How do submersible pumps work?

[Harry Bloomfield]

Yes but, as I said, given the total symmetry of everything (other than the outlet), only about 8% of the water flung out by the impeller will be "in the general direction of the outlet".,

It doesn't need to hit the 8%, it only needs to cause enough pressure differential for the water to want to escape via that 8%. As said, there are much more capable pump designs, capable of higher pressure differentials if the fluid might show some reluctance (due to back pressure) to escape via the outlet.

 

[SUNRAY]

Yes but, as I said, given the total symmetry of everything (other than the outlet), only about 8% of the water flung out by the impeller will be "in the general direction of the outlet".,
As I also said, it seems that (if the pump is in pretty shallow water) it manages to keep a pressure differential of about 0.7 bar (around 10 lb/square inch), which is far from insignificant, between the inlet and outlet orifices, about 50 mm apart.

This is very difficult! I obviously know that 'it works', but everything in my mind and 'intuition' tells me that it shouldn't (at least, nothing like as well as it does work), given everything I've described!

I suppose it must be a bit like currents in the sea. Relatively small and circumscribed high velocity currents can, presumably, result in quite marked pressure differentials between places very close together within what appears to be 'the big wide ocean'!

We're talking about a chamber about 14 cm diameter and about 7cm long, with two holes, 30mm and 34mm in diameter respectively, about 7cm apart. If one raised the pressure within any part of that chamber by any means other than 'flinging water about', I would not really expect any significant amount of water to come in through one of those holes and out through the other.

Kind Regards, John

But don't forget the water is being thrown away from the blue low ressure inlet and creates the high pressure area in pink and being simultaneousely rotated, outlet is usually formed into a scoop, added to this the cavity is often not round to divert the water out of the void. the later may be verrrrry subtle or non existant
My very crude interpretation:

Centridugal force forces the water into the void bottom left and rotation assists its movement anticlockwise and hence out of the exhaust.

Please note this is not within my field of expertise and these are my own observations and theories, I'm totally happy to be put right by those who work on/with pumps on a regular basis.

 

[JohnW2]

It doesn't need to hit the 8%, it only needs to cause enough pressure differential for the water to want to escape via that 8%. As said, there are much more capable pump designs, capable of higher pressure differentials if the fluid might show some reluctance (due to back pressure) to escape via the outlet.

Yes, I undertsand that, but it does not alter the fact that it is managing to maintain a substantial pressure differential between the vicinity of the outlet and the inlet hole around 40mm below the outlet - particular given that, at least in my case, the inlet offers a much easier 'escape route' (hence requiring less pressure) for the water than does the outlet. To pump water out of my pump's outlet requires raising it about 3m and overcoming the resistance to flow of about 10m of 1" tubing. To get water out of the inlet only requires 'overcoming' of a water head of about 0.5m, with no piping.

Kind Regards, John

 

[JohnW2]

But don't forget the water is being thrown away from the blue low ressure inlet and creates the high pressure area in pink and being simultaneousely rotated, outlet is usually formed into a scoop, added to this the cavity is often not round to divert the water out of the void. the later may be verrrrry subtle or non existant

If it were like you diagram, it would be easier to understand. However, as I've said, there is certainly no 'scoop' and if the cavity is not round (it is according to my calipers) or the impeller is tilted (ditto), or the impeller not central (ditto) it must be extremely 'subtle'!

As I've implied before, if the black line in your diagram were the walls of an 'enclosure', then it would all make sense - but that's not what I have.

I'm also amazed by how small the 'vanes' of the impeller are (as I said, just 7mm) - per the photo in post #41.

Kind Regards, John

 

[Harry Bloomfield]

Yes, I undertsand that, but it does not alter the fact that it is managing to maintain a substantial pressure differential between the vicinity of the outlet and the inlet hole around 40mm below the outlet - particular given that, at least in my case, the inlet offers a much easier 'escape route' (hence requiring less pressure) for the water than does the outlet. To pump water out of my pump's outlet requires raising it about 3m and overcoming the resistance to flow of about 10m of 1" tubing. To get water out of the inlet only requires 'overcoming' of a water head of about 0.5m, with no piping.

