Water-to-water Heat Exchanger. Wow, soft copper pipe is expensive

[turbotech]

With the current prices on soft copper pipe I am not excited about making a heat exchanger for a storage tank. I could go with a aluminum heat exchanger and sacrificial rod, but that requires more maintenance and more failure possibilities.
It looks like a 3/4" x 60' coil is about $150 and need 4 coils comes out to $600 to make two 120' coils. That has a surface area of 105 sq inches. That is crazy to me. A copper sheet measuring 12"x24" and is .080" thick costs $110. I could TIG weld a box that is 4" x 4" x 12" using the plate to get 192 sq in total. That is a higher surface area than two 180' coils which is 160 square inches.

The box would be equivalent to two 3/4" coils that are 218' long each!

I am thinking of going the copper box route. Any thoughts on why it would be a bad idea?

EDIT: Corrected box size

 

[SmokeyTheBear]

With the current prices on soft copper pipe I am not excited about making a heat exchanger for a storage tank. I could go with a aluminum heat exchanger and sacrificial rod, but that requires more maintenance and more failure possibilities.
It looks like a 3/4" x 60' coil is about $150 and need 4 coils comes out to $600 to make two 120' coils. That has a surface area of 105 sq inches. That is crazy to me. A copper sheet measuring 12"x24" and is .080" thick costs $110. I could TIG weld a box that is 4" x 4" x 12" using the plate to get 192 sq in total. That is a higher surface area than two 180' coils which is 160 square inches.

The box would be equivalent to two 3/4" coils that are 218' long each!

I am thinking of going the copper box route. Any thoughts on why it would be a bad idea?

EDIT: Corrected box size

You might want to redo your 60' copper tube coil as the exterior surface area is (assuming the 3/4 is an OD measurement) 3.14159265 x .75 x 60 x 12 (1696+) square inches and that is much larger than the 192 square inch box.

 

[turbotech]

I just realized that when I thought about the real world size. I forgot to do the times 12 for inches. It comes out to 1,271 sq inches. Your equation is has .75 where it should have (.75/2)^2. Thank you for checking this for me. I will see what a larger sheet costs to make a 12 x 12 x 12 box.

 

[afblue]

I am not understanding your math. 3/4"x60' to me is 1697 sq inches of surface area. (pi)dh=3.14*.75*720=1696.46 sq inches

I do agree with you copper is rediculously expensive. I imagine 50 years from now when I pass away, my grandchildren are going to gut my house and make a fortune

 

[SmokeyTheBear]

I just realized that when I thought about the real world size. I forgot to do the times 12 for inches. It comes out to 1,271 sq inches. Your equation is has .75 where it should have (.75/2)^2. Thank you for checking this for me. I will see what a larger sheet costs to make a 12 x 12 x 12 box.

Your calculation for the surface area of the cylinder should be pi x diameter x length or circumference x length.

 

[turbotech]

I think I am losing it. You two are right in that it is pi*D. I posted the cubic area. I will try again.

 

[tom in maine]

Copper is expensive. But experimenting on heat exchangers is more expensive.
Copper tubing is the best bang for the buck. I have been doing this for over 30 years.
Smooth copper does it the best for the range of temperatures that you need to work with in thermal storage systems.

 

[turbotech]

Copper is expensive. But experimenting on heat exchangers is more expensive.
Copper tubing is the best bang for the buck. I have been doing this for over 30 years.
Smooth copper does it the best for the range of temperatures that you need to work with in thermal storage systems.

I think you are right. The cost for the sheet would be much more expensive than using the copper pipe coils. I guess I had a case of sticker shock and wishful thinking. Now I just need to find a good place to get the pipe.

 

[turbotech]

The other thing I was thinking about was trying to use a water to water heat exhanger like they use to bridge to the house heating water. Then a pump and stuff is needed on the tank side. It didn't seem worth it.

 

[turbotech]

The best deal I can find on 3/4" x 60' of type L is $160. Looks like I am already over budget on the storage tank with the 4 copper coils that I need.
Is Type L the best one to use for this application?

 

[turbotech]

It is going to rot. The question is how fast. Being open I would say it will rot fast. Have you thought about doing a closed system with an anode?

 

[PassionForFire&Water]

Aluminum only has 60% of the conductivity of copper.
The thinner the tubing wall the faster the heat exchange.
Redo the calculations with this in mind

 

[turbotech]

I lookup the conductivities for copper and it shows 232 BTU/(hr*foot*deg_F). From what I read the max burn time for the Tarm Solo 40 is around 4 hrs. I should be able to calculate what length coil I need at a minimum. I see a deltaT of 20 degrees used a lot. With the given burn time and deltaT I should be able to figure out the length in feet I need.

I was thinking of using the Type L coil because it is thinner wall for faster transfer.

