thermodymic panel for water heating

1echidna

zeupater wrote: »

Hi

I'd want to see a a set of performance curves detailing COP vs cylinder water temperature at varying levels of irradiation, something similar to what's available for pv panels before I would believe COPs at those levels for water heating ....

Air source heatpumps which are dedicated to heating water from a range of global Japanese household-name manufacturers, known collectively as 'Ecocute' systems, seem to claim an average COP for heating water to 55C of 1.8 at -15C ambient, 3.1 at 7C ambient & 3.75 at 20C ambient and even products supplied by these global leaders often struggle to perform at anywhere near these published figures under long-term test conditions ...

A heat pump is a heat pump and it's very likely that compressors used in the plate units under discussion are made by the same manufacturer as ones used in the Japanese units raised above, the mechanism for transferring heat from the refrigerant to the water will be exactly the same and the cylinders will be insulated to the same standards .... that leaves the only difference being the passive plate exchanger vs a blown heat exchanger so we're left with a the detrimental effect of running a fan vs potential irradiation gain if the panels are in sunlight when there is a heating requirement ....

Given the high temperature of the cylinder and the figures above, I'd expect the compressor to be working as hard as it could to provide heat into the already warm cylinder because the refrigerant return temperature to the evaporator would be high, so how does a COP at 7C ambient of 1690/550 from the abovequoted figures look .... it seems to be 3.1 ... what a surprise, it seems to be exactly the same as Japanese 'Ecocute' systems ....

HTH
Z

I agree there may be some shady dealing going on here. Really I should be looking at the phase diagrams and enthalpy entropy relationships for R134a and going through the cycle in detail. There is a standard for measuring COPs of heat pumps, EN14511 and for air to water it specifies an air temperature of 7C and a water temp of 35C. Now if the compressor is variable speed there could be a stage in a heating cycle when the refrigerant was condensing at 40C and the hot water was 35C. I think this in itself is a bit sharp practice as it implies the tank is fully cold at the start of each heating cycle for 35C to be a reasonable mid way point, and many families might operate the system with the temperature nearer 55C most of the time.

As for the differing evaporator and compressor loads I am uncertain. Maybe one is with a solar gain on the evaporator.

I might compose an email on various points to the supplier who has offered to answer any questions.

grahamc2003

zeupater wrote: »

Hi


A heat pump is a heat pump and it's very likely that compressors used in the plate units under discussion are made by the same manufacturer as ones used in the Japanese units raised above, the mechanism for transferring heat from the refrigerant to the water will be exactly the same and the cylinders will be insulated to the same standards .... that leaves the only difference being the passive plate exchanger vs a blown heat exchanger so we're left with a the detrimental effect of running a fan vs potential irradiation gain if the panels are in sunlight when there is a heating requirement ....
Z

Agreed that the heat pump bit is standard stuff - nothing special or different there (i.e. all modern heat pumps have the capability of higher cops than previously purely because of the higher temps and pressures modrn refrigerants enable). So the difference is the heat collection, as you say, with the trade off as you say. Where I disagree is your statement about sunlight, when there is a gain in these panels due to radiation all the time, rather than just when in sunlight, albeit, the largest gains will be when in sunlight.

The measurement of the cop at a specified ambient temperature (7C) is a bit of a red herring for these panels. They could have very high cops at very low temperatures if it were arranged to have suffient radiation falling on them. (Put a human in a matt black sphere, and the radiation transfer from him to the sphere will be about 1kW whatever the temperature inside the sphere - it's pretty critical that that perhaps counterintuitive example is fully understood before an understanding of how these panels work can be made).

There are extremely high heat transfers via radiation all the time, and I think you are probably underestimating them in this system.


Z[/QUOTE]
HTH

zeupater

grahamc2003 wrote: »

Agreed that the heat pump bit is standard stuff - nothing special or different there (i.e. all modern heat pumps have the capability of higher cops than previously purely because of the higher temps and pressures modrn refrigerants enable). So the difference is the heat collection, as you say, with the trade off as you say. Where I disagree is your statement about sunlight, when there is a gain in these panels due to radiation all the time, rather than just when in sunlight, albeit, the largest gains will be when in sunlight.

The measurement of the cop at a specified ambient temperature (7C) is a bit of a red herring for these panels. They could have very high cops at very low temperatures if it were arranged to have suffient radiation falling on them. (Put a human in a matt black sphere, and the radiation transfer from him to the sphere will be about 1kW whatever the temperature inside the sphere - it's pretty critical that that perhaps counterintuitive example is fully understood before an understanding of how these panels work can be made).

There are extremely high heat transfers via radiation all the time, and I think you are probably underestimating them in this system.

