Willis Water Heating System

[JohnW2]

I don't know where the heating pipe enters, because I have never had one, but I'm puzzled as to why it would enter halfway up. I cannot see the purpose at all ...

It's actually the crux of what we are discussing - and is just like an immersion. If one creates/introduces heated water at some point within a cylinder, it is primarily the water in the cylinder above that point of creation/introduction which will get hot. Just as an immersion high up in a cylinder primarily only heats water at the top of the cylinder, so would the same to be true if the feed from a boiler is high up. If one wants to heat all of the water in the cylinder, then the ideal is to have the immersion, or point of connection of the boiler feed,, as low as possible in the cylinder.

However, in the case of the boiler, the two points of connection have to be an appreciable distance apart, so as to facilitate circulation by convection - so, with the 'out' pipe near the bottom, the 'in' one has to be appreciably higher, often near the middle as per the diagram I posted.

It may sound silly, but the equivalent of connecting the heat source (either back boiler of Willis heater) to the outlet pipe of the cylinder (i.e. 'above the top of the cylinder') would be to have an immersion hypothetically located 'above the top of the cylinder' (maybe in a hypothetical 'second cylinder' above the main one) - and, in either case, there is no apparent reason why the water within the cylinder should get hot.

You said 'conduction', it is not conduction, rather it is convection - a convection current in water carry the heat to the cylinder.

I meant what I wrote. I was saying that because, in the second diagram, heating of water in the cylinder by convection would not be possible, the only way in which the water in the cylinder could 'warm' at all would be by conduction (primarily through the material of the pipes and cylinder).

Your back boiler would none the less, collect the hotter water at the top of the cylinder.

Once water heated by the boiler got into the cylinder, it would, of course, rise to the top of the cylinder.

Kind Regards, John

 

[JohnW2]

You have acknowledged that with a simple loop the water would circulate so what difference does the overflow pipe make?

I think there may be some misunderstanding here. As I see it ..

If you have a cylinder full of hot water and you take a bit of pipe, filled with cold water and connect it to top and bottom of that cylinder, then I see no reason why there should be any 'circulation' through that pipe. Similarly, if you had two cylinders side-by-side, one filled with cold water and the other filled with hot water and connected them together with pipes at both top and bottom.

Kind Regards, John

 

[Harry Bloomfield]

If you have a cylinder full of hot water and you take a bit of pipe, filled with cold water and connect it to top and bottom of that cylinder, then I see no reason why there should be any 'circulation' through that pipe. Similarly, if you had two cylinders side-by-side, one filled with cold water and the other filled with hot water and connected them together with pipes at both top and bottom.

Unlike you - I fail to see any difference in the basic convection action, between a top connection and a lower connection to the cylinder. It would work with either layout, but with a top connection, you are heating water from the top down exactly as the Willis is suggested to work.

Can you explain why why you think there might be a difference in convection flow, between having a top hot flow in and one lower down?

 

[JohnW2]

Unlike you - I fail to see any difference in the basic convection action, between a top connection and a lower connection to the cylinder. It would work with either layout, but with a top connection, you are heating water from the top down exactly as the Willis is suggested to work. Can you explain why why you think there might be a difference in convection flow, between having a top hot flow in and one lower down?

Convection is all about temperature differentials and buoyancy ('floating/sinking').

• If a liquid cools, such that its temperature is lower (hence density higher) than liquid below it, it will 'sink' down through that warmer liquid below unless/until it reaches a point where the temperature (hence density) is the same as that of the liquid it is falling/sinking through; if it is cooler than all of the liquid below, it will sink 'right to the bottom'.
• If a liquid warms up, such that its temperature is higher (hence density lower) than liquid above it, it will 'float' up through that cooler liquid above unless/until it reaches a point where the temperature (hence density) is the same as that of the liquid it is rising/floating through; if it is warmer than all of the liquid below, it will rise/float 'right to the top'.

Do you agree with that?

If so then, if one connects heated water (from the external heater) to the very top of a cylinder (which is initially full of cold water), there could only ever be movement ('sinking'), by convection, of the 'water from the heater' into the top of the cylinder if the 'water from the heater' were cooler (hence denser) than the (cold) water at the top of the cylinder - and that obviously is not going to happen.

Kind Regards, John

 

[EFLImpudence]

It seems too obvious - but

you seem to be forgetting that if the heated water in the Willis heater rises (which you agree it will), then it has to be replaced by drawn in cold water from the cylinder.
Therefore the water in the cylinder has to be replaced by drawn in heated water at the top.

Unless we are all missing something.

 

[JohnW2]

It seems too obvious - but ... you seem to be forgetting that if the heated water in the Willis heater rises (which you agree it will), then it has to be replaced by drawn in cold water from the cylinder. ... Therefore the water in the cylinder has to be replaced by drawn in heated water at the top.

You seem to be taking this 'rising' out-of-context, almost as if a lump of water (you used the term "bubble") will in some way 'fly upwards', regardless of anything else (even if that 'rising' potentially created some sort of void that had to be replenished by 'drawing water' from somewhere). However, per what I just wrote, the actual situation of convection is ...

... water will rise (due to 'buoyancy'), through and totally within any water above it, if, and only if, the water immediately above it is at a lower temperature (hence higher density)

Conversely (and I suppose more relevant ...

