Is it possible to harness the latent heat of freezing in a heat pump? 2

[microhydro]

Is it possible to harness the latent heat of freezing in a heat pump?

I have a water source heat pump application. The water is 40 F. and limited in flow. Is there a way to freeze some ice crystals at the evaporator heat exchanger which could flow out the water source discharge, without freezing the entire water source line?

 

[vpl]

It might be possible, but you would expend far more energy than you would gain.



Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

 

[waross]

Your heat pump is already benefiting from the latent heat of vaporization at differing vapor pressures. To shift your evaporation temperature/pressure lower will result in lower efficiencies. My information is very old, but about 30 degrees ambient used to be the temperature that many heat pumps switched over to straight electric heat as the drop in efficiency had reached the break-even point with electric heat. Your challenge would be to freeze part of the water without freezing all of the water.
Remember that freezing and rupturing a heat exchanger and introducing raw water into the refrigeration charge could negate your savings for a century or so.
There are two issues here;
1> Efficiency.
The efficiency will drop as the evaporator temperature drops. When you consider the wear and tear on the compressor and fans, you are approaching the point where the KWHr per BTU for the heat pump s closing in on the KVHr per BTU for electric heat.
2> Shortage of water.
See 1> Efficiency. and go to electric heat when short of water.
Or, off the wall, cut in the evaporator section of a suitably sized ice machine as an evaporator during water shortages.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[MintJulep]

Theoretically, probably.

Practically, no, not with a single evaporator.

You'll ice up the water side of the evaporator. This will reduce the already limited water flow and reduce heat transfer from the water to the evaporator. You'll lose more than you'll gain.

If you had two evaporators you could allow one to freeze, the run the other while the frozen one thaws.

 

[microhydro]

I had hoped that someone had already solved this problem. Here's my design. Please help me refine it.

I've been designing a system to control ice formation, and maybe even capture some latent heat of freezing. I plan to use coils of tubing inside a barrel of antifreeze solution. The 40 F spring water is gravity fed to my site, with a static pressure of almost 50 psi. It will sustain 3.3 gpm. When cooled to 32F, this is a heat flux of about 13,000 btu/hr.

Construction: Fill a barrel with antifreeze. Immerse a copper refrigeration evaporator coil from the heat pump. Arrange the evaporator coil so that refrigerant enters the barrel at the bottom and exits at the top. Surrounding that coil in the barrel, immerse 350 feet of 1/2" pex flowing the spring water. Arrainge the pex so that the spring water enters the top of the barrel and exits out the bottom. the 350 foot length will convert the 50 psi static pressure to heat as friction loss and flow about 3.3 gpm. Control the cooling action of the evaporator coil to limit the minimum temperature of the antifreeze in the barrel to a few degrees below freezing. My hope is that some ice crystals would form in the spring pex coil, but would be swept along in the 5.5 fps flow and not buildup inside the pex.

I think that the refrigerant could be significantly below freezing and still not freeze the spring pex line.

How much freezing am I looking for? If my 2.1 kw heatpump operates at a COP of 3:1, then it will need 14,280 btu/hr from the spring water. This 14,280 btu/hr plus the 7140 btu/hr from the work of compression yields a 21,420 btu/hr system. Cooling 3.3 gpm spring water from 40 F to 32F gives 13,210 btu/hr. 1070 btu/hr would then need to be recovered from the latent heat of freezing. So about 0.4% of the spring water would need to be frozen. It seems possible. Perhaps I could use an Automatic Expansion Valve and adjust the evaporator pressure to hold the bottom part of the barrel a few degrees below freezing when it is in full-power steady-state operation. There would be no concern of freezing for less than about 1.8kw shaft input power.

Does anyone have experience with these techniques?

 

[waross]

Ice tends to form on cold surfaces. Many ice machines work with flowing water over the surface on which the ice forms.
I am afraid that you are building an ice machine, with no provision to harvest the ice.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[zdas04]

The thermodynamics work, the mechanics don't. The key point of your design is

My hope is that some ice crystals would form in the spring pex coil, but would be swept along in the 5.5 fps flow and not buildup inside the pex

Wishful thinking will not keep the ice crystals from aggregating at a bend somewhere and becoming an ice block. In fact nothing will prevent this. Within minutes of the onset of freezing you will begin to see some amount of ice damming, and a few minutes after that you'll see an ice block. Then you'll be done until you thaw the whole system out.


David

 

[MintJulep]

Whatever you might have gained by freezing a bit of the flow you loose in the additional heat transfer through the PEX to the glycol and losses from the barrel surface.

 

[dcasto]

why not ask the heat pump vendor. tell them that you want the heat pump to operate in 40F water and see what they say.

 

[willard3]

You should spend some time studying Ice/Thermal storage systems. Thermodynamics for this type of system are OK, your concept of controls, piping and etc are way too simplified.

http://www.calmac.com/

 

[waross]

Use an ice machine or the ice making portion of the machine. All the ice handling issues are solved. With the time constants of a heat pump you should not be inconvenienced by two or three minutes of harvest cycle every fifteen or twenty minutes.
You will have enough ice for all you can drink. grin.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[zdas04]

Bill,
An ice machine is an interesting device. Fluid flows into an ice chamber of some shape and then the ice is shoved down a consistent cross section to the bin by mechanical force. In a heat pump the bends, twists, and fittings make the outlet cross section anything but consistent. An ice cube that won't make a turn is an ice dam. Ice machines work really well because they don't have any wishful thinking--the designers know that they won't make ice if the designer can't add enough mechanical energy to push the ice out of the freeze chamber. A fluid system just doesn't have that ability.