Hi

I really do understand that the panels will absorb radiation even in the dark .... it's the level of additional heating this provides which I believe is being missed and I'll try to explain why, again, but in more detail ....

In very bright sunlight there's a maximum 1000W/sqm of energy available to be collected from the sun ... I hope we can all agree so far ...

A black body at 280K (approx 7C) emits ~348W/sqm (0.0000000567x280^4) ... this is approximately the total amount of radiation available if all surfaces (walls, ground, clouds, trees ... etc) are at 7C ... I hope we can all agree to this point too ...

We now have a maximum radiation which can be absorbed by the panel ... but only if it was cooled to 0K, that's -273C, which it's not ... I hope we can all agree to this point too ....

So, what's the level of radiation being emitted by the panel ? ... let's take the average temperature across the entire plate as being -25C (248K) ... it therefore emits ~214W/sqm (0.0000000567x248^4) ... I hope we can all agree to this point too ...

So we have an energy budget for the available radiation to provide heat to the panel (if the transfer process was 100% efficient in a vacuum) when solar gain is removed ... 134W/sqm (348-214) ... do we all still agree ? ... I'm simply using laws of physics & logic ....

Now, if there's agreement on the above, a panel 1.7mx0.8m has a physical limit of absorbing 364W (1.7x0.8x2x134) of radiation emited at 7C ...

Now then, according to manufacturer's source marketing literature & implied data, a single panel system delivering 18kWh of energy into the cylinder over 5 hours results in a rated average power of 3600W(18/5) of which 390W could be supplied by the compressor, therefore the panel is averaging 3210W ... do we all agree on this ?

So, the maximum contribution of heat radiated by bodies at 7C which could be collected by the panel accounts for ~11.3%(364/3210) of the heat being supplied by the panel, if in a vacuum .... do we all still agree to this point ? ....



Let's now put the panel on a roof and face a clear sky, we have the 214W/sqm being lost to a black body, space ... if its cloudy it's a new energy budget calculation to the clouds, they're probably around -6C (288W/sqm) on average ... so, would this have an effect ?

So what happens below the panel ... do the tiles on the roof cool due to the cooled air temperature ? ... will a body cooled to somewhere between 7C & -25C emit as much radiation as one at 7C ?

I really do hope that this helps place the radiation absorption into context, am I underestimating ... or ... are others overestimating ??.

HTH
Z


#Edit - Reference source for technical details if anyone still doesn't agree but understands laws of physics ... http://en.wikipedia.org/wiki/Stefan%E2%80%93Boltzmann_law

1echidna

You can't buck the basic thermodynamics whatever refrigerant or fancy evaporator you use.

The maximum performance achievable is

COP(heating) = T(hot)/(T(hot) - T(cold))

Ref http://en.wikipedia.org/wiki/Coefficient_of_performance

1st Case - condenser temp 40C, evaporator -20C

COP = 313/(313-253) = 5.2

2nd Case - condenser temp 60C, evaporator -20C

COP = 333/(333-253) = 4.1

The better the heat exchange and compressor technology the closer to these figures may be achieved, but they will never be reached in practice.

zeupater

grahamc2003 wrote: »

... Put a human in a matt black sphere, and the radiation transfer from him to the sphere will be about 1kW whatever the temperature inside the sphere - it's pretty critical that that perhaps counterintuitive example is fully understood before an understanding of how these panels work can be made ...

No, no, no ..... this is not what a blackbody is or does ....

If the temperature of the black sphere was 0K maybe, if the black sphere was at a higher temperature than the person inside then the energy budget would cause both the person and the air inside it to warm, if it was sufficiently higher than the person, and assuming that there was sufficient thermal mass in it, the person inside would eventually roast ... the sphere would still be black because it hasn't reached a temperature which is high enough to emit the frequency of radiation we see as light ...

HTH
Z

... my oven is black inside, not that it has any bearing though ...

zeupater

1echidna wrote: »

You can't buck the basic thermodynamics whatever refrigerant or fancy evaporator you use.

The maximum performance achievable is

COP(heating) = T(hot)/(T(hot) - T(cold))

Ref http://en.wikipedia.org/wiki/Coefficient_of_performance

1st Case - condenser temp 40C, evaporator -20C

COP = 313/(313-253) = 5.2

2nd Case - condenser temp 60C, evaporator -20C

COP = 333/(333-253) = 4.1

The better the heat exchange and compressor technology the closer to these figures may be achieved, but they will never be reached in practice.

Hi

Agree, although this would be a zero energy compressor which would seriously effect the input vs output COP ...

Anyway, COP aside, the claim for the additional performance benefit of the flat plates seems to be based on non-solar radiation sources ... do you understand and/or agree with my previous post (#74) regarding radiated heat transfer and if not in agreement where is it substantially incorrect .... ?

