Willis Heating System

[JohnW2]

Mine just has the sprayed on insulation but it doesn't change the fact the change only occurs when used earlier.

Fair enough but, as has been said, that doesn't alter the situation about 'mixing' (or the lack of it .... .. rough calculations assuming 'perfect' insulation (zero heat loss to environment) ...

.... most nights, the (3 kW) immersion is powered (under thermostatic control) for a total of between 1 and 1.5 hours. Assuming that water is being heated from, say, 18°C to 62°C, that implies that I am using around 58 - 88 litre of hot water per day, so that is the amount of cold water which would exist at the bottom of the cylinder at the end of the day (just before heating re-started).

If that cold water (assume 18°C) had fully mixed with the remaining heated water (assumed 62°C) then the water available from the cylinder at the end of the day would be between 34.5°C and 43.8°C - dramatically lower than I actually experience. I therefore conclude that minimal mixing (or any other process of raising the temp at the bottom of the cylinder) is occurring

 

[robinbanks]

So, no matter what you may regard as 'obvious', there clearly is no significant 'mixing' (or heat transfer by conduction) in my cylinder

There isn't in anyone's. Warm water is very good at sitting on top of cold water and not being cooled by it/not warming it up

Willis doesn't change that in the slightest; water can stratify and sit around not mixing or conducting at the hot/cold boundary very well


Let's present a "all figures for illustration purposes only" time graph of a tank with your upper immersion and a willis. Let's assume the target temp is the same, the power is the same, the time slots are the same:

Both yours and the Willis end up with, after 50 minutes, similar profiles. This is roughly roughly, not an exact depiction of the molecular level and the exact timings. It's getting the point across of how the tank heats a volume. Blue is 26 degrees, purple is 32 degrees, pink 38, hot magenta 44 and red 50, for example (just in case you didn't get the fade from blue to red, through pink being related to a water temperature)

At any point in the heating period a person could have a decent (but short, getting longer) shower with the Willis because of how it heats the tank
You'd have to wait 50 mins for the immersion, then have a decent shower (let's assume the shower is intolerable at anything less than red temp)

Now, you may insist on a 25 minute long shower, and you need the red water volume presented at the 50 minute mark to achieve this, so whether you have a Willis or an immersion, you wait the 50 minutes, to get the reddest volume, and that is the exact and perfect volume for your shower. It makes no odds to your whether you have a willis or an immersion

Your wife is happy to have a 5 minute shower, but demands it be red temp too; she can get away with hopping into the shower after 10 minutes with a willis; there is enough red water to give her what she wants. With your immersion she must still wait 50 minutes, heat up 5x more water than she needs and if it is unused it's a waste

 

[JohnW2]

There isn't in anyone's. Warm water is very good at sitting on top of cold water and not being cooled by it/not warming it up

Exactly, despite what you wrote about your hypothetical water inlet that would use a jet of mains-pressure water as a 'stirrer'

Willis doesn't change that in the slightest; water can stratify and sit around not mixing or conducting at the hot/cold boundary very well

Again, exactly.

Let's present a "all figures for illustration purposes only" time graph of a tank with your upper immersion and a willis. Let's assume the target temp is the same, the power is the same, the time slots are the same:
Both yours and the Willis end up with, after 50 minutes, similar profiles. .... At any point in the heating period a person could have a decent (but short, getting longer) shower with the Willis because of how it heats the tank ... You'd have to wait 50 mins for the immersion, then have a decent shower (let's assume the shower is intolerable at anything less than red temp)

Sure, that's true IF one starts with a cylinder completely full of cold water, then switches on the heater and wants only a very small amount of hot water (at 'red' temp). If one wants more than a very small amount of water, with the Willis one will have to wait, for anything up to 50 mins with your figures.

In contrast ... in practice, with immersions and a very well-insulated cylinder like mine, that situation will very rarely arise. There will nearly always be as much more-or-less 'red temperature' water sitting at the top of the cylinder as a Willis would take 50 minutes to create, available for 'immediate use' (no wait at all).

 

[Harry Bloomfield]

In contrast ... in practice, with immersions and a very well-insulated cylinder like mine, that situation will very rarely arise. There will nearly always be as much more-or-less 'red temperature' water sitting at the top of the cylinder as a Willis would take 50 minutes to create, available for 'immediate use' (no wait at all).

So, you are agreeing - no real difference between your two level immersion system, and a Willis, when both have a full charge?

 

[SUNRAY]

There isn't in anyone's. Warm water is very good at sitting on top of cold water and not being cooled by it/not warming it up

I believe that is also not as clearcut as you describe, the calorimeter experiment at school with a row of thermometers at different depths showed an increase in temp when hot oil was floated on the top

After 50 years I'm unable to provide any detail but I think the curves are representative.
the thermometer virtually at the junction with the hot oil rose rapidly and the lowest was still rising when the top was fallingI think there were around 6 thermometers and reading were taken over about 1 hour.


We did a similar experiment with a heater at half way which showed a slower rise at the top but of course no fall and a little quicker rise at the bottom.

The calorimeter was around 100 to 150mm tall. Whether these were representative of a domestic hot water cylinder I truly have no idea.

