# Screw the gas company

I only use gas for heating water, so I'm considering a solar water heater. It seems dumb not to go solar here in Arizona. I was wondering do any of you have a solar water heater? How do you like it? Do you have any plans on how to build one?

russ_watters
Mentor
I've never looked, but there are commecially available solar water heaters. No need to build one yourself (though it could be fun). Google it.

Something to consider:

If you have an attic, I have seen plans for a water heater that uses a fan to move attic air through a coil. Attics get pretty warm depending on the insulation arrangement. If the insulation is on the attic floor, the attic gets warm. If it's on the underside of the roof, the attic may not be too warm.

If you do have a warm attic, this kind of system can have two positive effects, cools the attic and warms the water in the coil. This may not provide enough heat to completely heat the water, but the combined effect of reducing air conditioning load, and preheating domestic water may allow for electric or gas backup for the water heating and still allow for savings.

This kind of system also works good for a swimming pool heater.

russ_watters
Mentor
I'd never heard that, Aartman, but its a lose-lose catch-22. In the summer you want to keep your attic as cool as possible to reduce your a/c bill.

Ivan Seeking
Staff Emeritus
Gold Member
Originally posted by Artman
Something to consider:

If you have an attic, I have seen plans for a water heater that uses a fan to move attic air through a coil. Attics get pretty warm depending on the insulation arrangement. If the insulation is on the attic floor, the attic gets warm. If it's on the underside of the roof, the attic may not be too warm.

If you do have a warm attic, this kind of system can have two positive effects, cools the attic and warms the water in the coil. This may not provide enough heat to completely heat the water, but the combined effect of reducing air conditioning load, and preheating domestic water may allow for electric or gas backup for the water heating and still allow for savings.

This kind of system also works good for a swimming pool heater.

One needs to be careful with ideas like this since the costs can exceed the lifetime benefit. The problem is the heat exchange efficiency. The rate of heat transfer through a separating wall increases as the velocity, density, specific heat, and the thermal conductivity of the fluids increase. The state of the fluids inside and outside of the thermal conductor is also significant. For example, liquid to liquid transfers are more efficient that gas to liquid.

Here are some numbers from an old engineering book that I use; Refrigeration, Air Conditioning, and Cold Storage; by Gunther, 1969 [it was old when I got it]. This was long considered the bible of heat transfer applications.

For a well designed industrial heat exchanger, using clean copper pipes and with low rates of fluid flow [e.g. < 1 gpm liquid flow per foot of linear contact between the media and the copper pipes, with air moving by a low power fan - air to water transfers], we typically get something like 2 BTU per hr per sq. ft of contact per degree F – a best case scenario.

So if we use some typical 3/4" copper pipe, we get about 0.19 sq ft of contact per ft of pipe. We might expect a temperature differential of no more than 30 degrees F at the inlet, and 10 degrees at the outlet in order to be useful at the other end. So we can loosely assume an average of a 20 degree difference [not really but the error works in your favor]. Thus we get about 0.4 BTU per hr per linear foot of 3/4" pipe. The pipe size was chosen to make the system practical. Smaller pipe requires greater length; stay tuned.

Next, 1 BTU per hr is about 0.29 watts [CRC]; which yields about 0.12 watts per linear foot of exchange. So in order to rival the contribution of a 100 watt light bulb, we need at least 830 feet of 3/4” copper tubing. The current price of type L copper [common] is about $2.25 per foot. So for only$1,900.00, just for the pipe mind you, you still need a pump, controls, wiring, a fan, these need to then be powered which may take more than 100 watts, you can have the heat of a 100 watt light bulb for about half the day.

At ten cents per KWH, this would only require 43 years to pay for the pipe.

My point: If these types of systems are to be of use they must be properly engineered.

Last edited:
Ivan Seeking
Staff Emeritus
Gold Member

One thing that I have done is to use the cool air from underneath my house for free A/C. This can be dangerous if any molds or toxins are present, and the smell requires the use of good carbon filters, but I did this for several years. I found it to be effective in my house for up to 5 days running, in 100 degree F weather. Edit: Note that in fact we are using the heat capacity of the earth, the footer(s), and the structure of the house itself for heat storage. For every 1000 sq feet of house, we get about 2500 sq ft of heat exchange surfaces for heat storage.

I blocked all but one vent [located on the NE corner of the house] for the crawlspace. Then, using a window style box fan, I just blew the air from under the house through some good filters, and directly into the house. Every day the temp under the house rose - after a week of very hot weather it ceases to be very effective - but for 5 days the blowing air stayed pretty close to 70 degrees most of the day. I estimated that this all equated to a 5000 BTU A/C unit [1500 watts] running for 6 hours a day. In moderate weather – in Oregon – this can work all summer long.

