thermodymic panel for water heating

1echidna

On industrial systems I have seen a turbo expander put in rather than an expansion valve, the power generated being used to boost the gas off the evaporator. The expander and booster compressor being mounted on the one shaft. Not sure how this would work on a micro scale but does give a good boost in efficiency.

zeupater

grahamc2003 wrote: »

The temperature of the plate will not vary to the extent which you imply, and will be, if well designed, reasonably constant and cold over most of the surface. This is because the heat is picked up mainly as latent heat of evaporation at the boiling point (i.e. constant) temperature of the refrigerant. It isn't substantially picked up as a temperature rise, which will only happen when all the refrigerant has boiled. As you say, the heat absorbtion will drop with a rising plate temperature - hence why all refrigeration cycles use fluids with high latent heat of evaporation. the fluid state within the plate will be both liquid and gas at the boiling point, with more liquid at the start and no liquid near the exit (and only from this point the temperature will rise). (This is why the refrigerant volume in heatpumps is very important, and I suspect why some heatpump systems work well and others dont).



Don't know why you are still saying this. It is completely incorrect. The plate works by completely different physics to the coils in an ashp as previously explained. Your apparent reluctance to accept that is leading you to make many mistakes in your analysis.



I'm naturally sceptical too, but this isn't on a par with finned rare-metal filled heaters (as you perversely seem to be treating them).
They may work well or they may not, but certainly the system is within the bounds of thermodynamics. I think one main problem colouring your analysis is the lack of recognition of the plate temperature which is crucial to the physics under which it operate. For example, the heat pickup on a sunny day will not come solely from the sun (which is mainly the case with other solar panels at ambient temperature). A cold plate will pick up all the direct radiation from the sun, plus radiation from everything else which is warmer than the plate (which is everything, including roof tiles the other side of the panel points at).

Also I think there's a misconception in your understanding of the details of the refrigeration cycle itself at the compression stage. The compression is designed to be isentropic (i.e. performed at constant entropy) and all analysis assumes that, hence why a 390W compressor delivers 390W (almost) into the working fluid as a temperature and pressure increase - it's not about reclaiming lost heat as you implied, this is a critical energy input into the system.

Hi

The Temperature of the evaporator plate will vary considerably over the length .... the refrigerant will boil immediately at the point of pressure drop as it's the transition between a high pressure liquid form and low pressure gas which cools the refrigerant, so this is the coldest point in the cycle .... at this point the latent heat capacity of the refrigerant is reduced because the density is reduced, therefore as the gas warms it becomes less able to cool the downstream plate surface, remember, the differential temperature of the gas entering to leaving the single plate system is delivering 3.6kW of heat (marketing claim 18kWh over 5 hours) .... As you say, it will be 'reasonably cold over the entire surface', but there will be complex temperature gradients over the entire plate ... Looking at the exchangers in question they look to simply be a rollbond unit with two half-plate circuits connected in series ...

Regarding the 'plate working differently' ... how and why ? .... if that relates to absorption of radiated heat being collected from surroundings, let's put it into context ... during the day it's warm, during the night it's not ... sunlight provides ~1000kWh/sqm per year in the UK ... the panel running for 5 hours/day extracts almost five times that figure (4830kWh/sqm per year (18*365/1.36)), so this extra is radiated by what ?, reflection ?, something providing 5 times the energy as the sun ?, no, it's direct contact with the air, it's conductive heat transfer which cools the air which causes negative convective flow across the plate ... look at the pictures in the referenced marketing literature - the downward airflow is further cooled by the plates, reaches dew point, condenses on the lower section of the plate, freezes and forms frost .... take a standard refrigeration condenser and instead of welding a flat plate parallel to the serpentine tube runs you fold the serpentine a number of times and then attach smaller plates at a tangent, then you have a fan-blown condenser, take an evaporator and do the same and you have a fan-blown evaporator ... there's nothing complex, this is how our fridge & freezer work (fridge is rollbond+plate, freezer is fan blown) ... take a fan blown evaporator and a fan blown condenser and you have the basis for an air to air heat pump, replace the condenser with a heat exchanger in water and you have an air/water heatpump system and so on for GSHP etc .....

Regarding the 'cold plate will pick up all the dirct energy from the sun' ... fine, if this is the case, then why can I see white on the plate surfaces ? ... doesn't this suggest that the majority of irradiation would simply be reflected and solar gain is pretty irrelevant ? ....

