Results and Data from Simple Cooling System Prototype

I suspect you are going to find that the efficacy of the system will decrease when you are working at more "normal" temperatures. Heat transfer is going to depend on the temperature difference between the fluid and the heat sink. Notice how your curve in the first graph tends to flatten out as temperatures decline. One thing that may help might be more consideration over your diffuser arrangement. If you can get smaller air bubbles, you increase the heat transfer surface area. With regards to getting the heat transferred from the water to the ground- I am not sure your test procedure really defines your actual operating constraints...What would be nice would be some numbers on water temperature, both day and night. You obviously have not exceeded your heat capacity in the water tank yet, but how does the water temperature build up over the day? Does it come back down at night?

I like your air pump- but I would have done it with duct tape...

> Notice how your curve in the first graph tends to flatten out as temperatures decline.

Yes, but if the temperature gradient was insufficient to push the heat through there'd be a rise of at least a couple of degrees. Here the air is coming out bang on the ground temp of 24.5

> One thing that may help might be more consideration over your diffuser arrangement.

I thought this would be more of a concern, but am actually surprised how well my initial attempt seems to have worked. Like I say, the air was coming out at the water/ground temp, even when it was going in at up to 55 C, so I think I might even be able to back off it a little, and maybe have no diffuser at all and a shallower inlet, which would be great for reducing resistance to the air flow.

> What would be nice would be some numbers on water temperature

At the time of this test, 24.5 C, no daily variation.

http://urbangreenhouse.blogspot.com/2011/01/ground-temperature-data.html

> how does the water temperature build up over the day?

It didn't during the 8 hours I ran this test, and you probably wouldn't run it longer than that in a day. If there was buildup during the day the rate of loss overnight still remains to be seen, and would be entirely dependant on your soil type.

Duct tape would've been better, yes, as the stuff I used was kind of stretchy. But I ran out. Hence the string.

Science!

PS. It is great to see someone post actual test data here...

This is a repost from the old thread but since it's related to the test data I thought it would be prudent to post it in this thread too.

The results of the first experiment are pretty much what I would have expected to see and if you know the surface area of the water that was in contact with the walls of the container you should be able to work out the thermal conductivity of the system which you would then be able to use to calculate how much and fast the heat could be dumped into the surrounding earth.

However, in experiment 2 there is a major problem and that is the humidity of the air returning to the house from the water container. What you have there is a pretty good evaporative cooler rather than heat exchanger and you need to take into account the enthalpy of the air going into the container with the enthalpy of the air coming out to get an idea of the amount of energy being transferred. Otherwise you are not comparing apples with apples.

What happens when you bubble hot air through cold water is the water evaporates into the air and cools it due to the latent heat of vaporization but raises the relative humidity of the air in the process. In a closed system like yours this is ultimately going to cause a problem because the air will soon become saturated with water and you will end up with much less heat transfer than in your experiment .

Actually thinking about it evaporation of the water into the air as it's bubbled through is going to be a major problem. In the beginning you will get what appears to be really effective cooling, that's till the air that is being recycled becomes saturated with water and then and only then will you start transferring heat into the water. This is going to be a big problem because plants need to be able to evaporate water from their leaves so while they like humid climates they can't cope with a climate that has a relative humidity of 100% all the time. You will also have problem with condensation overnight when the outside air temperature drops, not to mention building the perfect environment for the growth of moulds and or bacteria.

However, I think there is a solution and that something akin to the way a boiler or heat exchanger works. Basically you have container of water that has lots of pipes running through it. You then blow the air through these pipes and transfer the heat energy to the water which is then both temporarily stored in the water and transferred to the surrounding earth whenever the soil temperature is lower than the water temperature. This way the water never comes in direct contact with the air so you don't have a problem with evaporation. In a low pressure system like this copper would be the best material to construct both the heat exchanger and buried container because it has both a very good thermal conductivity and resistance to corrosion

It's going to make the system somewhat more complex and it may be simpler to just lay a grid of polly pipes in the ground that you blow the air through directly, but if you have the air coming into direct contact with the water you are going to have evaporation and while that will cool the air initially in a closed system it is going to have problems with air that is saturated with water.

I must apologize for not thinking of this earlier as it's going to have a major effect on the design of your system, but I still think experiment 1 shows that you have the basics of a workable system. You're just going to have to work out a different way of transferring the energy in the air to the water as bubbling it through isn't going to work.

