Copper v. PVC/CPVC.....?

[Naja002]

Hey Everybody,

Been considering going geothermal for a while now--6 months or more. I have the seasonal dual car rad setup and it's working nicely--but I want to improve on elimination of heat (summer) and conservation of heat (winter), and I have made great strides toward that.....but with limitations. I'm also considering adding a chiller in my out-building this spring to replace the car rad out there. But that would involve added electrical costs. So, atm, I'm seriously leaning toward the geothermal: If it worked well--it would provide benefit all yr-round.

I am contemplating something similar to this in general concept--buried pipes:

Ground loopin' - Look ma, no fans or rads.

I have an area--that can be expanded--that is 14' x 3' x 16" deep. This area has previously been excavated (plant bed) and can be again.

The thermal conductivity of copper is ~401 according to Wikipedia, but the cost is very high. I don't have any idea what the thermal conductivity of PVC/CPVC is, but the cost is low.

So, what I'm wondering is: How well would I fair if I lined the plant bed with say 100' if 1/4" cpvc v. 1/4" or 3/8" copper?

It doesn't take that much for exposed pvc pipes to freeze and crack.....

Anybody have any thoughts on this.......?

[Naja002]

Ok, according to this:

Thermal conductivity of some common materials

The thermal conductivity of PVC is 0.19.

So, does that mean I would need about 2,000x as much pvc to equal the copper? How does that work? I'm just not really sure what the numbers mean....

[Slovnaft]

It means PVC will actually insulate your water against any thermal change. So i guess theoretically yes you would need 2000x the contact area to have the same rate of thermal exchange.
You need to use copper, the purpose is to have thermal exchange with the ground, so you need a thermally conductive material. this is an ambitious project, but i think it might be cheaper and more effective to build/buy an SS.

[[XC] Hicks121]

Basically it means that pvc is an insulator. If your going to do this bud, I think copper is the only way to go.

Instead of using straight lenths of copper, they sell the coils of it(but i think it has a thinner wall than the 10' lenths of conduit).

The other thing you will have to worry about using the copper, is how long it will take to deteriorate in the ground. Not too sure on that, but when I was in Florida the would use some copper underground & ther is alot of salt in the soil/sand there. Might not even be an issue?

There you go buddy, got me thinking again & you know how dangerous that is....

[Naja002]

It means PVC will actually insulate your water against any thermal change. So i guess theoretically yes you would need 2000x the contact area to have the same rate of thermal exchange.
You need to use copper, the purpose is to have thermal exchange with the ground, so you need a thermally conductive material. this is an ambitious project, but i think it might be cheaper and more effective to build/buy an SS.

Lost me on the SS. I assume you're referring to Single-Stage phase. If so, I would just build a chiller, because that would suit my purposes much more...

I looked up the thermal conductivity of PVC after the OP, so I understand that it's nowhere near as conductive as copper, but Thanx for the clarification on the 2,000x thing. From what I can gather from this:

1 W/(m.K) = 1 W/(mo.C) = 0.85984 kcal/(h.m.oC) = 0.5779 Btu/(ft.h.oF)

A factor of 1=0.5779 Btu/(ft.h.oF).....but I have no idea what the .h.oF stands for..... (EDIT: Something about per hr per degree of Farenheit )

Basically it means that pvc is an insulator. If your going to do this bud, I think copper is the only way to go.

Instead of using straight lenths of copper, they sell the coils of it(but i think it has a thinner wall than the 10' lenths of conduit).

The other thing you will have to worry about using the copper, is how long it will take to deteriorate in the ground. Not too sure on that, but when I was in Florida the would use some copper underground & ther is alot of salt in the soil/sand there. Might not even be an issue?

There you go buddy, got me thinking again & you know how dangerous that is....

I've looked at both the rigid pipe and coils for months now. My main concern with the coils is "durability"....both just sitting in the ground and if I wanted to dig it up for any reason. I would assume that it would be ok for yrs just sitting in the ground. However, retrieving it would be another story. A thin piece of plywood or paneling may provide enough of a barrier to let me know when I've reach paydirt, so I could be more careful. I'll have to put a bit of thought into that.

From the 1 reference I've found so far--it took ~30yrs for acid soil to sufficiently corrode galvanized pipe. No idea yet though on copper coils....

If (more accurately--when) I do this I will be digging up a plant bed.....what I use to back fill it with is my choice.

Sand, dry 0.35
Sand, saturated 2.7

I wouldn't want to back-fill it completely with sand, but I would assume that if I put a layer of sand over the pipe that it would stay moist/wet. Also, I have an aquarium setup that auto-drains ~15gals. into that plant bed daily--so that would help keep things moist.