Note that the air was (fortuitously) forced to travel a long path under the house before getting to the fan. This is very important. Also, by continuing to run the fan all evening, the cool night air helps to "recharge" the crawl space for the next day. If it never cools down at night the effectiveness tapers off very quickly.

The elimination of heat exchangers can increase the efficiency of a system tremendously; sometimes even making something once useless, practical.

A properly designed building could increase the effectiveness of this concept.

Last edited:
In a 105 deg F attic these type units exchange approximately 3 tons of heat (depending on the attic size and inlet water temperature) @ 2500 CFM.

Also, an attic in Arizona is going to be closer to 110 or 115 deg F much of the year, giving a wider temperature difference for a better rate of exchange.

I suggest this as a preheater, with the cooling of the attic as a side benefit raising the overall efficiency, but with 115 deg F or higher air temperatures, water circulating through this fan coil system will continue to rise in temperature until it reaches equilibrium with the air temperature at 115 deg F, which is hot enough for most domestic uses.

A quick look at the ratings of a fan coil unit will verify that a 20 deg F rise in water temperature is easily optainable over 2 or 3 gpm even at as low as 1000 CFM using just a 2 row cooling coil and a small fractional HP fan with a difference in entering air to entering water temperture of only 35 deg F. (80 deg F DB EAT and 45 deg F EWT) We have that with 75 EWT and 110 EAT increase the air flow and the water temperature rise will increase (the air temp drop will decrease).

So how long will the payback take with 36,000 btu gained by the water, AND 36,000 btu removed from the attic minus a 1/25 hp circulator (at most) and a 1/8 hp fan (at most)? By my calculations, 29 watts for the circulator and 93 watts for the fan is only about 122 watts
according to your figures 1 btu is .29 watts so 36,000 is 10440 watts saved per hour x 2 (we are saving 36000 in AC and gaining 36000 in water heat) for a total of 20880 x say 75% eff loss (won't be this high, but it works in your favor) = 15660 watts per hour (during peak attic heat).

At 10 cents per KWH this arrangement saves roughly $1.50 every hour it runs. If it costs$5000.00 to install the entire payback will be 3333.3 hrs of operation or probably a year or two of water heater operation.

These things are already in use in areas with warm climates. They can completely cover the water heating load during summer months.

Ivan Seeking
Staff Emeritus
Gold Member
Originally posted by Artman
In a 105 deg F attic these type units exchange approximately 3 tons of heat (depending on the attic size and inlet water temperature) @ 2500 CFM.

Also, an attic in Arizona is going to be closer to 110 or 115 deg F much of the year, giving a wider temperature difference for a better rate of exchange.

I suggest this as a preheater, with the cooling of the attic as a side benefit raising the overall efficiency, but with 115 deg F or higher air temperatures, water circulating through this fan coil system will continue to rise in temperature until it reaches equilibrium with the air temperature at 115 deg F, which is hot enough for most domestic uses.

A quick look at the ratings of a fan coil unit will verify that a 20 deg F rise in water temperature is easily optainable over 2 or 3 gpm even at as low as 1000 CFM using just a 2 row cooling coil and a small fractional HP fan with a difference in entering air to entering water temperture of only 35 deg F. (80 deg F DB EAT and 45 deg F EWT) We have that with 75 EWT and 110 EAT increase the air flow and the water temperature rise will increase (the air temp drop will decrease).

So how long will the payback take with 36,000 btu gained by the water, AND 36,000 btu removed from the attic minus a 1/25 hp circulator (at most) and a 1/8 hp fan (at most)? By my calculations, 29 watts for the circulator and 93 watts for the fan is only about 122 watts
according to your figures 1 btu is .29 watts so 36,000 is 10440 watts saved per hour x 2 (we are saving 36000 in AC and gaining 36000 in water heat) for a total of 20880 x say 75% eff loss (won't be this high, but it works in your favor) = 15660 watts per hour (during peak attic heat).

At 10 cents per KWH this arrangement saves roughly $1.50 every hour it runs. If it costs$5000.00 to install the entire payback will be 3333.3 hrs of operation or probably a year or two of water heater operation.

These things are already in use in areas with warm climates. They can completely cover the water heating load during summer months.

I don't see how this is possible. Where do you get this information?
Also, your numbers are only good a few months out of the year. Even using these numbers, the two years is probably more like 8 or 10.

