Originally Posted by Captain Bill
I have a serious problem of understanding how using a fresh water tank as a source of heat for a heat pump is going to work from a physics point of view. Wrwakefield's water tank is 10 to 20 degrees warmer than the sea water because he is pumping heat out of his refrigeration unit into the tank. I'm not sure how his tank is set up but because it's staying warmer than the seawater it must have some insulation
from the seawater. It may just be the fiberglass hull or the hull plus an airspace. If he turned his refrigeration off over time it would find a balance temperature between the seawater temperature and the heat coming in from the boat's interior and it's not going to be anywhere near 20 degrees above the seawater. A simple calculation shows that a 110 gallon tank 20 degrees above the sea water temperature of 48 degrees has a heat capacity of 880 BTUs per degree F. So if the temperature of the water tank is 68 degrees and the minimum temperature that the AC units can work at is 41 degrees then the heat capacity of the tank is 23760 BTU. One 15000 btu/hr heat pump will reduce the temperature of the tank to 41 degrees in about an hour and a half of constant running. Now because the refrigeration unit is dumping heat in the tank it will actually take a little longer to reach 41 degrees but not much longer. The idea of using a fish
tank heater to maintain the temperature means that someone does not understand the difference between temperature and heat. Aquarium heaters come in sizes rate at a couple of hundred watts. A 15000 BTU/hr heat pump would require a 4300 watt heater just to maintain the temperature of the tank. Now if you had an aluminum
boat with one side of the tank being the hull in contact with the seawater and that contact area was big enough this scheme might work, but it's not going to work in a fiberglass boat with an independent tank.
I, too, wondered about the efficacy of such a set-up when I did some rough calcs before giving it a try.
Since I was cleaning
and replumbing the raw water circuits anyway, it was way too easy to connect to the second potable tank since all the hoses were right in front of me as I sat there ripping old hose out... So I installed a 3-way valve [so I could switch back to raw water if the experiment
failed...] and gave it a try with one 16k btu unit. It worked at freezing [outdoor air] temps that winter, so I plumbed up the other two heat pumps the same way.
I suspect we are lucky and get away with this partly because of our current cruising grounds [SE Alaska] Here the Pacific Ocean
water just doesn't get that cold S of ~59°N [high 30s, low 40s °F in winter
; ignoring glacial fjords, etc. where we wouldn't be using heat pumps anyway...] And, if we were planning not to venture to higher latitudes, I didn't even need to try this experiment
. I did so in case we venture further N where the water would be too cold to winter with the heat pumps in a harbor slip [one eventuality for us...]
So why has it worked so far? [Why doesn't the thermal reservior get too cold when the heat pumps are running?]
One thought [which you already voiced] is the one small refrigerator compressor
is adding a modicum of heat to the tank. The other thought is the heat pumps quickly heat the interior and do not run continuously even when they are the only heat source. [They may cycle on maybe up to 40% in cool spells, but usually less than 10 mins/hr.] And ambient waters rarely [if ever] drop below 40°F where we have wintered the last couple of years during this experiment.
I haven't thought about this for a while, and your questioning made me curious.
I just measured the water temps from both water tanks
[at the galley
faucet after running water for a couple of mins...] Water from both tanks
measured 49°F. Ambient raw water temp [measured from the dock
using the same simple dial face thermometer] is currently 40°F. Outdoor air temp today and this past week has been hovering around freezing, and we've been heating with the forced air diesel
. Nothing has been recirculating water in that 2nd tank. [The refrig that uses that tank has optional water cooling, and that has been off all winter...]
So apparently the water tanks are both getting some heat transfer from heated occupied space [galley and lower settee area] through the uninsulated tops; +9°F by today's measure. [N=1 so none of this is difinitive, but it does jive with the few times I have spot checked during this experiment the last couple of winters...]
The tanks are inline low against the hull on the centerline; large rectangles not a foot tall, so the largest ratio of surface area exists on the tops and bottoms. The sides are insulated, tops are not. I cannot see underneath, but will assume there is some foam there since it is on the sides and the centerline of the hull is not flat. FRP hull averaging 1+" thick- based upon through-hull cores from adjacent areas. And no heat source is near the tanks.
So this experiment must be continued in colder ambient waters to demonstrate viability for its intended purpose; support heat pump use should we winter further N in waters approaching freezing temp. [I'm not sure we are up to that challenge at the moment...]
So I'm shrugging my shoulders right along with you because I agree, it didn't seem like it should work...
It will be interesting to hear how this works out for others.
Thanks for your feedback.