"Keep in mind that the 50-55 C air was a result of the motor, and in a real setting you'd probably be looking about mid thirties at most, and cooling during the day as the system takes effect."

This is true but given the output temperature ranges through only 1°C while the input temperature varies through 9.5°C tend to lead me to believe that the cooling effect you are seeing is primarily due to the evaporation of the water rather than the transfer of heat into the water.

Keep in mind the most important factor here is not the temperature of the air but the enthalpy which is a measure of the total energy content of the air. If you have air coming out that is at 90% humidity which is I suspect the case then you are getting very little change in the enthalpy of the air and therefore little transfer of the heat into the water.

Unfortunately there is no way to get around this, but you will have to redo experiment 2 except this time monitoring the following:

• Air input temperature
• Air input dew point
• Air output temperature
• Air output dew point
• Water temperature
• Air flow rate
• Water volume

I know this sounds like a lot of things to monitor but they all have a part to play and with them and a psychrometric chart you will be able to work out how much energy is being transferred into the water and how much of the cooling is due to evaporation.

Mind you, evaporative cooling isn't all that bad as it is a very efficient way of cooling things down, it just requires you to vent the hot dry air and replace it with the cool moist air. If you have ever seen the greenhouses north of Adelaide where they grow things like tomatoes and roses year round you will find that they use evaporative cooling systems that vent the hot air straight into the atmosphere and replace it with cool humid air.

Getting back to experiment 1, which by the way I believe was a resounding success, if you can give me the exact dimensions of the container then I can calculate the coefficient of thermal conductivity that you got which you can then use when you scale everything up to calculate how much and fast you will be able to dump energy into the surrounding earth.

Keep in mind, I still think you have the basis of a workable system, it's just going to be slightly more complex that you first thought, but then that's pretty much what always happens whenever you start designing something from scratch.

Just a quick reply til I get a chance to respond properly; the tank was a cylinder of diameter about 27 cm and about 35 cm high. This gives a total surface of 4100 cm^2.

It was thin steel, the soil was dry unpacked sand on a sandstone base and the area was in shade throughout the day.

> while the input temperature varies through 9.5°C tend to lead me to believe that the cooling effect you are seeing is > primarily due to the evaporation of the water rather than the transfer of heat into the water.

Possibly, or that the heat is being transmitted into the ground faster than it's building up in the water.

I had a thought about evaporation, in that it'd be limited to how much energy is in the air. 1 litre, say, of air at 55 C would have a mass of of 1.28 grams and 38.4 Joules more than at 25 C. This is enough energy to evaporate about 0.02 ml of water, which would create 34 ml of water vapour. I frankly can't figure out how to calculate how much of an increase in relative humidity this would be in 25 C air, but in absolute terms it represents 3.4% of the air volume.
And this doesn't take into account the energy required to raise the temperature of the water or how much vapour would recondense.
And at 37 C this would be halved.

> you can expect to radiate about 10 watts of power into the surrounding soil per square metre of surface area per degree of > temperature difference between the soil and the water in the radiating container.

Which at about .4 square meters for the tank would be 4 joules per second per Kelvin, and according to my previous energy calculation, at 2 l per second there should be about 40 joules per second coming in.
So either the water is 10 degress hotter than the soil, which it wasn't, or the energy is leaving in some other fashion, which I guess is the basis of your increased humidity hypothesis.
Hmmm...

> there is no allowance here for the soil heating up which is going to decrease your temperature differential

If that happened I'd just use a bigger water container, til it stopped.

Thanks for taking the time to work all this out.

That was well written Masu, GA.

Not only are you sorting out your better half's paperwork work but have reorientated, sorted a few things for me as well.

With such little investment in hardware and set up everybody should have one. I suppose it could be scaled up easily. The distribution of the air and the pump will chew a bit of energy but in comparison to other methods of cooling it has merits.

In combination with something I am working on I would surely introduce it to any future house I would build.

Thanks to the original poster for putting this up, very interesting, Ky.

Hi Charlie

Hope all is well in your neck of the woods.

I was told by all and sundry that this would not work in the tropics but it worked and never mind the humidity. I was the inventor, manufacturer and distributor 20 years ago. Sold the idea and they thought to use different (cheaper) materials and altered some other designs. Very sad for me because I had done so much work to get it to that stage and receiving royalties was out of the question.

Death by bean counter, Ky.

