Two immersion tanks setup, possible?

[webbygeeza]

I’m curious to know whether it is possible or efficient for that matter to use two electric only immersion heater tanks? I vision having one tank on top of another in the airing cupboard, the outlet from the top tank is connected to a pump which will add some pressure to the 6 hot taps within the building, the outlet from the bottom is fed into the top tank. Both tanks would still be fed from the cold water tank within the loft.

I want to achieve double hot water capacity in double quick to heat up.
If this is viable, would there be a technical name I could ask my plumber so we're on the same wavelength.

 

[oilman]

Why not just have a large tank with two heaters?

 

[webbygeeza]

Why not just have a large tank with two heaters?

Already have two tanks in my posession.

 

[PaulAH]

I want to achieve double hot water capacity in double quick to heat up.

You may achieve double water capacity but it won't be in double quick time (if that's what you meant). You have two elements but twice as much water to heat.

Curiously, you would get slightly more than double the amount of hot water because a single cylinder has warm/cool water at the bottom due to convection. This would be eliminated in the top tank with one above the other.

You would also have to consider whether a) the mains supply to the cupboard will cope with 2x3000w = 25 amps (assuming standard elements) and b) it can take the weight.

would there be a technical name I could ask my plumber so we're on the same wavelength

Yeah, two HW cylinders in series.

As for pumps, over to someone who knows.

Hey, I answered a question!

 

[webbygeeza]

Hey, I answered a question!

You certainly have and thank you very much!

 

[TexMex]

The main problem I can see here is that of the temperature gradient. In a single tank system, even when it has been heating for quite some time, and then left to stand, it will have piping hot water at the top, while the bottom is relatively tepid.

As you use the hot water, the cool water that comes in, enters at the bottom. For a while you will still be able to draw hot water, as this fresh water is not interacting with the main body of the tank (due to being held down by the temperature gradient). Once you have drawn a good deal of the hot water, you will, quite suddenly, notice a sharp drop in the temperature. This is due to the cool area finally reaching the top of the tank.

Now in your two tank system, you will have two separate tanks, each with their own internal temperature gradient. As you draw the hot water from the top one, the replacement water (which is theoretically hot), will, by necessity have to pass through the cool area of the top tank. In so doing, it will mix things up and even cool, to some degree itself. So by the time you have drawn as much water, as you'd expect from the top tank alone, the extra reservoir will not be as hot as you'd expect.

If you could generate sufficient convection between the top and bottom tank this would probably help the situation, but even then, you'd want the heating concentrated towards the bottom of the whole system.

 

[PaulAH]

Mmmm, interesting.

the replacement water (which is theoretically hot), will, by necessity have to pass through the cool area of the top tank.

But would there be a cool area in the top tank? I have a hunch that, if left long enough, the hot in the bot (which is not theoretically hot but plain old-fashioned hot) would filter upwards enthusiastically, even through a modest aperture, given the funnel shape of a cylinder top..

What's actually happening is the H2O molecules with the most energy will rise upwards and migrate through whatever pipe is connecting the two tanks, replacing or energising those with less, which will move down. A 22mm pipe would carry quite a lot of these and, after a reasonable period of time, would almost even out.

If you could generate sufficient convection between the top and bottom tank this would probably help the situation,

It would do more than that. Consider a pipe as wide as the cylinders themselves and you'd have only one temperature gradient in the entire system.

But I'm not a plumber - or a fluid dynamics engineer.

 

[TexMex]

Consider a pipe as wide as the cylinders themselves and you've have only one temperature gradient in the entire system..

Quite right, if webbygeeza was to connect the two tanks by cutting off the top of the bottom one, and the bottom of the top one and welding the two together, none of my concerns would be valid. In reality, I expect he's likely to be using something like a single length of 22mm with a couple of elbows thrown in to boot. That can create quite a bit of resistance to convection currents. In fact if the connection is not big enough, it could produce a substantial reduction of flow around the whole system. (The cooler water in the top tank will be attempting to flow against the tide to get into the bottom tank)

hot would filter upwards enthusiastically, even through a modest aperture, given the funnel shape of a cylinder top..

The hot water from the bottom tank will only enter the system at the rate, that hot water is being drawn out at the taps. Even so, "energetic" rising of this water could be counter productive. Once you get water moving it doesn't like to stop suddenly, so this will give rise to mixing of the relatively cooler water at the bottom with the hotter water at the top.

The thing is, during the heating process convection is a good thing. It ensures that as much water as possible becomes hot. Once you've heated the water and start drawing it off, mixing is a bad thing. You don't want your nice hot water mixing with the cooler water in the bottom of the tank. In a single cylinder system, this works a treat. With this two cylinder system, I think the complications of the mixing during the draw, will make it less efficient than anticipated.

I would favour putting the two tanks, side by side and plumbing them in parallel. If they are identical in size, and have identical heating arrangement, then no hot water needs to pass through any cooler region. It should also ensure a good flow rate. Even then you'd want to take steps to balance the flow from each.

 

[PaulAH]

The hot water from the bottom tank will only enter the system at the rate, that hot water is being drawn out at the taps. Even so, "energetic" rising of this water could be counter productive.

I think we're talking about two different things, TexMex. I am trying to imagine the scenario before any water is drawn, at the end of the heating cycle but before the effects of turbulence kick in. My theory is that this body of water would fairly soon settle down into a stable state and develop a gradient similar to that of a single body. The hottest energy pockets will have fought their way to the highest point, even through 22mm, and vice-versa.

Must agree that parallel makes more sense. I was going to suggest if but from Webygeezer's question it seemed the footprint would be too big for the cupboard.

 

[TexMex]

I think we're talking about two different things, TexMex. I am trying to imagine the scenario before any water is drawn, at the end of the heating cycle but before the effects of turbulence kick in. My theory is that this body of water would fairly soon settle down into a stable state and develop a gradient similar to that of a single body.

I am not sure it would, That single 22mm pipe will have to accommodate the rising hot water (from the bottom tank) and the falling cold water (from the top back down). If that pipe went straight from the top of the bottom one, into the bottom of the top one, it would possibly work, albeit quite slowly. But the addition of a couple of 90 degree bends would probably reduce the motion to an absolute crawl. But I am also, only guessing.

 

[PaulAH]

But I am also, only guessing.

Me too! Probably a suck-it-and-see (expensive suck). Agree, any sharp bends would have the hot pockets fighting for rite of passage.

 

[webbygeeza]

First of all thank you for your input, your comments have been referred to my plumber from which came a solution that has been working great for the last couple of weeks.

We decided not to feed fresh water to the top tank thus leaving one single 28mm feed from the storage to the bottom tank. 28mm theoretically hot feed from bottom to top tank. 22mm hot out to pump from top tank. And 22mm return to storage from top tank.

Once again thank you.