Next, if we consider the total heat stored in that attic air at any moment, say at 115 degrees F, using 0.017 BTU per cu ft per degree F for dry 100 degree air, and using a 2000 sq foot house with an average of 4 ft overhead - 8000 cu ft - over the span of 100 to 115 degrees, we have a maximum of 2040 BTUs available.

Last edited:
Ivan,
It is basically how a chilled water coil works. The fan draws entering air aproximately 80 DB across a coil holding 45 deg water and lowers the air temperature down to 55 or 56 DB and raises the temperature of the water 10 deg or more based on the flow rate.

As long as the temperature differences are similar, and the fluid properties are similar, the amount of btu's exchanged will be similar.

LAT = (Mbh/(CFM * .00108)) +EAT

GPM = ((MBH*1000) / ((EWT-LWT) * SpHtW * 60 * 8.33))

Even a standard hot water baseboard with no fan gets about 300 btu/LF using 1 gpm of 140 deg F water and 65 deg air with only convection moving the air. You get much higher exchange rate with a fan.

I may be over-estimating the savings a bit, but I think you may be under-estimating a bit more.

Ivan Seeking
Staff Emeritus
Gold Member
Originally posted by Artman
Ivan,
It is basically how a chilled water coil works. The fan draws entering air aproximately 80 DB across a coil holding 45 deg water and lowers the air temperature down to 55 or 56 DB and raises the temperature of the water 10 deg or more based on the flow rate.

As long as the temperature differences are similar, and the fluid properties are similar, the amount of btu's exchanged will be similar.

LAT = (Mbh/(CFM * .00108)) +EAT

GPM = ((MBH*1000) / ((EWT-LWT) * SpHtW * 60 * 8.33))

Even a standard hot water baseboard with no fan gets about 300 btu/LF using 1 gpm of 140 deg F water and 65 deg air with only convection moving the air. You get much higher exchange rate with a fan.

I may be over-estimating the savings a bit, but I think you may be under-estimating a bit more.

First, I would need some convincing to believe this is really effective as indicated. I have no doubt that people think it is, but still, I'm skeptical. On the other hand, don't misunderstand; I was speaking to the need for caution. I have seen people lose a lot of money by slapping these kind of systems together. Note that I purposely did not even allow for radiator fins. I presented a typical scenario similar to examples that I have seen. If these systems are properly engineered, I am the biggest fan [forgive the pun] of alternative approaches. Unfortunately, there are many systems sold and used that are a waste of time; and way too much money! I don't mean to be too cynical [or skeptical], but when it comes to thousands of dollars potentially wasted, I get pretty cynical.

Obviously in Arizona you enjoy much of a best case scenario.

Here in Oregon, I did a lot of research on low head hydro. [I have a creek that ranges from 20 to 100 cfs; about 8 months of the year]. Since I had an interest in tapping this energy, if practical, I did a lot of research. My wife ran across a retired gentleman, then I would guess in his late sixties, who had build a pontoon, put it in the river, and sat a giant undershot paddle wheel on top. This was to power his house. It was absolutely huge; I am thinking it was about 10 feet in diameter, and 20 feet long.. He of course expected to sell these things to everyone in who lives near a river. He effectively spent his life’s savings on this device – he had to buy a small crane in order to move it around. He wanted to show it off, but also see why he could only get a couple hundred watts out of this monster. I could hear the 4x8 ft sheets of plywood slapping the water as I approached his house. I didn’t have the heart to tell this guy that his $20,000 was mostly wasted. . This is where a half*ssed approach can get you. If good science and engineering are used in developing these systems, then you’ll find me leading the parade. Ivan Seeking Staff Emeritus Science Advisor Gold Member Originally posted by Artman Even a standard hot water baseboard with no fan gets about 300 btu/LF using 1 gpm of 140 deg F water and 65 deg air with only convection moving the air. You get much higher exchange rate with a fan. using the 2 BTU per sq ft per degrees F, we get 150 BTU per sq ft. Since a baseboard goes as about 1 sq ft per ft, we might expect 150 BTU per ft by this estimate. Also the exchange is more effecient due to the relative increase in water velocity. I think this accounts for the rest. Edit: I just checked my Lakewood oil radiator. The surfaces have nearly exactly 1 sq ft per ft. At full throttle, the oil temp is about 200 degrees F. The total surface area is 14 sq ft [and linear feet]. The heat transfer is about 1500 watts x 0.8 = 1200 [you don't get to count the electrical heat loss ], giving 4100 BTU per hour. (2 BTU/sqft/F)x(14 sq ft)x(130 degrees F) = 3640 BTU /hr Looks pretty close. These systems have a very low volume to surface area ratio, this also affects the results. I would agree that this number of 2 BTU does appear a bit conservative when it comes to a properly designed radiator. Last edited: Originally posted by Ivan Seeking First, I would need some convincing to believe this is really effective as indicated. I have no doubt that people think it is, but still, I'm skeptical. On the other hand, don't misunderstand; I was speaking to the need for caution. I have seen people lose a lot of money by slapping these kind of systems together. Note that I purposely did not even allow for radiator fins. I presented a typical scenario similar to examples that I have seen. If these systems are properly engineered, I am the biggest fan [forgive the pun] of alternative approaches. Unfortunately, there are many systems sold and used that are a waste of time; and way too much money! I don't mean to be too cynical [or skeptical], but when it comes to thousands of dollars potentially wasted, I get pretty cynical. Obviously in Arizona you enjoy much of a best case scenario. Here in Oregon, I did a lot of research on low head hydro. [I have a creek that ranges from 20 to 100 cfs; about 8 months of the year]. Since I had an interest in tapping this energy, if practical, I did a lot of research. My wife ran across a retired gentleman, then I would guess in his late sixties, who had build a pontoon, put it in the river, and sat a giant undershot paddle wheel on top. This was to power his house. It was absolutely huge; I am thinking it was about 10 feet in diameter, and 20 feet long.. He of course expected to sell these things to everyone in who lives near a river. He effectively spent his life’s savings on this device – he had to buy a small crane in order to move it around. He wanted to show it off, but also see why he could only get a couple hundred watts out of this monster. I could hear the 4x8 ft sheets of plywood slapping the water as I approached his house. I didn’t have the heart to tell this guy that his$20,000 was mostly wasted. .