Kind Regards, John

Just the action of centrifugal force on a relatively heavy liquid.

 

[JohnW2]

Just the action of centrifugal force on a relatively heavy liquid.

I fear that I remain both amazed and confused!

Is not the only thing that the impeller, and the centrifugal force it generates, does is to increase the pressure in the vicinity of the outlet orifice?

If, with the impeller not rotating, I increased the pressure at that place by any other means, the pressure would increase by a similar amount throughout the chamber, including in the vicinity of the inlet orifice, a couple of inches away from the outlet one, wouldn't it?

Kind Regards, John

 

[JohnW2]

Thinking aloud ... I'm coming to think that the explanation probably relates to the reduced pressure created below the centre of the impeller, extending down towards/to the 'input hole', because ...

... if, with everything else unchanged (including connections to the output port), I removed the impeller, and then created an extra hole in the casing opposite the outlet port, if I then 'pumped' a jet of water in through that hole (e.g. from a hose connected to mains water), then I would expect that most of the water would exit the chamber through its 'inlet' hole, since that represented the 'lower impedance' exit from the chamber - and I would expect that to largely remain the case even if my jet of water was directed straight at the outlet port (thereby exactly emulating the effect of the impeller in 'flinging water out' towards that exit).

If that is right, the the only difference from the actual situation (with the rotor present and rotating) would seem to be that region of reduced pressure below the impeller and extending down towards the 'input' hole.

Am I perhaps 'getting a bit warmer'?? (even though I find it difficult to understand how even that is 'enough' to achieve the performance we see!)

Kind Regards, John

 

[Harry Bloomfield]

Yes, warmer. You do know its the pressure of the atmosphere which to a large extent, keeps the bottom of the pump filled with water, don't you?

I suppose you also know that you can only suck water up a pipe to a pump which is around 20 to 30 feet above the water level depending on the pump type? The absolute theoretical maximum from memory is around 36 feet, if a pump is able to create a perfect vacuum in the pipe. The pump would have to be a 'self-priming' type. Your pump is not a self priming type, its impeller needs to be submerged under the water level.

 

[JohnW2]

Yes, warmer. You do know its the pressure of the atmosphere which to a large extent, keeps the bottom of the pump filled with water, don't you? ... I suppose you also know that you can only suck water up a pipe to a pump which is around 20 to 30 feet above the water level depending on the pump type?

That is true of a 'lift pump', operating in the atmosphere. You're talking about (giving my age away) 'O-Level Physics', and I've been educated to a much higher level than that (in a wide range of disciplines) In fact, if I recall correctly, we did 'lift pumps' and 'force pumps' in the first year of my O-Level course, so when I was 11.

Anyway, getting back on topic, what you say is not true of a submersed pump. One such as we have been discussing would work the same even if atmospheric pressure were zero. Atmospheric pressure, when present, will simply add, equally, to the pressure which is trying to push water into the inlet of the pump and the pressure (if any) trying to stop water being pushed out of the outlet of the pump, with no net effect.

With a submersed pump, whether or not there is any atmospheric pressure, the hydrostatic pressure of the water above the pump (or, more precisely, the difference between that and the pressure within the pump) will "keep the bottom of the pump filled with water". If the pump is, say, 5 metres below the surface of the water, the hydrostatic pressure will be about 0.5 bar - so the total pressure trying to push water into the pump will be 0.5 bar without an atmosphere, or 1.5 bar with a standard atmosphere. However, if the outlet were connected to 'the atmosphere' and the pump not running, the pressure within the pumping chamber would be zero or 1 bar respectively, and in both cases the net pressure pushing water into the pump would be 0.5 bar (i.e. the hydrostatic pressure)

Furthermore, since (even though there are no valves) we are essentially talking about a 'force pump', rather than a 'lift pump' (i.e. it is not atmospheric pressure which is lifting the water), I don't see that (unlike the case with lift pumps, which you described) there is any theoretical limit to how high the water could be lifted. A look at a fire engine should help you to believe that.

Your pump is not a self priming type, its impeller needs to be submerged under the water level.

