cylinder with a big honkin' weight on the piston rod.

 

[TheTick]

Low end of supply flow and not enough power? Add another turbine/waterwheel to the system

Sparkies don't believe in conservation of energy. Strange religion, EE is.

 

[btrueblood]

Good catch, Tick.

more sites regarding ram pumps:

[URL unfurl="true"]http://journeytoforever.org/at_waterpump.html[/url] - lots of how-to type info.

This one says they have a pump that operates on as little as 2 ft. of fall, might be worth a closer look:

[URL unfurl="true"]http://www.gravi-chek.com/html/about.html[/url]

 

[KENAT]

If a turbine/water wheel is deemed appropriate for generating mechanical power from the fluid flow, why limit to only considering rotary pumps?

Pretty sure ways to convert rotary to linear motion are well understood so why not a simple reciprocating positive displacement pump or some such.

Again an accumulator or simple header tank system or similar could be incorporated as required to smooth flow.

Or if you want to stick to rotary motion maybe an archemdies screw or some such but then you'd probably be looking at gearing again.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[Compositepro]

There are a number of solutions possible. Water turbine driving a gear pump would work. Using a centrifugal pump would greatly narrow the window of operating conditions. I think the most robust and efficient would be using the piston principle behind the pressure boosters mentioned above.
Friction is almost non-existent with diaphragm sealed pistons.
Combine the principle of an air operated double diaphragm pump (AODD using water pressure rather than air pressure) with pressure booster. Another way to look at it is that a hydraulic piston motor with a large piston area drives a hydraulic piston pump with a smaller piston, and it is double acting so pumping occurs while stroking both ways.

 

[TheTick]

It would be interesting to find a way to do this without moving parts. Wonder if it's possible.

Not talking about "magic", overunity, or perpetual motion. As I mentioned earlier, acquiring energy from part of the stream to deliver smaller part to a higher altidude.

 

[asimpson]

A hydraulic ram pump has virtually no moving parts. Apart from flow actuated valves. No magic or over-unity. Problem is they are too big for me in terms of physical size.

 

[btrueblood]

Tick,

Maybe, using magneto-fluid-dynamic pumps/turbines? But I coubt there is enough efficiency in current MFD tech to make this particular application work.

 

[GregLocock]

A hydraulic ram has one thumping (literally) great moving part.

As a black box you seem to be wanting to convert a large supply of hugh entropy fluid into a small supply of low entropy fluid and the remainder at even higher entropy.

At first sight this is perilously close to a Maxwell's Demon situation, but given that a hydraulic ram, or a generator/pump, actually achieves this, obviously 2LE is not broken. The reason is that you do have reservoirs at different entropies, whereas the demon is more concerned with creating reservoirs from a uniform mush.

Perhaps one way you could do it with no moving parts is with a series of hydraulic jumps and overflows, to build gravitational PE.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?