Thanks again for the excellent feedback.

I'm probably not going to get to do too much more on this in the short term; I'm just starting in on the first solarflower workshop, while simultaneously building the greenhouse prototype and somehow getting the website up. Also am in New Zealand, where cooling of growing spaces and homes isn't such an issue.

At some point tho I will test this humidification issue, Masu your idea of measuring water volume loss is a good one and I'll also see if I can get use of a RH meter or similar.

The prototype is back in Australia keeping my mum's house nice and cool, apparently.

Regarding the pool cooler, how does that work?

Regarding energy use of this system, ideally the air pump would be powered mechanically by a wind turbine, with an electrical backup for low wind periods. I looked briefly into the operating rpm of vacuum cleaner air pumps and horizontal axis wind turbines, I think it should work ok.

A nice side draft cooler mounted close to ground level

in conjunction with a tall peak mounted vent will do quite a bit of cooling with just a small water pump to keep the pads wet

The taller the vent the more natural flow

throw in a water treatment block every month or so

I run a similar unit

120 hours of running the blower for 20 bucks a month, the water pump is less than $5 a month

I had some further thoughts on the evaporative cooling component in experiment 2 and how to calculate how much of the cooling was due to evaporation.

But first up we need to know a bit about relative and absolute humidity. Absolute humidity is the amount of water dissolved in the air while the relative humidity is a measure of how saturated the water is with water vapour. Now the interesting thing about absolute humidity is that provided you don't have a major weather front go through the absolute humidity remains relatively stable while the relative humidity varies throughout the day as the temperature rises and falls.

So, how does that relate to the experiment I hear you say?

Well, if I knew the temperature and relative humidity sometime during the day that you did your experiment then and that there was no major weather front moving through I could calculate the absolute humidity or water content of the ambient air and from there calculate how much more water the air could absorb as it went through your simple heat exchanger.

Now you mentioned that you lived in the Nelson Bay region of NSW Australia and in your post on 30^th March that contained the experimental results you said that you did the experiment about a week earlier. So, I went to the Australian Bureau of Meteorology's web site and looked up the records for Nelson Bay in March 2011 and looked at the readings about a week before the 30^th.

Now I was hoping that the weather around then was stable because I could then deduce the absolute humidity or moisture content of the air, but guess what? Yes you guessed it, a bloody great front went through the region some time on the 23^rd March and the relative humidity and temperatures varied dramatically over the period in question.

But, we're not totally snafued yet as if you can remember the date or even the day of the week on which you did the experiment I can use the BoM records for the exact day to calculate the moisture content of the ambient air and from there figure out how much the evaporative cooling effect was.

So it all comes back to you at the moment, can you remember the date or day of the week that you did your experiments?

You sir, are a legend. I'm just going to put that out there.

re historical humidity data, I was in Fremantle Western Australia at the time, currently I'm in the Nelson Bays area, NZ.

The date was the day I flew out, which (*checks) was the 25th of March. There's a site (I think it's Nine MSN weather) which has data records on this, I'll check into it when I get the chance (madly rushed with giving this solar workshop at the moment).

Keep in mind tho that the air going into the system was tapped off the second storey of the house, so not being from directly outside it may have a different water content...

And I'm pretty sure there were no fronts went over that day.

"Keep in mind tho that the air going into the system was tapped off the second storey of the house, so not being from directly outside it may have a different water content..."

You'd be surprized at how constant the water content in the air is provided nothing major happens with the weather pushing a new block of air through like you see with cold and warm fronts moving through.

While the relative humidity in the roof would be much lower than in the atmosphere this is primarily due to the higher temperature and the moisture content would be fairly close to that in the atmosphere. It might be a little dryer due to it being trapped, but your experiment removed some 57 m³ of air from the roof cavity which would have been replaced with the outside air.

So, for the purposes of your experiment using the moisture content of the outside air would more than likely be close enough, at least for a first analysis anyway.

In the meantime I'll see if I can get the records for Freemantle for the period around the 24^th March and see what pops up.

Regarding the pool cooler, how does that work?

Hi saf

Try this at home.

Take some hot water and spray it through a fine nozzle, like a child's water pistol, check the temperature at the outlet and a meter and then 2m away from the outlet. The water will cool down while traveling through the air. You can have the same experience when you take a shower. You will notice that the water leaving the spout is hotter than the water that traveled all the way to your feet.

The rest is history, Ky.