This is where a half*ssed approach can get you. If good science and engineering are used in developing these systems, then you’ll find me leading the parade.

I agree. There are also code issues, plumbing codes, building codes, lifesafety questions (the attic system is weighty and may be considered a pressure vessel), and should not be rushed into without research and study.

I have thought about using the attic heater approach to heat a small pool, but there are many dangers with putting chlorinated water in such a system and then putting it into your attic (many corrosion factors to consider)and I don't want to put out the cost just to heat my pool.

Doc
Don't sit around arguing about it, go here:

http://www.otherpower.com/

They have alot of alternative energy solutions and are very informative. There is a water wheel device shown also.

russ_watters
Mentor
Originally posted by Ivan Seeking
One thing that I have done is to use the cool air from underneath my house for free A/C. This can be dangerous if any molds or toxins are present, and the smell requires the use of good carbon filters, but I did this for several years.
Interesting idea, but you have to be VERY careful. Besides mold, the humidity itself can cause you to INCREASE the cooling load on the house even if the air entering the house is below room temp. The vast majority of the energy your air conditioner uses (depending on climate of course - and the northwest is humid) goes toward DE-HUMIDIFICATION, not simply cooling. I'd have to do the psychometrics to know for sure what the maximum temp/humidity you could benefit from would be.
At 10 cents per KWH this arrangement saves roughly $1.50 every hour it runs. Even a standard hot water baseboard.... I just checked my Lakewood oil radiator. Whoa, whoa, whoa, easy there guys. Radiators? Boilers? Fan coils? You're missing the big picture here. Usage rates and thermodynamics are interesting, but the savings realized is based on hours of use. You guys are assuming it is always running flat out. A heat recovery system of this type may very well save$1.50 an hour - but only for that hour right after you take your shower and run the dishwasher. The rest of the day, your hot water heater uses only about 100W. ~$.015 / hr. THAT is why it is not economically viable. Think about it - saving you$1.50 an hour is $1080 a month. What is your electric bill these days? Last edited: Ivan Seeking Staff Emeritus Science Advisor Gold Member Originally posted by russ_watters Interesting idea, but you have to be VERY careful. Besides mold, the humidity itself can cause you to INCREASE the cooling load on the house even if the air entering the house is below room temp. The vast majority of the energy your air conditioner uses (depending on climate of course - and the northwest is humid) goes toward DE-HUMIDIFICATION, not simply cooling. I'd have to do the psychometrics to know for sure what the maximum temp/humidity you could benefit from would be. True. Still, in my case it worked great within the limits cited. Whoa, whoa, whoa, easy there guys. Radiators? Boilers? Fan coils? You're missing the big picture here. Usage rates and thermodynamics are interesting, but the savings realized is based on hours of use. You guys are assuming it is always running flat out. A heat recovery system of this type may very well save$1.50 an hour - but only for that hour right after you take your shower and run the dishwasher. The rest of the day, your hot water heater uses only about 100W. ~$.015 / hr. THAT is why it is not economically viable. Think about it - saving you$1.50 an hour is $1080 a month. What is your electric bill these days? Well, I don't recall anyone citing$1000 per month benefit. Also, if the system is properly designed, which requires the use of a second insulated water tank, the practical problems that you cite can be accomodated...at least mostly. Again though, this gets into more cost with a longer period for your return on investment. This is the trap that many people and systems fall into.