Z

1echidna

zeupater wrote: »

Hi

Agree, this would be a zero energy compressor ...

Anyway, COP aside, the claim for the additional performance benefit of the flat plates seems to be based on non-solar radiation sources ... do you understand and/or agree with my previous post (#74) regarding radiated heat transfer and if not where is it substantially incorrect .... ?

Z

All I know is that the heat transfer coefficient I quoted of 14W/m2C (post #37) will include both radiation and convective heat transfer, but not solar. The figures are probably gained from experience and analysis of what goes on in refrigeration evaporator plates inside refrigeration systems. Taking some revised figures we have

Evaporating temperature -20c
Ambient temperature 7C

Temperature difference 27C

Area of panel 3.2m2

So heat transfer rate = 27 x 3.2 x 14 = 1210W = 1.2kW

The figure in the details I have for evaporator load are 1690-390=1300W = 1.3kW, so not a great disceptancy.

There is also a higher evaporator load quoted of 2900 - 550 = 2350W = 2.35kW. I don't have a good feel for solar heat gain perhaps an extra 1.05kW for the exposed half of the panel (1.6m2) is not unreasonable.

I am not at all comfortable with 18kwh in 5h with 390W compressor input!!

grahamc2003

zeupater wrote: »

No, no, no ..... this is not what a blackbody is or does ....

If the temperature of the black sphere was 0K maybe, if the black sphere was at a higher temperature than the person inside then the energy budget would cause both the person and the air inside it to warm, if it was sufficiently higher than the person, and assuming that there was sufficient thermal mass in it, the person inside would eventually roast ... the sphere would still be black because it hasn't reached a temperature which is high enough to emit the frequency of radiation we see as light ...

HTH
Z

... my oven is black inside, not that it has any bearing though ...

Hmmm, you didn't read or understand what I (perfectly correctly) stated. Do you want another read more carefully and come back again?

zeupater

1echidna wrote: »

All I know is that the heat transfer coefficient I quoted of 14W/m2C (post #37) will include both radiation and conductive heat transfer, but not solar. The figures are probably gained from experience and analysis of what goes on in refrigeration evaporator plates inside refrigeration systems. Taking some revised figures we have

Evaporating temperature -20c
Ambient temperature 7C

Temperature difference 27C

Area of panel 3.2m2

So heat transfer rate = 27 x 3.2 x 14 = 1210W = 1.2kW

The figure in the details I have for evaporator load are 1690-390=1300W = 1.3kW, so not a great disceptancy.

There is also a higher evaporator load quoted of 2900 - 550 = 2350W = 2.35kW. I don't have a good feel for solar heat gain perhaps an extra 1.05kW for the exposed half of the panel (1.6m2) is not unreasonable.

I am not at all comfortable with 18kwh in 5h with 390W compressor input!!

Hi

Total solar irradiation in the UK is approx 1000kWh/sqm of flat ground per year ... that's an average ~228W/sqm (1000/(365*12)) per daylight hour, with maximum gain being ~1000W/sqm during exceptionally strong sunshine .... therefore it's likely that the 1.7mx0.8m panels at 100% collection efficiency would be expected to average just over 1.5kWh (1.7x0.8x0.228x5) of the 18kWh daily calculated demand used within the referenced manufacturers marketing material ... that's an 8.3%(1.5/18) contribution ...

So, in effect we are discussing a standard ASHP setup where we can remove the inefficiency caused by running a ~100W fan on the evaporator, that would reduce the rated input power by 20% (100/(390+100)) add a few percentage points (11.3% max & likely well under half of this) for the emitted radiation absorption and another few percent for direct solar (~8%) gain if only running through the day .... and this 8% advantage is being considered as being preferable to running the same unit or a standard ASHP overnight on E7 for around a third of the cost ? ....

HTH
Z

zeupater

grahamc2003 wrote: »

Hmmm, you didn't read or understand what I (perfectly correctly) stated. Do you want another read more carefully and come back again?

I read, reread and understood what was written ... please note that these conditions would be in a vacuum and if the blackbody temperature was low ... possibly close to 0K as stated in the reply ... please check.

I intentionally used the 'black oven' to demonstrate the point ... the oven is black, there is no light and it isn't switched on, but, importantly it isn't a freezer because it has a thermal mass and a temperature significantly above absolute zero.

Please also have a look through the points raised in post #74 and explain where a pretty well known law of Physics is wrong or where the logic I have applied this law to the flat panel is substantially wrong ... I specifically structured the post so that the point at which consensus is lost can easily be identified and discussed ... a typical problem solving approach - take a complex problem and split it into smaller problems which can be solved ....

HTH
Z