All I'm attempting to do is bring the claims on the units into the context of everyday 'real world' average conditions, not a 25C average, not 1000W/sqm .... the marketing details claim and/or imply 18kWh of heating provision within 5 hours using a 390W compressor with the capability to achieve this every day of the year .... the cost reduction for the provision of the heating (electric vs electric) is £919.80/year to £109.50/year ... again, this is an implied annual COP of 8.4 (919.8/109.5) ... so the question is 'Is a COP of 8.4 considered reasonably achieveable ?' .... my rough and ready simple calculations suggest a COP range from 2.5 to 3.8max, and I would expect that the average would be closer to 2.5 ... 1echidna's latest calculation suggests a maximum COP under test conditions of 4.0, I'll not worry about the odd 0.2 (4-3.8) difference considering the different approaches employed, but do either of these approaches confirm a COP of 8.4 ?, so at the moment, in my mind, it cannot be considered as passing the reasonability test ....

HTH
Z

zeupater

grahamc2003 wrote: »

.... Also they deliver heat at a higher rate than a passive system (under sunny conditions). They claim 2.9kW max - seems reasonable to me. A passive system (with less area and at ambient temperature) will perhaps deliver 400W (?, I'm asking). But for those cliaimed advantages (and the claims seem reasonable to me and within the bounds of physics and thermodynamics), as you point out, you've got to use a 390W compressor, as opposed to a 20W pump in a passive system ....

Hi

An update regarding the 400W query above .... for the first time in days the sky has been pretty clear for a while, allowing the thermal panels to reach their potential .... currently feeding 2.2kW.t into the HWC with a 20W pump, a COP of 110 .... this is from a roof footprint equivalent to four 250W panels .... For comparison, at the same time our 4kWp pv is generating 3.4kW from 16 panels ... the outside temperature is 12.4C ...

HTH
Z

grahamc2003

zeupater wrote: »

Hi

An update regarding the 400W query above .... for the first time in days the sky has been pretty clear for a while, allowing the thermal panels to reach their potential .... currently feeding 2.2kW.t into the HWC with a 20W pump, a COP of 110 .... this is from a roof footprint equivalent to four 250W panels .... For comparison, at the same time our 4kWp pv is generating 3.4kW from 16 panels ... the outside temperature is 12.4C ...

HTH
Z

Hi Z, I think we are going to have to agree to disagree on several points (not in the above post, but the previous). Also, you quote many claims which I simply haven't seen claimed - in my quote, the blurb talked about max power from 1 panel of 2.9kW and a max cop of 5.2, with cops up to 7 for multiple panels. Obvioulsy, a maximum isn't an average, and would only be in ideal conditions, perhaps only for a few minutes per ideal day.

I can't comment on 1echidna max cop calc of 4 or whatever because I have chosen to filter his/her posts after his/her first contribution. (a very new strategy I've adopted just a couple of days ago, prompted by an overvoluminous regular who I consider has nothing worthwhile to say).

Re the post quoted, I don't think cops are applicable to passive systems where, by their nature, they are high, infinite even. Put your panels below your HW tank, throw away the then unneeded 20W pump, and you have a cop of infinity all the time it operates.

edit - as to icing and frosting which you mentioned, I gave my thoughts on that in post 3 and expanded my view in post 6.

zeupater

grahamc2003 wrote: »

.... Also, you quote many claims which I simply haven't seen claimed - in my quote, the blurb talked about max power from 1 panel of 2.9kW and a max cop of 5.2, with cops up to 7 for multiple panels. Obvioulsy, a maximum isn't an average, and would only be in ideal conditions, perhaps only for a few minutes per ideal day.

Hi

Refer to previous referenced link (http://www.thermogroupuk.com/thermogroup_pdfs/TDY_Pres_MAY.pdf) ... have a look at the ice on the panels (remember that these are marketing photographs) and consider the 'Domestic costs and savings' tables ... an implied COP of 8.4 (919.80/109.50) is given for a single panel system ... surely you can see this ... also, for direct comparison, in order to deliver 18kWh of daily heat provision (looks pretty high, but lets go with it as it's their figures) in 5 hours then the system would need to be providing a continuous, or average, 3.6kWh (18/5), with at least 3.2kWh (3.6-0.39) being provided by the heat exchanger ... surely you don't disagree with this ??

Is a COP of 8.4 achieveable ? .... if so, why wouldn't a pair of 0.6sqm plate condensers from a couple of large freezers and a 200W compressor mounted through a wall provide exactly half of the duty of one of these systems and therefore provide a significant proportion of the heating requirement for my house ? .... what happens to the efficiency of one of their systems as the temperature of the water in the HW cylinder increases ?, does this mean that the 3.6kWh heating duty of the plate system must be a cycle average ? .... if so, what would the maximum COP rating be when the water in the cylinder is cold ? ..... leave the complex science behind for a while and just consider the underlying logic ...