 

[afblue]

232 BTU/hr*ft^2*F I think is the Steam to water coefficent. Water to Water for Copper is 60-80BTU/hr*ft^2*F. The way to calculate the loops you need are 2 numbers. First is the BTU/hr that the boiler produces to calculate the gpm to flow. So the Solo 40 is a 140,000BTU/hr. The formula is BTUper hr/500*delta T=GPM So I get 14GPM. Now you need to have the cross section to flow that. You want to have the water flow through at less than 4ft/second, optimal is 2ft/sec. 3/4 type L flows 3.2-6.5 respectively. so if you shoot for the middle of 3ft/sec, then 14gpm/4.85=2.9 rounds up to 3 separate 3/4" loops. 140K/3=46,700BTU per coil. Thats 33.4 sq ft per coil of copper. pi*d*h tells me that each coil needs to be 170ft longer at full boiler output, and no house load. Thats 510 ft of copper which is really $$$. Now that is worse case scenerio. If you have a 40-50K BTU load running while you are charging the tank, then you can take that off the top of the boiler output. Then you are talking about ~350ft of copper. The delta T would also be alot higher when the tank is at 100F at the beginning of your burn, so if you time the loading correctly, as the tank is getting to maximum temp the fire is about to burn out, or the its at least coals before the boiler goes to idle.

Real world experience from the guys here will tell you what you can get away with, but my guess is going to be around 400ft of 3/4" copper.

 

[NHFarmer]

I will give you my real world system. I have a Tarm Solo 40 and a 950 gal. non pressurized storage tank. I bought 20' lengths of 1/2" copper pipe, I used type L pipe and hand bent them with a conduit bender. I made 4 coils and connected them in a parallel setup using 1" copper pipe. I used a total of 320' of 1/2" pipe plus 10' of 1". I run my tank up to 170* and the boiler will idle some but it is really no big deal. I run the tank down to 120* minimum and It will heat up with no problem. I heat with CI radiators so I can utilize lower temps. This is the third season with this setup and it has work out well for me.

 

[turbotech]

I have been asking around about copper pipe and may have a lead on some 3/4" type K that is new cut-offs. They are left overs from water supplies.
I read afblue's post and it makes sense what I need. Where can I find the correct heat transfer values for the K type that is thicker wall?

The Tarm Solo 40 states 140K BTUs, but is that input BTUs? Assuming 80% efficiency and that being input BTUs, then it seems like 3 coils of 135' each would be plenty even without any other loading being turned on. Each coil would being flowing 3.4 ft/sec using 14 GPM as the overall rate.

14 gpm
1.871657754 Ft^3/min
610.3750127729 ft/min
10.1729168795 ft/sec Single 3/4†copper pipe
3.3909722932 ft/sec Three 3/4†copper pipes
2.5432292199 ft/sec Four 3/4†copper pipes

26.6666666667 80% efficiency 3 coils -> sq ft per coil
135.881104034 length in feet

Overall, it looks like I will use 3 coils of type K copper pipe that are 135 feet long.

Moving on to sizing the DHW coil and side-arm.............

 

[turbotech]

For the DHW I plan on using the existing tank and using a side-arm. The thought is to use a 2" copper outer pipe and a 1" inner pipe (maybe larger on the inner pipe). The unpressurized tank will need coil(s) for the DHW heater. I want to be able to have the conventional heating element come on if the storage tank is not heating the DHW high enough. Which is the best way to do this? Pre-heat with a coil and a seperate side-arm coil? Mixing valve?

 

[Singed Eyebrows]

Sink a cast iron radiator in the tank. Cast iron has 3 times the surface area of steel. Cast iron will also resist corrosion etc that will chew up steel. Look on Ebay for rads near your house, I've seen them go for $25.00, Randy

 

[turbotech]

Sink a cast iron radiator in the tank. Cast iron has 3 times the surface area of steel. Cast iron will also resist corrosion etc that will chew up steel. Look on Ebay for rads near your house, I've seen them go for $25.00, Randy

I looked at the coefficients for cast iron. For water to water it is 40-50 (Btu/ft2 hr oF), and for copper it is 60-80 (Btu/ft2 hr oF). So copper is much better. I would need a few good sized radiators to equal the copper pipe. With the copper I shouldn't have to deal with chemicals. Have you tried the cast iron radiator in a tank?

I would need a cast iron radiator with an equivalent surface area of a 4.6 foot square box.

 

[Singed Eyebrows]

Sink a cast iron radiator in the tank. Cast iron has 3 times the surface area of steel. Cast iron will also resist corrosion etc that will chew up steel. Look on Ebay for rads near your house, I've seen them go for $25.00, Randy

I looked at the coefficients for cast iron. For water to water it is 40-50 (Btu/ft2 hr oF), and for copper it is 60-80 (Btu/ft2 hr oF). So copper is much better. I would need a few good sized radiators to equal the copper pipe. With the copper I shouldn't have to deal with chemicals. Have you tried the cast iron radiator in a tank?

