Home made swimming pool heat pump

[desA]

Des, the guage reading on the LP side is irrelevant as the guage is not connected to anything as I have no valve to connect it to to take a reading - I plan to put one in but haven't yet. Only valid pressure reading is the HP one at 22 bar.

LOL - may want to re-calibrate your gauge, mind.

Then, with all due respect, we are unable to assist further, until you collect the data listed by Gary above.

[Gary]

Boiling point, while being a term familiar to the general public, is not commonly used in the refrigeration industry. We generally refer to the P/T conversion as saturation.

If you look at your gauges, you will see a scale that is identified as R22. This gives you the same information as the P/T chart (saturation temp) and is far more convenient.

[Gary]

For a starting point, we need the air and water temps as these have an enormous effect on everything else, but a full set of readings is needed for a full evaluation.

[frank]

I would have agreed with you some years back, but the truth is, there is really no problem at all with Chlorine treatment of pools and copper. However, strangely as it seems, if the pool operator doesn't keep the chlorine level right, ie high enough, then there would be a problem with corrosion. The Chlorine oxidises organic material in the pool, if there is insufficient free Chlorine then Chloramines are produced. It's these Chloramines that give the "chlorine" smell in pools and also causes sore eyes.

Children are a common factor here!

So when you smell chlorine in a pool hall it indicates the treament is off or not being kept up to adequate levels to combat the chidren

Some years back, a friend asked me to have alook at his fathers pool heater.

A basic set up with a condensing unit, maybe 1hp, situated within his large green house and a stainless steel cylinder with copper discharge pipework coil running through it. It looked like a home made jobby.

When I connected my gauges on the suction rotolock, water came out

When It was stripped in our workshop, the discharge copper coil was disintegrating,

I managed to replace it, dry out the existing condenser, rebuild it and re-install it back into his greenhouse. It's still going strong

I used 316 S/S tubing though, not copper in the re-build.

I haven't worked on any S & T HX's for pools since, only brazed plate HX's

[Gary]

The best way to calibrate your gauge is to attach it to your R22 container, along with a temp sensor taped to the side of the container. The R22 (saturation temp) scale on the gauge should match the temp of the container.

[Gary]

Some years back, a friend asked me to have alook at his fathers pool heater.

A basic set up with a condensing unit, maybe 1hp, situated within his large green house and a stainless steel cylinder with copper discharge pipework coil running through it. It looked like a home made jobby.

When I connected my gauges on the suction rotolock, water came out

When It was stripped in our workshop, the discharge copper coil was disintegrating,

I managed to replace it, dry out the existing condenser, rebuild it and re-install it back into his greenhouse. It's still going strong

I used 316 S/S tubing though, not copper in the re-build.

I haven't worked on any S & T HX's for pools since, only brazed plate HX's

I'm wondering if the corrosion might have been the result of using dissimilar metals.

[Acuario]

Excellent information and so simple!
I'll take readings as soon as I can but I'm going to need to get and fit a valve on the LP side first. I don't have one so it may be a few days.

Thanks all for the information to date. The pool has heated up about 1'C in the last 10 hours so something is working half right at least!

[Acuario]

I'm wondering if the corrosion might have been the result of using dissimilar metals.

Sounds like it to me - according to nobility table stainless steel is quite anodic and far away from copper which is cathodic with the result being that the copper will corrode.

[frank]

I'm wondering if the corrosion might have been the result of using dissimilar metals.

The corrosion was not at the point where the copper entered the s/s cylinder. It was the majority of the coil within the cylinder which was in constant contact with the circulating chlorine water.

[Acuario]

I've seen this happen with copper cold water pipes that were connected to galvanized iron fittings. I went into a customers garage and there was a fine mist of water being sprayed from the numetous micro perforations in the pipe. How it hadn't burst I'll never know. The whole length was pretty much the same. Water conducts electricity.