... water will fall/sink (due to 'buoyancy'), through and totally within any water below it, if, and only if, the water immediately below it is at a higher temperature (hence lower density)

Convection only occurs within water 'which is already there' - so your ideas about a situation in which water would have to be 'drawn' from somewhere really are not appropriate.

I might add that even if there were a pump in the pipe between external heater and top of the cylinder, the 'pumped hot water' would primarily go up the expansion pipe and overflow into the storage tank (causing the cylinder to have to draw more water from the storage tank), not into the cylinder.

Kind Regards, John

 

[Harry Bloomfield]

Convection is all about temperature differentials and buoyancy ('floating/sinking').

• If a liquid cools, such that its temperature is lower (hence density higher) than liquid below it, it will 'sink' down through that warmer liquid below unless/until it reaches a point where the temperature (hence density) is the same as that of the liquid it is falling/sinking through; if it is cooler than all of the liquid below, it will sink 'right to the bottom'.
• If a liquid warms up, such that its temperature is higher (hence density lower) than liquid above it, it will 'float' up through that cooler liquid above unless/until it reaches a point where the temperature (hence density) is the same as that of the liquid it is rising/floating through; if it is warmer than all of the liquid below, it will rise/float 'right to the top'.

Do you agree with that?

Yes, not problem with that. However what you are missing is the fact that the Willis and the cylinder form a loop, a circuit. Water heated in the Willis will rise, thus drawing cold water in at the base of the Willis from the cylinder, so the heated water at the top, will begin to fill the upper part of the cylinder.

You obviously agree that your back boiler worked, despite it being slow to heat water, there is no difference in the convection process between that and the Willis.

 

[EFLImpudence]

Yes, and you, John, agreed that the water would circulate in a loop.

Isn't that why the back boiler system is called a gravity system?

 

[JohnW2]

Yes, not problem with that. However what you are missing is the fact that the Willis and the cylinder form a loop, a circuit. Water heated in the Willis will rise ...

You seem to be missing the same thing that I just pointed out to EFLI, namely that "water heated in the Willis will rise within any cooler water above it" ... hence will not ..

... thus drawing cold water in at the base of the Willis from the cylinder ...

You obviously agree that your back boiler worked, despite it being slow to heat water, there is no difference in the convection process between that and the Willis.

As I explained and illustrated before, there is a big difference. As I said, I do not believe that the back boiler arrangement would work if (as per the Willis arrangement we're discussing) the upper ('flow') feed from the boiler was connected to the top of the cylinder.

Kind Regards, John

 

[EFLImpudence]

You seem to be missing the same thing that I just pointed out to EFLI, namely that "water heated in the Willis will rise within any cooler water above it" ... hence will not ..

Ah, that is what I meant about the pipe being small enough to prevent convection, i.e. rising within the cooler water above it rather than being lighter and just trying to rise.

Like double glazing being the correct gap size to prevent convection of the air between the panes.

 

[EFLImpudence]

Again, why do you think the water would circulate if just a simple loop?

 

[JohnW2]

Yes, and you, John, agreed that the water would circulate in a loop.

Not in 'any loop'.

Considering (for convenience) a 'loop' consisting of two vertical 'arms', joined at top and bottom, then water will only 'circulate' (by convection) if the situation is such that the temp of one arm is higher at the bottom than at the top, and the temp of the other arm is higher at the top than the bottom. Water cannot 'circulate' unless it is travelling in the same direction ('as the pipe flies') in both arms.


Kind Regards, John

 

[EFLImpudence]

... but it does.

That 'we' do not understand how does not prevent it.

 

[JohnW2]

Ah, that is what I meant about the pipe being small enough to prevent convection, i.e. rising within the cooler water above it rather than being lighter and just trying to rise.

I think you are talking about a physical phenomenon that doesn't exist - there is no way in which water can 'rise' other than 'within the water which is there'.

I've never really thought about it but, as you imply, warmer water can only rise ('within the water which is there') if cooler water can fall ('within the water which is there'), which is what would usually happen, even in a pretty small pipe. However, if a situation arose in which the pipe were "small enough to prevent convection" (which, I guess, would, at the extreme, probably would be the case with a 'capillary tube'), I can but presume that the warmer water would just be 'stuck' at the bottom, because it couldn't 'rise' (in the way you seem to be envisaging) if cooler water could not correspondingly 'fall.

Don't forget that the 'rising' we are talking about is a consequence of buoyancy, due to differing densities, and can only take place 'within a mass of fluid'. As I keep saying, if there is something (e.g. your hypothesised inability of cooler water to 'fall' {within the mass of liquid} to compensate for a 'rise'), then that fluid cannot 'rise anyway', creating a void which has to be replenished by drawing water from somewhere.

Kind Regards, John

 

[JohnW2]

Again, why do you think the water would circulate if just a simple loop?

... but it does. That 'we' do not understand how does not prevent it.

OK, before I proceed, a couple of questions:

Do you agree that the Laws of Physics have to be obeyed within any part/arm of your 'loop' - specifically that, within any (vertical) part of the loop, water will only rise within that part if the temp is higher at the bottom than at the top, and will only fall if it is colder at the top than the bottom?

Secondly, am I right in thinking that the 'loop' you are considering consists of the two arms:
1... the ('upward arm') Willis and its pipes (which are attached to the top and bottom of the cylinder)
2... the ('downward arm') path through the cylinder ??

Kind Regards, John