 

[SUNRAY]

In contrast ... in practice, with immersions and a very well-insulated cylinder like mine, that situation will very rarely arise. There will nearly always be as much more-or-less 'red temperature' water sitting at the top of the cylinder as a Willis would take 50 minutes to create, available for 'immediate use' (no wait at all).

Again in commercial systems the external heater(s) can provide some instant hot water which an immersion cannot.
My gut feeling is the water at the very top of the storage vessel is hot much quicker with external heater but after the initial input I don't know

 

[Harry Bloomfield]

Let's present a "all figures for illustration purposes only" time graph of a tank with your upper immersion and a willis. Let's assume the target temp is the same, the power is the same, the time slots are the same:

Your diagrams, neatly illustrate the difference between the Willis system, and a simple immersion heater in cylinder.

 

[JohnW2]

So, you are agreeing - no real difference between your two level immersion system, and a Willis, when both have a full charge?

Indeed - I've been saying all along that I could not see what appreciable difference./advantage a Willis system would offer over what I already have.

 

[Harry Bloomfield]

Indeed - I've been saying all along that I could not see what appreciable difference./advantage a Willis system would offer over what I already have.

For your limited use, of the capacity of your cylinder, boosted each night, no, but others do make good use of their hot water, and need a fast replenishment, of a small quantity - that's where the Willis shines.

 

[JohnW2]

I believe that is also not as clearcut as you describe, the calorimeter experiment at school with a row of thermometers at different depths showed an increase in temp when hot oil was floated on the top

As discussed, that can't happen by convection unless something 'mixes' things - so if it did happen it would presumably have to be by conduction - which I think is very slow/slight.

In any event, we have the experience of the hot water at the top of my DHW cylinder to consider, which is showing little fall in temp (over relatively long periods) whilst 'floating' on top of much colder water (with no heating). If heat was being lost to lower parts of the cylinder, there would be an appreciable fall in temp at the top.

 

[JohnW2]

For your limited use, of the capacity of your cylinder, boosted each night, no, but others do make good use of their hot water

What do you mean by 'making good use' of my hot water? As I've illustrated, I appear to be using some 60-90 litres (roughly 43%-64% of cylinder capacity) per day - are you suggesting that is very low?

, and need a fast replenishment, of a small quantity - that's where the Willis shines.

An element close to the top of a cylinder will provide pretty fast 'replenishment'.

 

[SUNRAY]

There isn't in anyone's. Warm water is very good at sitting on top of cold water and not being cooled by it/not warming it up

Willis doesn't change that in the slightest; water can stratify and sit around not mixing or conducting at the hot/cold boundary very well


Let's present a "all figures for illustration purposes only" time graph of a tank with your upper immersion and a willis. Let's assume the target temp is the same, the power is the same, the time slots are the same:

View attachment 408579

Both yours and the Willis end up with, after 50 minutes, similar profiles. This is roughly roughly, not an exact depiction of the molecular level and the exact timings. It's getting the point across of how the tank heats a volume. Blue is 26 degrees, purple is 32 degrees, pink 38, hot magenta 44 and red 50, for example (just in case you didn't get the fade from blue to red, through pink being related to a water temperature)

At any point in the heating period a person could have a decent (but short, getting longer) shower with the Willis because of how it heats the tank
You'd have to wait 50 mins for the immersion, then have a decent shower (let's assume the shower is intolerable at anything less than red temp)

Now, you may insist on a 25 minute long shower, and you need the red water volume presented at the 50 minute mark to achieve this, so whether you have a Willis or an immersion, you wait the 50 minutes, to get the reddest volume, and that is the exact and perfect volume for your shower. It makes no odds to your whether you have a willis or an immersion

Your wife is happy to have a 5 minute shower, but demands it be red temp too; she can get away with hopping into the shower after 10 minutes with a willis; there is enough red water to give her what she wants. With your immersion she must still wait 50 minutes, heat up 5x more water than she needs and if it is unused it's a waste

Thanks, a good description, however i think there will be an amount of temp gradient during the process with an immersion heater rather that the single layer you show, my scrappy hint at an attempt showing the idea of varying temp - I gave up playing but I hope the idea shows with it, I'm convinced the copper cylinder feels gradual variation rather than a distinct change.

and the gradients move down until it's fairly constant at your 50 min with some conduction warming it further down.

 

[JohnW2]

Thanks, a good description, however i think there will be an amount of temp gradient during the process with an immersion heater rather that the single layer you show ....

Sure, that's inevitable (at least to some extent), isn't it? - but that doesn't detract from the general point robin was making.

 

[SUNRAY]

Sure, that's inevitable (at least to some extent), isn't it? - but that doesn't detract from the general point robin was making.

No I don't believe so other than full temp water is available before the 50mins

 

[JohnW2]

No I don't believe so other than full temp water is available before the 50mins

(assuming one started with a cylinder full of cold water,which one very rarely would) .... some pretty hot water, yes, but I'm not sure that it would be literally 'full temp', since, if there is any space at all above the immersion, heated water getting to the top of the tank (and hence available to be drawn off) would have to have passed through, and partially mixed with, at least some colder water on its way up.

However, as I keep saying, when one has a system like that one would not normally wait until the cylinder was totally full of cold water before turning the immersion on, so there would probably still be some 'full temp' water there ready to be drawn off immediately.