Also, years ago my dad worked for the gas company in the UK. In the late 60's/early 70's there was a shortage of copper pipe so they decided to install central heating with steel pipe. Total disaster, it all had to be ripped out and replaced due to electrolysis destroying the steel pipe.

[Gary]

It looks like I need to start planning Heat Exchanger Mk II but Mk I is working well. The pool pump pumps 20,500 litres per hour and I'm getting a dT of nearly 40'C across it in the refrigerant line.

The dT is the water out temp minus the water in temp. So... if the dT is 40C and the water out is 38C, then the water in is -2C? Somehow I doubt this.

[Acuario]

Sorry, you mis understood. The temperature of the refrigerant line into the exchanger is around 76'C, the exit temperature of the refrigerant line is 38'C so the temp drop across the exchanger in the refrigerant line is nearly 40'C, in terms of water temperature I doubt it's much, maybe 1 or 2 degrees but I haven't measured it. I'm sure there's some equation somewhere that would say what it was. At the end of the day there is heat transfer, that is what is important.

[Gary]

If the refrigerant leaving temp is 38C and the SCT is 57.5C, then the SC is 57.5-38=19.5K. This indicates liquid is backing up into the condenser, reducing heat transfer. The system is overcharged. The SC should 5.5-8.5K.

The change in refrigerant temperature between the inlet and outlet tells us nothing about heat transfer, because there is a change of state from vapor to liquid. A change of state involves heat transfer without temperature change.

[frank]

If the water in/water out temp difference is only 1C then you need to slow the water flow down.

This will increase the dwell time and allow a greater heat transfer, which will result in a lower discharge pressure

Throttle the valve before/after the pump and let us know what the results are

[monkey spanners]

Head pressure seems very high.....
Wonder if the cap tube is too restrictive?
You really want a txv in this sort of application, preferably one with a balanced port design.

Something to think about is what you want is a high energy transfer from the refrigerant to the water, which is not necessarily best made by a high refrigerant temperature.

[Gary]

First remove refrigerant until the SC is no more than 8.5K, then try adjusting the water flow.

When reducing water flow, if the SCT (discharge pressure) drops, that indicates increased heat transfer. If the SCT rises, you have gone too far and the water flow is insufficient. Whatever flow gives you the lowest SCT.

[Gary]

Head pressure seems very high.....
Wonder if the cap tube is too restrictive?
You really want a txv in this sort of application, preferably one with a balanced port design.

Something to think about is what you want is a high energy transfer from the refrigerant to the water, which is not necessarily best made by a high refrigerant temperature.

I would suspect that the pressure is high because the condenser is woefully inadequate, but at this point we don't have enough information to make the call.

[Gary]

In the condenser, I assume the refrigerant and the water flow in opposite directions (counterflow). If this is not the case, it will need to be re-piped.

[Acuario]

Wow, what a change! Looks like it was seriously over charged. I removed a load (sorry for the technical term) of refrigerant and now it's running with a head pressure of 15bar. I did calibrate my guages first.

The changes were dramatic. Previously there was frost/ice forming between the end of the evaporator and all down the compressor suction pipe. Now there is frost after the capillary and on the botttom rows of the evaporator coils. The temperatures have also changed dramatically. They are as follow:

Exit from compressor (measured at 150mm) 106'C, entry to the heat exchanger 84'C, exit of heat exchanger 28.8'C, exit from capillary (entry to evaporator) -13.8'C, exit from evaporator 6.8'C. All measurements taken with the case off which I'm sure makes a difference. Pressure at the one point I can measure, 15bar.

I tried reducing the water flow but as I don't have a bypass valve my poor pool pump started complaining so I don't think this is an option at the moment.

Just to confirm, the heat exchanger is configured to counter flow water against refrigerant.