Have I mentioned the Hydrogen alternative?

russ_watters
Mentor
Originally posted by Ivan Seeking
Well, I don't recall anyone citing $1000 per month benefit. Aartman said$1.50 /hr. Actually though he also said 1-2 year payback for 3333 hrs use. At 2 years, thats 9hr/day or $410 / month. In any case, that way off how a water heater is actually used - an order of magnitude at least. I only use water for 1 person in an apartment, but my TOTAL electric bill for a month in the spring is <$20. Thats hot water, refrigerator, my computer, and my entertainment center. The hot water can't be more than $5 of that (which would be an average of 55W). And I average about 1.5 showers a day since I work out. I also have the thermostat on the water heater all the way up. Last edited: Ivan Seeking Staff Emeritus Science Advisor Gold Member Originally posted by russ_watters Aartman said$1.50 /hr. Actually though he also said 1-2 year payback for 3333 hrs use. At 2 years, thats 9hr/day or $410 / month. In any case, that way off how a water heater is actually used - an order of magnitude at least. posted by IvanEven using these numbers, the two years is probably more like 8 or 10. I agree...maybe more like 20 years. I only use water for 1 person in an apartment, but my TOTAL electric bill for a month in the spring is <$20. Thats hot water, refrigerator, my computer, and my entertainment center. The hot water can't be more than $5 of that (which would be an average of 55W). And I average about 1.5 showers a day since I work out. I also have the thermostat on the water heater all the way up. This brings up an interesting point. The maximum momentary energy demand is often quite different - by up to 5 times or more - from the average demand. Distribution of demand makes alternative sources much more viable. I have played with PWM on the heating coils for a full sized space heater. Instead of running the 20,000 watt heat load for a few minutes 10 times a day, the idea was to modulate one of the four coils as a variable load. This would ensure the maximum power transfer from the hydro generator [about 3000 watts continuous edit: but widely variable over days and weeks]. The regular coils and increased fan speed would kick in as needed. Also, by increasing the max temp on the water heater, we could dump excess energy into this system; if properly coupled. I have been playing with the hydro stuff for years but I have been afraid to spend the money for a dam. It is a risky investment. By the way, if I run electric heat, I can easily spend$150 a month on heat and hot water. For a time we had a large house, in excess of 2800 sq ft, that could fetch a $300 electric bill each month in mid winter. In very cold areas of the US, the equivalent cost of electric heat is hidden in the prices of cheap oil heat. Have I mentioned the real energy costs in a gallon of oil? Last edited: Originally posted by russ_watters Aartman said$1.50 /hr. Actually though he also said 1-2 year payback for 3333 hrs use. At 2 years, thats 9hr/day or $410 / month. In any case, that way off how a water heater is actually used - an order of magnitude at least. I only use water for 1 person in an apartment, but my TOTAL electric bill for a month in the spring is <$20. Thats hot water, refrigerator, my computer, and my entertainment center. The hot water can't be more than $5 of that (which would be an average of 55W). And I average about 1.5 showers a day since I work out. I also have the thermostat on the water heater all the way up. True, the most you could save has to be less than you spend for water heating now. However, I believe I overestimated the cost as well. Insulated storage tank, fan coil unit, circulator, some piping and valves, electrical connections, controls, if he installs it all himself shouldn't cost$5000.00. Also, the projected savings reduce as the attic cools and the water gets hotter (the temperature difference reduces between the EWT and EAT). I should have worked a few of the numbers before posting the response.

However, you do get the benefit of "free cooling" of the attic space. This was included in my savings number.

I like this system because you are transferring your heat from an unwanted heat source and in the process possibly decreasing the load on the home's AC system. With Solar panels, you block some of the sun from striking the roof, if your panels are mounted on the roof, but probably not significantly effecting the cooling load on the house. I like systems that provide multiple benefits.

Actually another way to look at this is that you are air conditioning your attic and in the process getting some free hot water.