It does, but this is a bit of a play on words, since any submerged pump will be "self-priming", even if atmospheric pressure is zero, for the reason described in the last paragraph - i.e. hydrostatic pressure will always be acting at the point of the (submerged) inlet, even if there is no atmospheric pressure. To be pedantic, if one applied air pressure to the outlet pipe that was equal to or greater than the hydrostatic pressure at the pump's inlet, the pump would not "self prime".

So, I don't think that I'm much further forward than I was last night. Atmospheric pressure (whether zero or anything else) is irrelevant, since it's effects in pushing water into the pump and stopping water leaving the pump will cancel. So we're back to the fact that it can only work if the impeller manages to create a pretty low pressure zone at the position of in input hole. Consider the below, which is roughly my situation - pump sumberged at about 0.5m and lifting water 3m. The atmospheric pressure (whatever, even zero) cancels, so the action of the impeller has to be to create a pressure differential of at least 2.5m of water (about 0.25 bar) between the inlet and outlet orifices, which are little more than a couple of inches apart. I've known that all along, but the thing I have found so hard to 'believe' (even though it must be true!) is that such a tiny impeller, with no discernible tilts, tapers or 'scoops', can actually achieve that!

Kind Regards, John

 

[EFLImpudence]

Have you noticed when you are doing the dishes - or ask the servants?

If you stir the surface of the water with your hand - with hardly any effort, the plates will rise.

 

[JohnW2]

Have you noticed when you are doing the dishes - or ask the servants? If you stir the surface of the water with your hand - with hardly any effort, the plates will rise.

No human servants, but we do have a DWM

Yes .. it's all about 'vortices' - even more dramatic examples are to be seen by looking at some of the phenomena which occur in the sea ('whirlpools etc.).

However, it's the scale/magnitude which astounds me. If, like I did in the photo in #41, you had held that tiny impeller in your hand, and then looked at the high velocity torrent of water flying half way across my garden that the pump can achieve, maybe you would share just a little of my surprise, if not also some of my (clearly misplaced!) 'incredulity'.

As I said, I had expected to see a (much larger) impeller in some sort of shaped 'enclosure', the 'output' of which went only to the output port, with no 'continuity' with the chamber from which it was pumping the water - OR, at the very least, some sort of crude valve on the inlet. I find it hard to believe that the former of those would not be (even) more efficient than what I'm looking at, but I suppose the advantage of 'not having to do that' (which is clearly the case), is that it considerably reduces the risk of things getting clogged up with debris.

Kind Regards, John

 

[Harry Bloomfield]

Anyway, getting back on topic, what you say is not true of a submersed pump. One such as we have been discussing would work the same even if atmospheric pressure were zero. Atmospheric pressure, when present, will simply add, equally, to the pressure which is trying to push water into the inlet of the pump and the pressure (if any) trying to stop water being pushed out of the outlet of the pump, with no net effect.

Yes, I was wrong, they fill via gravity.

 

[Harry Bloomfield]

No human servants, but we do have a DWM

Yes .. it's all about 'vortices' - even more dramatic examples are to be seen by looking at some of the phenomena which occur in the sea ('whirlpools etc.).

Dyson uses that principle in his ('bladeless') fans. You can also use it to pump fluids - My car has a saddle shaped tank, so the fuel settles on each side as it is consumed, but only has a pickup and pump from one side. Returned unused fuel at some pressure, goes first to the pickup-less side, were a vortex pump (no moving parts) sucks fuel out and transfers it to the side with the pickup.

 

[JohnW2]

Yes, I was wrong, they fill via gravity.

Yes - well, hydrostatic pressure, which is a consequence of gravity. I'd have to have a good think about what the situation would be in the absence of gravity (an issue which those involved with spacecraft must have to consider)!

I think you were simply thinking about the wrong sort of pumps - which is why I said that I think you were also wrong in suggesting that (with the sort of pumps we are discussing) there is a theoretical limit to how high the pump can 'lift' water.

Kind Regards, John

 

[Harry Bloomfield]

there is a theoretical limit to how high the pump can 'lift' water.

That is a matter of pressure created by the pump.