David

 

[microhydro]

Thanks, everyone for your thoughtful posts. I have gained an appreciation for the pernicious nature of ice dams.

In the system I outlined, I now suspect that any ice that formed inside the spring water pex coil at high flows would stick to the cold wall of the pex pipe and not be scrubbed off by the flow. Now I am asking, "Is there a way to control where the ice forms and remove it?" Perhaps the Ice Machine suggestion really is practical.

My new idea is to to use the antifreeze barrel and coil system as descried, but to alternate the flow on and off every 30 seconds. Perhaps in the still water, long ice crystals would form that extend into the middle of the tube. Then the high flow could break them off and sweep them away. I am only looking for 0.4% ice formation.

 

[zdas04]

I've operated above ground pipelines in country that gets really cold. What I've seen has always been what I call a "candle-wax" freeze. A few mm of ice forms on the pipe wall. Next time it gets cold, the existing ice acts as a great nucleation site and you get a few extra mm of ice. A few days later the line is almost full of ice and the fast flow down the center plus the insulating effect of the ice keeps it from freezing solid until a chunk breaks off (calves?) and plugs the center for a few seconds. By the time the chunk breaks away, additional ice has formed upstream and the line is plugged. In a 10 inch pipeline running up the north side of a 1,000 ft cliff, it was usually the 5th or 6th day of freezing temperatures before I lost flow altogether. Never did find a solution since the gas had about 2,000 lbm/MMCF of water vapor that was going to condense somewhere inconvenient.

I would expect your Pex to act the same way, the ice will tend to form at the coldest part of the flow which happens to correspond to the no-flow boundary with the pipe.

David

 

[microhydro]

So when the wall of the pipe is the coldest part, ice will form there. Is there an ingenious way to conduct the heat out of the water but not through the wall of the pipe? Like injecting super-cooled air into the line, except simple.

 

[waross]

I have worked on one or two ice machines. Once flew from Canada to Central America to repair one at the owners expense.
A popular design for tube ice has vertical tubes surrounded by refrigerant. Sort of a short fat tube type heat exchanger.
There is a sump and a small pump delivering water to the top of the tubes. The water freezes on the walls of the tubes. As the water freezes and the ice gets thicker it acts as an insulator. Not much of of an insulator but there is enough change to drop the suction pressure as the ice builds up.
A pressure switch triggers the harvest cycle. The water is stopped and the hot gas discharge is diverted from the normal condenser to the evaporator. The temperature of the evaporator rapidly rises and shortly above the freezing point starts to melt the bond between the tubes and the ice. The ice columns drop by gravity onto the cutting disk. In a few minutes harvest terminates and ice making resumes.
The only water pressure needed from the source is enough to fill the sump with make-up water through a float valve.
Some designs of flake ice machines work continuously, the shaving blades shaving the ice off the inside of a large cylinder as it is formed.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

I should point out that an analysis of the efficiency MAY suggest that it is cheaper overall to just switch to electric heat. Making ice implies evaporating at a temperature below 32F. That was the old rule of thumb for the point where the BTUs per kW with a heat pump dropped below the BTUs per kW of electric heat.
I am not saying this will happen, but you are close enough to an old rule of thumb that you should evaluate your local conditions and do some BTU/kW calculations.
I would hate to see you spend a lot of time and money to adapt an ice machine that works great and is JUST A LITTLE MORE EXPENSIVE TO OPERATE THAN ELECTRIC HEAT.
You are approaching that operating point.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Dave41A]

Micro-hydro:

Certainly a novel idea...as you observe, the low (40F) temp of your water supply only gives you 8 deg F of room to work with until the water starts to freeze. The low water flow doesn't help, either.

However, the short answer is yes, you can use the latent heat of fusion of the water as a heat source, but (as many others have noted) you need to get rid of the ice.

If this is for a residential application, you could have an ice-maker and just have the owner dump the ice (manually) every day or so (dump the ice outside...if you let it melt in the kitchen sink you're literally sending all your heat down the drain). Sort of like emptying ashes from a woodstove.

If this is for a larger-scale application or emptying the ice manually is not practical, it sounds like you are really looking for a glorified snow-cone maker. Possibly you could pump or otherwise transport the ice/water slurry somewhere, but handling the two-phase flow (liquid/solid mix) will likely pose the greatest headache.

If this is for a commercial application, is it possible the ice could be sold as a by-product?

Good luck,
Dave

 

[Dave41A]

microhydro:

By the way, what is the reason for the barrel of anti-freeze? Introducing the intermediate fluid between the refrigerant and the spring water will just slow the overall rate of heat transfer.

This may be what you intend, but refrigerant-water heat exchangers are readily available for this type of application if you need one (any a/c system that uses "chill water" will have one).

Good luck,
Dave

 

[microhydro]

Thanks for your advice, waross and Dave41A.

The application is residential heating. The heat pump will be located outside, as will the evaporator heat exchanger. I conceived the antifreeze barrel idea to finely control the cooling of the spring water to create ice crystals but not freeze the spring line up. Solid advice from this forum has led me to abandon the two phase ice-in-the-tube idea. Perhaps I should use a thermostatic expansion valve and an oversize copper evaporator coil directly immersed in a barrel of the spring water.

Additional heat could be collected from a small ice rink. This would avoid the ice maker complexity and ice disposal problem.