[desA]

If we assume some further flashing after cap tube (typically is an additional 2-3K)

Te,sat~-16.8'C ; Tc,sat~55.6'C ;
SH=6.8-(-16.8)= 23.6K
SC=55.6-28.8=26.8K
Tdisch~160.3'C

You'll really need to get the LP side connected & sort out decent temp measurement info.

A few thoughts:
If your data is at all believable, it shows to me that you may be between a rock & a hard place in terms of excessive SH & SC. Adding charge may assist SH, but will bother SC - removing charge will have the opposite effect. If your info turns out to be close, then you'll probably want to also look at your cap tube design, 'evap' pipework & fan speed, condenser... a little more.

You must be having fun...

[Gary]

If we assume some further flashing after cap tube (typically is an additional 2-3K)

Te,sat~-16.8'C ; Tc,sat~55.6'C ;
SH=6.8-(-16.8)= 23.6K
SC=55.6-28.8=26.8K
Tdisch~160.3'C

Wouldn't 15 bar make Tc,sat ~41'C?

[Gary]

Wow, what a change! Looks like it was seriously over charged. I removed a load (sorry for the technical term) of refrigerant and now it's running with a head pressure of 15bar. I did calibrate my guages first.

The changes were dramatic. Previously there was frost/ice forming between the end of the evaporator and all down the compressor suction pipe. Now there is frost after the capillary and on the botttom rows of the evaporator coils. The temperatures have also changed dramatically. They are as follow:

Exit from compressor (measured at 150mm) 106'C, entry to the heat exchanger 84'C, exit of heat exchanger 28.8'C, exit from capillary (entry to evaporator) -13.8'C, exit from evaporator 6.8'C. All measurements taken with the case off which I'm sure makes a difference. Pressure at the one point I can measure, 15bar.

I tried reducing the water flow but as I don't have a bypass valve my poor pool pump started complaining so I don't think this is an option at the moment.

Just to confirm, the heat exchanger is configured to counter flow water against refrigerant.

Ambient temp? Pool water temp?

[desA]

Wouldn't 15 bar make Tc,sat ~41'C?

Quite correct ~ 41.78'C. Finger trouble - had yesterday's results stuck in me noggin.

Te,sat~-16.8'C ; Tc,sat~41.78'C ;
SH=6.8-(-16.8)= 23.6K
SC=41.78-28.8=12.98K
Tdisch~135.6'C

[Gary]

A few thoughts:
If your data is at all believable, it shows to me that you may be between a rock & a hard place in terms of excessive SH & SC. Adding charge may assist SH, but will bother SC - removing charge will have the opposite effect. If your info turns out to be close, then you'll probably want to also look at your cap tube design, 'evap' pipework & fan speed, condenser... a little more.

A rock & a hard place indeed. We may be able to compensate with fan speed for current conditions, but it all changes with warm weather. This system would be MUCH better off with a TXV.

[Gary]

You indicated in an earlier post that the fan was set to full speed. This implies that it's speed can be adjusted.

I would slow the fan until the entire coil is evenly frosted and the suction line is partially frosted, the frost ending before it reaches the compressor inlet.

[Gary]

On any condensing unit, the condenser is sized to handle both the heat of evaporation and the heat of compression. Since we are now using it as the evaporator, we can assume that it's capacity is such that it is capable of overwhelming and overloading the compressor. The capacity needs to be reduced by reducing the airflow through it, in order to properly match the compressor... or it will destroy the compressor in warm weather.

[Gary]

I am assuming this cap tube is also from the original system. This being the case, it is sized to provide proper refrigerant mass flow under warm (A/C) evaporator conditions. This makes it ideal for warm weather operations, but just as an A/C, if run under abnormally cold indoor conditions, in cold ambients it can flood the compressor. We can compensate for this by limiting the refrigerant charge, such that in cold ambients it runs low (but not dangerously low) superheat.

In warm weather, it will then run high superheat, but this is probably the best we can do.

[Acuario]

Des, I'm having great fun, just what I wanted from this project, and learning a lot!