HTH
Z

1echidna

grahamc2003 wrote: »

Hi Z, I think we are going to have to agree to disagree on several points (not in the above post, but the previous). Also, you quote many claims which I simply haven't seen claimed - in my quote, the blurb talked about max power from 1 panel of 2.9kW and a max cop of 5.2, with cops up to 7 for multiple panels. Obvioulsy, a maximum isn't an average, and would only be in ideal conditions, perhaps only for a few minutes per ideal day.

I can't comment on 1echidna max cop calc of 4 or whatever because I have chosen to filter his/her posts after his/her first contribution. (a very new strategy I've adopted just a couple of days ago, prompted by an overvoluminous regular who I consider has nothing worthwhile to say).

Re the post quoted, I don't think cops are applicable to passive systems where, by their nature, they are high, infinite even. Put your panels below your HW tank, throw away the then unneeded 20W pump, and you have a cop of infinity all the time it operates.

edit - as to icing and frosting which you mentioned, I gave my thoughts on that in post 3 and expanded my view in post 6.

my bold

I'm sorry to have upset Graham as I think we could have had an interesting discussion.

The basic thermodynamics as much as evaporator performance limit the achievable COP. There must be some novel compressor/expander design here together with process control system to give the kind of performance claimed. The basic thermodynamics of the refrigerant also means that for high COPs the temperature difference between expansion and condensation must be low and for this system I guess it means higher evaporator and/or lower condenser pressures when ambient and/or water tank conditions allow.

zeupater

1echidna wrote: »

.... The basic thermodynamics as much as evaporator performance limit the achievable COP. There must be some novel compressor/expander design here together with process control system to give the kind of performance claimed. The basic thermodynamics of the refrigerant also means that for high COPs the temperature difference between expansion and condensation must be low and for this system I guess it means higher evaporator and/or lower condenser pressures when ambient and/or water tank conditions allow.

Hi

The temperature differential can't be that low ... it's designed so that one plate can provide 18kWh in 5Hrs (their figures) to heat the water in the tank to 55C, therefore it must be expected to provide heat to the refrigerant/water condensing heat exchanger in a range between say 10C and 60C, and as previously mentioned, the efficiency of exchange will reduce as the differential between the compressed refrigerant in the condenser and the water falls ... the average COP for the cycle must therefore be 8.4, which would likely be considered possible at the average cylinder temperature, ie ~32.5C (10+((55-10)/2)) if heating from cold ... below this temperature the COP must be over 8.4, with the opposite being the case when the water temperature is above 32.5C ... this being due to the variable efficiency of the heat exchange process in the condenser related to temperature ... as the refrigerant temperature returned to the evaporator increases, the efficiency of the unit to collect heat at a given ambient air temperature also decreases ....

Another, and actually more logical approach considering the application, would be for the system to simply top up the tank temperature on a thermostatic basis, for example, reheat the tank to 55C whenever a sensor in the lower part of the tank falls below 45C .... this would mean that the COP of 8.4 would need to be achieveable at a much higher average temperature ....

I don't know what the answer is without seeing full technical details, but am pretty certain that nothing seems to stack-up regarding the claimed performance and heating duty ...

HTH
Z

1echidna

Heat Pumps
Synergy of High Efficiency and Low Carbon Electricity
Akio Koike, Tokyo Electric Power Co. JAPAN

Coefficient of Performance (COP) 16 of air-conditioners for residential use in Japan has doubled in the past 10 years and now
exceeds 6 in COP of both heating and cooling modes.

Such drastic improvement was achieved by not only a single breakthrough but also by the accumulation of every small progress in element technology; decreasing the loss of the compressor, applying the permanent-magnet motor to increase efficiency (Figure 10), efficiency improvement in the fin and the tube of the heat exchanger, improvement in the refrigerant cycle, etc. 19 20 21

In addition, the inverter-controlled air-conditioner, which enables variable speed operation, remarkably improves the efficiency of the heat pump at a partial load. Now, almost all of the new air conditioners for residential use produced in Japan are equipped with this inverter, while only 20% in Europe and almost zero in the U.S. have this function.

zeupater

Hi

That is for low temperature air to air... What about air to water with a temperature of 55C?

Z

1echidna

zeupater wrote: »

Hi

That is for low temperature air to air... What about air to water with a temperature of 55C?

Z

Remember the water starts out at 10C approx perhaps which is where modern control systems could come into play for this part of the heating cycle.

Coming back to your 18kWh in 5h that does seem out of kilter and I'm wondering if there can be some misinterpretation here.