The Great 'Get Paid To Generate Energy' Hunt

albyota

g-a-g, Is this the sort of thing you had in mind?

green_as_grass

Zeupater

Fascinating.

I wish we had people doing research studies comparing the type of system you describe with a conventional one-tank system to measure the difference in efficiency in real-life situations. In mentioning insulation I had in mind that two tanks would tend to lose more heat (more surface area for the same volume of water), but possibly that could be mitigated somewhat with effective insulation.

What's a 'PLC' controller?


Nice diagram, Albyota.

My set up is quite different though.
1. My thermal store doesn't have an external heat exchanger, just a large DHW coil within the store, so no pump required to circulate around the heat exchanger when hot water is required.
2. My two solar coils will be set up to be supplied with solar heated water one-at-a-time. Your diagram shows them connected in series, which would have the effect of evening out the temperature top-to-bottom to some degree rather than concentrating the heat at the top, initially.

I'd try to post a diagram if I knew how.

John_Pierpoint

I am confused by that gold coloured thing, shown on the RHS of your large hot water tank?
What does that do?
I don't believe water could more than dribble through that thing and come out more than luke warm?

zeupater

green_as_grass wrote: »

What's a 'PLC' controller?

Hi

Effectively its a very basic form of computer which has input connectivity, programmable logic and outputs. I had intended to use one instead of a dedicated solar controller.

Link for further info :- http://en.wikipedia.org/wiki/Programmable_logic_controller

Regards
Z

albyota

green as grass,

I think I know the set up you have designed and in the main, would work, however, with the above diagram (which I drew up from what I thought from reading your first description, it is not my system)

if the water flow from the solar tubes feeds into the top of the store, it will heat the water at the top (only if the water at the top of the store is cooler).

As the top section of the store heats up, there can be no transfer from the hot solar water inside the top coil to the surrounding water, therefore the water will continue to flow through from the top coil to the lower coil where heat transfer can transfer, this set up, layers the whole store from top to bottom without causing a thermal current.

P.s.

John_Pierpoint, ignore the gold coloured thing on the RHS of the tank in the diagram, that was an attempt of a High Efficiency flat plate heat exchanger used on the heatbank type of store as show in the link. http://www.heatweb.com/products/cylinders/heatbank/heatbank.htm

green_as_grass

Albyota, you say

"if the water flow from the solar tubes feeds into the top of the store, it will heat the water at the top (only if the water at the top of the store is cooler).
As the top section of the store heats up, there can be no transfer from the hot solar water inside the top coil to the surrounding water, therefore the water will continue to flow through from the top coil to the lower coil where heat transfer can transfer, this set up, layers the whole store from top to bottom without causing a thermal current."

I see what you mean, but...
initially the heat from the panel would be transferred to BOTH the top and bottom coils. It would effectively act like one larger coil (except the circuit comes out of the tank and goes back in again between the two coils). This would reduce the initial concentration of heat at the top of the tank and spread out the heat top to bottom somewhat. For a longer period of time the store temperature would be 'lukewarm' and not practically useable for showers, taps etc. That is the opposite of what I'm trying to achieve.

Once the temperature at the panel matches the temperature at the top coil then heat would be transferred only to the bottom coil - agreed. But this would be a very temporary situation if the sun keeps shining. Then the temperature of the panel would increase further and heat would be transferred via BOTH coils again. That would be OK if the top coil temperature has reached useable heat temperature (say 60 degrees). If not it would still be delaying rise to the desired temperature at the top coil.

Once top coil temperature has reached 60 degrees all would be well, the top and bottom coils would continue both to take heat from the panel. Even heating top-to-bottom would be fine from then on.

Now, with my design, flow to the bottom coil is only allowed once the top coil has reached desired temperature, so temperature at the top coil increases more rapidly. Flow then switches to the bottom coil until temperature matches the top coil and beyond. When the bottom coil temperature exceeds the top coil then convection will cause the heat to flow from bottom to top, a thermal current as you describe. I don't see this as a problem.

Or have I missed something?

green_as_grass

Zeupater

Thanks for you description of PLC.