Ambient temp this morning was 7.5'C, pool water temp 22'C. This was when I calibrated my guages so I remember it well!

The fan as standard has 2 speeds, I'll try and build a controller to make it variable. At the moment I'm getting frost/ice buildup on 2/3 of the evaporator; is this a good thing or is it better not to have it frosting up and hence restricting air flow? I've put a mechanical timer on the compressor to turn it off for 15 minutes every hour or so to let it defrost with the fan still running.

I take the point about the TXV, I'll see if my bits supplier has one. Any spec I need to look out for?

The capillary is the original one, as is basically everything. From what you say Gary, should I block some of the evaporator or just reduce the fan speed? I'm not planning to use this when there is hot weather as it doesn't need it. The plan is to just use it in winter and the solar panels in summer with a changeover during spring/autumn.

[Gary]

Try switching the fan to low speed and let's see what the pressure and all of the temperatures then look like.

[Gary]

At the moment I'm getting frost/ice buildup on 2/3 of the evaporator; is this a good thing or is it better not to have it frosting up and hence restricting air flow?

It is better not to have frost, but that is not going to happen. You will have frost in cold weather and we want the coil to frost fully and evenly.

[Gary]

I'm not planning to use this when there is hot weather as it doesn't need it. The plan is to just use it in winter and the solar panels in summer with a changeover during spring/autumn.

Excellent. That makes the fine tuning far less complicated.

[Gary]

With SCT at roughly 42C and water in temp at 22C, this gives us a cond TD of 20C. This is only borderline high. There may be hope for this system yet.

[Gary]

Given the stipulation that the system is not used in warm weather, we can run it up to full capacity without endangering the compressor... however... the condenser TD is already running borderline high with a light load. The condenser is the weak link in the chain. I would build another identical condenser and pipe it in parallel with the first. This would be a good thing even if we stayed with the light load.

And given increased condenser capacity, we could take steps to load this system up. We could install a TXV/wMOP and then crank the fan back up to full speed.

[Gary]

Hmmm... TXV sizing is not my forte. Perhaps others could jump in here.

I would guesstimate that the Fujitsu AOG25 capacity in A/C is about 7Kw... and in med/low temp operation would be about 3-4Kw. Am I close?

[Acuario]

The compressor is a Tecumseh AW5532A / AW108KT-027-B4, can't find any info on it but the AWA5532EXV is 9.144Kw, input 3Kw so maybe this is the current equivalent (verified it takes 3Kw by connecting an electric meter in series).

I'll build heat exchanger MK1 No2 and connect it in series as soon as I can.

Thanks to all for the help so far, it's been fun!

[MikeHolm]

Parallel, pipe it is parallel. I would think the pressure drop is important. Des, Gary?

[Gary]

Parallel, pipe it is parallel. I would think the pressure drop is important. Des, Gary?

Yes... it should be piped in parallel, not series.

[Gary]

Did you switch the fan to low speed?... take new pressure and temperature measurements?

[Acuario]

No time to do much today but I've fitted a valve in the LP side, recharged and took this photo of the guages. It's dark now so I can't do any more tonight. The ambient was around 9'C, water at 23'C.

Picture here:
http://www.costadoradaholiday.com/guages_1.jpg

[Acuario]

Parallel, pipe it is parallel. I would think the pressure drop is important. Des, Gary?

Sorry, typo - meant to say parallel - no space to put it in series!

[Acuario]

Parallel, pipe it is parallel. I would think the pressure drop is important. Des, Gary?

Sorry, typo, meant to say parallel - no room to fit one in series..

[Acuario]

Hi,
here are some results following addition of the valve to measure suction line pressure and after adding a bit of refrigerant to take it back to roughly what it was before adding the valve.

