Water heat pump

[Cool Mike]

Hi all! I am busy building a domestic water heater with a 2,5kw split air conditioner compressor unit using R22 gas. Could anyone help me with the sizing of the condenser (the heater) I have used about 9 metres of 1/4 inch copper tubing which I have coiled up to fit into the water storage tank (150l) The problem I am having is that the high pressure side is going up to 500psi. I am guessing 200 -250psi would be acceptable? The unit is heating quiet well. Do I make the condenser bigger, or do I shorten the capillary tube? When the unit is running the condenser only feels hot about half way up.Could anyone advise me on this?

[AUcooler]

At 250psi i'd suspect your water temperature is only going to be around 38-40degC on r22. If you go higher in water temperature then your compressor will fail on thermal overload due to high current draw. Try using 3/8-1/2 copper for your condenser coil. If you want a sloppy method of calculating your condenser length then measure and count the rows on your evaporator, then add 20%. Your experiment will fail using R22 with that compressor.

I'd suggest replacing the TXV or capillary and changing your refrigerant to 134a, and dont set the thermostat beyond 50degC. The capilary tube needs to match your compressor capacity on the chosen refrigerant.

Monitor the compressor current draw when charging the unit, rather then the suction pressure/temperature. Check the current draw again as the water temp approachs SP.

Your PT chart and Tong tester will be your best friends on this experiment.

[Gary]

Hi all! I am busy building a domestic water heater with a 2,5kw split air conditioner compressor unit using R22 gas. Could anyone help me with the sizing of the condenser (the heater) I have used about 9 metres of 1/4 inch copper tubing which I have coiled up to fit into the water storage tank (150l) The problem I am having is that the high pressure side is going up to 500psi. I am guessing 200 -250psi would be acceptable? The unit is heating quiet well. Do I make the condenser bigger, or do I shorten the capillary tube? When the unit is running the condenser only feels hot about half way up.Could anyone advise me on this?

Your tubing coil (heater) should be a de-superheater, not a condenser.

In other words, the discharge gas should go through the tubing coil and then through an air cooled condenser.

Use a pressure switch on the condenser fan to limit high side pressure.

[mad fridgie]

First use TXV.
Remember that the water is static in the cylinder (very poor heat transfer)
So for coil sizing look at 60Btu/Ftsqu/f
Let you convert to SI units.
Actual configuation we all have our own secrets

[Magoo]

Talk to des A, he is a guru on air to water heat recovery.
As per Gary's suggestion, desuperheating is inefficient. Max 15 > 20% of total heat of rejection to maintain compressor operating conditions.
R22 is not the best option either, consider R410A. Add water regeneration and cycle water discharge on loop temp

[Gary]

Talk to des A, he is a guru on air to water heat recovery.
As per Gary's suggestion, desuperheating is inefficient. Max 15 > 20% of total heat of rejection to maintain compressor operating conditions.


... when the condenser fan is running.

When the condenser fan is not running, the de-superheater becomes a condenser.

[desA]

Talk to des A, he is a guru on air to water heat recovery.


I'm honoured... We all stand on the shoulders of giants... Gary, Magoo, Mad-Fridgie, Chef... the giant's list is long :)

For reference:
1 Btu/(h.ft2.'F)=5.677 W/(m2.'C)=5.677 W/(m2.K)

Water-side coil heat-transfer coefficient, as per mad_fridgie:
60 Btu/(h.ft2.'F)=340.62 W/m2.K

Personally, I'd think that this could be a little on the high side for a tank internal coil, with unmixed water. The heat-transfer here will be restricted by the water-side natural convection, which, at best, will have a heat-transfer coefficient of around 5-10 W/m2.K

Tank internal coil:
With a refrigerant condensation coefficient of say 1000W/m2.K (well to the upper end, I'd expect - rough ballpark), the expected coil/water combined heat-transfer would be around (neglecting wall loss & fouling):

R,eq = (hi+ho)/(hi*ho) = (1000+10)/(1000*10)=0.101 m2.K/W

U,eq=1/R,eq=1/0.101 = 9.9 W/m2.K !!! (ie. aweful) :eek:

Next:
Apply small water pump to circulate the tank contents - mixed-mode, slightly-forced convection, heat-transfer:

ho~50 W/m2.K

R,eq = (hi+ho)/(hi*ho) = (1000+50)/(1000*50)=0.021 m2.K/W

U,eq=1/R,eq=1/0.021 = 47.62 W/m2.K !!! (getting better) :)

Next - again:
Apply small water pump to circulate the tank contents through a decent water-cooled condenser:

ho~5000 W/m2.K (water-side, conservative)

R,eq = (hi+ho)/(hi*ho) = (1000+5000)/(1000*5000)=0.0012 m2.K/W

U,eq=1/R,eq=1/0.0012 = 833.33 W/m2.K !!! (are we there yet?) :)

[mad fridgie]

As the patent holder of the worlds highest efficient hot water heat pumps (SFT and Boost) "probably as carlsberg would say", I can say R22 is an acceptable refrigerant, especially as you already have the unit.
You need to look at the evap configuration (it is more than likely configured as a condensor) Your are more than likely need to increase the number of circuits. (evap pressure drop)
This should give a higher SST, allowing you to increase your SCT (compression ratios)
The condensor top feed, do not worry if you over size, you can use as a type of reciever/ liquid sub cooler. You could use a small "p" trap to ensure a liquid seal. Control can be based upon temp and HP The HP cut should be under maximum rating for the compressor type for example if max rating is 65C then have it at 60C.

[desA]

As the patent holder of the worlds highest efficient hot water heat pumps (SFT and Boost)

http://www.sft.no/arbeidsomr/luft/kuldemedier/Faktablader_naturlige_kuldemedier_samlet_eng.pdf

An interesting link to the word SFT...

Probably totally out of context. You wouldn't be using ammonia for your heat-pump, would you? :eek:

[mad fridgie]

Hi DesA, a little conservative but natural convection in a cylinder always an issue as far as heat transfer, For this particular application based upon heat of rejection of 3kw I would look at at installing 39 lineal meters of 15mm (1/2inch).
Do not generally do static for the above reasons (variable heat transfer coefficients), internal mixing of the cylinder in my option is a no no, best use natural stratification.
Certainally I prefer the pump method, so what more predicatable.
I applogise for swapping between imperial and SI units (static coils I can only remember in imperial unit)

[mad fridgie]

http://www.sft.no/arbeidsomr/luft/kuldemedier/Faktablader_naturlige_kuldemedier_samlet_eng.pdf

An interesting link to the word SFT...

Probably totally out of context. You wouldn't be using ammonia for your heat-pump, would you? :eek:
Well found, Is not funny how search engines find differnt infomation depending where you are searching from! Maybe i need to change the name:D,
No not using ammonia, the process could accept ammonia, but I do not think i could make one for a price that i could ever sell (domestic size)

[desA]

^ lol... :D

[echotsang]

Buy an set of air source heat pump or take the r417a in to consideration.... @_@