I would need a cast iron radiator with an equivalent surface area of a 4.6 foot square box.

I have not tried this, I run pressure storage. Even if 2 rads were $100.00 though this would be far cheaper than copper. I'm sure they would last for many years with or without chemicals without any kind of real investment to lose, Randy

 

[Duetech]

Sorry to say I don't share the "real world" consensus concerning coiled copper. Seems like a terrible waste of cold cash. It would be a lot of work but hard copper and solder will make a submersible heat exchanger of probable equitable (or beter) output for a lot less money. Look at an air/water heat exchanger like used in a fossil fuel furnace plenum. A custom "built to your size tank" exchanger using 1" reduced to 1/4" and back (like in the above mentioned exchanger) will extract or collect large btu's. It is labor intensive but you can place your exchanger in the tank right where you want the conversion to take place and not through a portion of the tank that is prone to rob efficiency. The hard copper exchanger will take up less room and cost less to build. IMHO

 

[turbotech]

Sorry to say I don't share the "real world" consensus concerning coiled copper. Seems like a terrible waste of cold cash. It would be a lot of work but hard copper and solder will make a submersible heat exchanger of probable equitable (or beter) output for a lot less money. Look at an air/water heat exchanger like used in a fossil fuel furnace plenum. A custom "built to your size tank" exchanger using 1" reduced to 1/4" and back (like in the above mentioned exchanger) will extract or collect large btu's. It is labor intensive but you can place your exchanger in the tank right where you want the conversion to take place and not through a portion of the tank that is prone to rob efficiency. The hard copper exchanger will take up less room and cost less to build. IMHO

I priced hard copper and soft copper. Soft copper was cheaper. You can get hard copper a lot cheaper where you are? How much less it it than soft copper?

 

[afblue]

Sorry to say I don't share the "real world" consensus concerning coiled copper. Seems like a terrible waste of cold cash. It would be a lot of work but hard copper and solder will make a submersible heat exchanger of probable equitable (or beter) output for a lot less money. Look at an air/water heat exchanger like used in a fossil fuel furnace plenum. A custom "built to your size tank" exchanger using 1" reduced to 1/4" and back (like in the above mentioned exchanger) will extract or collect large btu's. It is labor intensive but you can place your exchanger in the tank right where you want the conversion to take place and not through a portion of the tank that is prone to rob efficiency. The hard copper exchanger will take up less room and cost less to build. IMHO

I priced hard copper and soft copper. Soft copper was cheaper. You can get hard copper a lot cheaper where you are? How much less it it than soft copper?

I second your findings, the cheapest place that I have found hard and copper coils and pipes. Yes soft type L coils are $3.18/ft, hard copper L piping is $2.83/ft, but if you need 2 90deg @ $1.13 each fittings every 8ft. then you are talking effectivelly $3.11/ft for hard with fittings in a serpentine configuration, and 30 solder joints in 60ft. If you were to make a 3ft by 7.75' grid with 7 branches, you are talking over 40 joints and it costing $3.27 per ft.

Not to mention with hard copper in a grid, or serpentine, every joint adds more head pressure. Soft copper is smooth laminar flow. simple, and alot less hassle to solder and install with alot less points of failure.

I know copper looks to be ridicoulously expensive, but if you ever take apart the system and scrap it, its going to be worth more in scrap 20 years from now then when you buy it now.

 

[markmudd]

Real world example, I have 500 gallons non pressurized storage and 120" of 3/4 copper in parallel hookup. The coils are supplied by 1" lines from 150,000 BTU boiler (non-gasser). 1 three speed pump does all flow(not a primary seconday plumbing setup). I can heat tank up from 125 or so to 170 Deg. with hot fire within 3-4 or so hours while supplying water to air heat exchanger assuming my house temp is in normal range (67 to 70) to begin with. I will get usable heat for exchanger for 9- 11 hours depending on how cold it is outside. Wish I had radiant in the floor, it would be longer between burns. Also have 120 ft of 1/2 inch copper in storage tank for DHW and it works fine too.

 

[turbotech]

Real world example, I have 500 gallons non pressurized storage and 120" of 3/4 copper in parallel hookup. The coils are supplied by 1" lines from 150,000 BTU boiler (non-gasser). 1 three speed pump does all flow(not a primary seconday plumbing setup). I can heat tank up from 125 or so to 170 Deg. with hot fire within 3-4 or so hours while supplying water to air heat exchanger assuming my house temp is in normal range (67 to 70) to begin with. I will get usable heat for exchanger for 9- 11 hours depending on how cold it is outside. Wish I had radiant in the floor, it would be longer between burns. Also have 120 ft of 1/2 inch copper in storage tank for DHW and it works fine too.

Thank you for posting actual results. Is that two coils that are each 120 feet long in parallel?
What is the tank material and what type of insulation?