Ambient temperature: 10'C
Water temperature: 22'C

Fan full speed
HP pressure: 270psig
LP pressure: 48psig
Temp at compressor 150mm: 98.8'C
Temp at condenser input: 93.9'C
Temp at condenser output: 33.4'C
Temp at evaporator input: -4'C
Temp at evaporator output: 9.6'C
Frost on evaproator: about 30%

Fan slow speed
HP pressure: 245psig
LP pressure: 39psig
Temp at compressor 150mm: 117'C
Temp at condenser input: 106'C
Temp at condenser output: 31.8'C
Temp at evaporator input: -10'C
Temp at evaporator output: 9.2'C
Frost on evaporator: about 60%

Hope the figures make sense!

Nigel

[Gary]

Hi,
here are some results following addition of the valve to measure suction line pressure and after adding a bit of refrigerant to take it back to roughly what it was before adding the valve.

Ambient temperature: 10'C
Water temperature: 22'C

Fan full speed
HP pressure: 270psig
LP pressure: 48psig
Temp at compressor 150mm: 98.8'C
Temp at condenser input: 93.9'C
Temp at condenser output: 33.4'C
Temp at evaporator input: -4'C
Temp at evaporator output: 9.6'C
Frost on evaproator: about 30%

Fan slow speed
HP pressure: 245psig
LP pressure: 39psig
Temp at compressor 150mm: 117'C
Temp at condenser input: 106'C
Temp at condenser output: 31.8'C
Temp at evaporator input: -10'C
Temp at evaporator output: 9.2'C
Frost on evaporator: about 60%

Hope the figures make sense!

Nigel

Add refrigerant slowly until the evap is fully frosted. The frost can be on the suction line, but not all the way back to the compressor. Then show us the new readings.

[Acuario]

What fan speed, slow or full?

[frank]

Have you fitted a TEV yet or are you still trying to make it work with the capillary?

The compressor won't last long with those temps ....... 100C+

[Gary]

Have you fitted a TEV yet or are you still trying to make it work with the capillary?

The compressor won't last long with those temps ....... 100C+

I agree. Adding refrigerant will bring down the discharge temp, but it will also make the condenser even less efficient than it already is. Adding a TEV aka TXV will overload the condenser. This system needs expanded condenser capacity as well as a TEV.

Until then, the best we can do is add refrigerant and/or shorten the cap tube.

[Gary]

What fan speed, slow or full?

Unless/until you have added condenser capacity... slow the fan speed.

[desA]

Ambient temperature: 10'C
Water temperature: 22'C

Fan full speed
HP pressure: 270psig -> Tc,sat~ 50.46'C
LP pressure: 48psig -> Te,sat ~ -4.244'C
Temp at compressor 150mm: 98.8'C
Temp at condenser input: 93.9'C
Temp at condenser output: 33.4'C -> SC=17.06K
Temp at evaporator input: -4'C
Temp at evaporator output: 9.6'C -> SH=13.84K
Frost on evaproator: about 30%

Fan slow speed
HP pressure: 245psig -> Tc,sat ~ 46.49'C
LP pressure: 39psig -> Te,sat ~ -8.752'C
Temp at compressor 150mm: 117'C <========= !!!!
Temp at condenser input: 106'C
Temp at condenser output: 31.8'C -> SC=14.69K
Temp at evaporator input: -10'C
Temp at evaporator output: 9.2'C -> SH=17.95K
Frost on evaporator: about 60%

Hope the figures make sense!

Nigel

System seems to be overcharged wrt condenser, undercharged wrt evaporator. Add lots of extra surface to condenser pipe!!!! Then lets see what gives.

[desA]

Yes... it should be piped in parallel, not series.

I'd agree. Will need decent end manifolds to get decent distribution.

[Acuario]

Just bought the bits to make the second condenser so will hopefully make it tomorrow. Still using a capillary but my suppliers have TXV's so will get one and try to work out how to install it. I guess it replaces the capillary, the one he showed me had 2 ports, what looked to be a point to adjust it and what looked like a bulb on the end of a thin coil. Where does this go?
Nigel