Going back to your post no. 231 a further thought:

We had been discussing rate at which a stratified tank would even out its temperature. You gave a value for conduction in water of 0.6W/m2 (of cross sectional area of the tank) heat transfer per m of separation between the top and bottom sections of the tank, per degree C difference, per hour.

Can this be rate be calculated regardless of direction of transfer of heat, given the effect of convection in water, causing heat to rise?

I would have thought that convection would inhibit the rate of conduction downwards, but would assist the rate of conduction upwards.

Not sure if I'm expressing this well, hope you know what I mean.

John_Pierpoint

albyota wrote: »

John_Pierpoint, ignore the gold coloured thing on the RHS of the tank in the diagram, that was an attempt of a High Efficiency flat plate heat exchanger used on the heatbank type of store as show in the link. http://www.heatweb.com/products/cylinders/heatbank/heatbank.htm

Are these things as powerful as the graphs on the bumph suggest?
Would they keep their efficiency? (Mind you operating at say 55 degrees there should not be a problem with scale?)
Would they require a pump better suited to a compressor to get the water from the heat store through the exchanger?
Presumably "normal" mains pressure would be sufficient to push through the other side of the exchanger and still provide a modest shower in the bathroom?

zeupater

green_as_grass wrote: »

Zeupater

Thanks for you description of PLC.

Going back to your post no. 231 a further thought:

We had been discussing rate at which a stratified tank would even out its temperature. You gave a value for conduction in water of 0.6W/m2 (of cross sectional area of the tank) heat transfer per m of separation between the top and bottom sections of the tank, per degree C difference, per hour.

Can this be rate be calculated regardless of direction of transfer of heat, given the effect of convection in water, causing heat to rise?

I would have thought that convection would inhibit the rate of conduction downwards, but would assist the rate of conduction upwards.

Not sure if I'm expressing this well, hope you know what I mean.

Hi

What you need to do is remember that there are three forms of heat transfer within the tank structure and it's contents (convection/conduction & radiant) which are attemping to reach thermal equilibrium. Above this the contents and the structure are working to achieve thermal equilibrium with the external environment, again using all of the above forms of transfer. The external insulation on the tank slows the rate ot which equilibrium is reached. Even in a closed circuit thermal conduction will transfer heat along the pipe connections to the tank, both through the tube material and the fluid content, this is why it's important to insulate all connected pipework around the tank. All conditions being equal, conduction will be equal in all directions.

In an ideal situation, with no supply heat demand, stratification of the temperatures within the thermal store will be maintained and conduction would be the primary driver towards equilibrium, however, as soon as heat is removed from the top of the tank thermal convection currents will be created which will disrupt/destroy the static stratification of the contents (depending on the amount of heat removed). To minimise the extent of the convection manufacturers should build baffles into the internal structure, these effectively disrupt the convection loops and thus attempt to re-create stratification as quickly as possible. After the induced convection has settled and stratified thermal layers have started to form you will have a situation where conduction occurs at an accelerated rate until a smooth thermal gradient forms between the layers or tank sections.

What is being described is that whilst there is no call on heat from the tank, the main driver towards thermal equilibrium in a well constructed thermal store will probably be conduction, however, whenever there is a withdrawal of energy from the top of the tank convective currents will form, upsetting the stratification and thus accelerating the rate of conduction as a result of creating steep temperature gradients along the boundaries of convection.

Test results should be available for thermal stores, there is an interesting document here (http://www.jradmin.com/modus/uploads/img48493db2142541.pdf), unless you have access to the BS library, but it doesn't really address stratification issues in detail.

HTH
Z

zeupater

John_Pierpoint wrote: »

Are these things as powerful as the graphs on the bumph suggest?
Would they keep their efficiency? (Mind you operating at say 55 degrees there should not be a problem with scale?)
Would they require a pump better suited to a compressor to get the water from the heat store through the exchanger?
Presumably "normal" mains pressure would be sufficient to push through the other side of the exchanger and still provide a modest shower in the bathroom?

Hi

The efficiency will just be a product of the contact surface area to volume ratio along with the flow rate. Also the flow in such a system will be disrupted and become non-linear, further increasing the efficiency. Mind you though, if the plate separation isn't appropriate and serviceable in-line filtration hasn't been fitted, I'd hope that the modules are easily exchangeable

Z