The TC of "Soil" varies with composition and moisture, but is assumed to be ~1.

The ground here is red clay. From what I can gather--it should be ~ +2.0 (~2.2-2.6). I'll look into it somemore, but if that's the case--then it will be a few inch layer of clay on top of the pipes. And that means that the bed will expand to +16" deep x 14' x 4-5'. Or sand. Hmmmmm.......


Anybody have any thoughts to add to all this stuff? Heads-up on anything?

[HDCHOPPER]

good ideas can lead to great things keep on it
usta work in a uraniam mine if you get 10 foot + below ground it's allways at 55*f the deeper ya go the more stable that gets
verry deep it heats up but thats over 2ooo feet

[OldChap]

How about a smallish but deep (6ft plus) pond? drop a coil into the bottom, well away from the thickness of ice you migh expect in winter. With the heat dump of the coil I would that convection would aid heat transfer.......Just a thought

[Naja002]

good ideas can lead to great things keep on it
usta work in a uraniam mine if you get 10 foot + below ground it's allways at 55*f the deeper ya go the more stable that gets
verry deep it heats up but thats over 2ooo feet

Thanx for the info. From what I can gather the earth is generally 45-75F depending, I assume, upon depth, time of year and location. Going down any where near 10' here is out of the question. I'll probably take it down to 18"....2' at most. I have someone local that should be able to tell me what the freeze line is....

1 W/(m.K) = 1 W/(mo.C) = 0.85984 kcal/(h.m.oC) = 0.5779 Btu/(ft.h.oF)

A factor of 1=0.5779 Btu/(ft.h.oF)

Would someone care to give me an explanation of the above?

I kinda sorta think I understand it---very, very generally, but a breakdown might help!

From what I can gather 1 = 0.5779Btu/Ft.sqd./hour/degree of Farenheit.

I think 1' long 1/4"OD pipe would be ~9.42 sq.in. = 0.06541667 sq.ft. If that's reasonably correct then it would take ~16 ft of 1/4"OD pipe to make 1sq.ft. So, if I'm understanding things correctly:

1sq.ft. of copper pipe can disipate about 67.8watts (copper-401 x 0.169watts) of energy (0.5779Btu) every hour...for every degree F difference between the liquid inside the pipe (water) and the temp of the surrounding medium (the ground).

Am I seeing this right?

[Naja002]

How about a smallish but deep (6ft plus) pond? drop a coil into the bottom, well away from the thickness of ice you migh expect in winter. With the heat dump of the coil I would that convection would aid heat transfer.......Just a thought

Ponds/lakes are used for this purpose....but its really not a practical consideration at my current location. But the thought is in the back of my mind for the future. I can learn a lot just by pursuing my current goal..... I've already realized quite a bit from my dual car rad setup.

[OldChap]

Ponds/lakes are used for this purpose....but its really not a practical consideration at my current location. But the thought is in the back of my mind for the future. I can learn a lot just by pursuing my current goal..... I've already realized quite a bit from my dual car rad setup.

Perhaps your local scrap yard would have an old tank (bigger version of the 45gallon drum) or a variation on this theme would be a pipe or pipes of say 6 inch dia (same source) to bury same as the coils (steel cheaper than copper better heat transfer than plastic. Both ideas for the future.

All my previous reading has been about coils laid for heat pumps in homes where the load was a 3Kw pump supplying 12+ Kw of heating from around 400 meters of blue alcathene pipe.

On the copper front, here in UK we can get plastic covered soft copper coils example: PVC Coated Copper Coil White 10mm x 25m £53.74 The thin plastic coating is to prevent reaction with concrete when used as underfloor heating pipe Does this compare with where you are? is this the cost that is prohibitive?

[Pudi]

Which pump suited to pump through 75 feet of tubing ?

btw



Project Medusa

[Naja002]

Perhaps your local scrap yard would have an old tank (bigger version of the 45gallon drum) or a variation on this theme would be a pipe or pipes of say 6 inch dia (same source) to bury same as the coils (steel cheaper than copper better heat transfer than plastic. Both ideas for the future.

All my previous reading has been about coils laid for heat pumps in homes where the load was a 3Kw pump supplying 12+ Kw of heating from around 400 meters of blue alcathene pipe.

On the copper front, here in UK we can get plastic covered soft copper coils example: PVC Coated Copper Coil White 10mm x 25m £53.74 The thin plastic coating is to prevent reaction with concrete when used as underfloor heating pipe Does this compare with where you are? is this the cost that is prohibitive?

I can't find anything in the USA for the plastic covered copper pipes. Not looking to mix metals, so its either copper or a type of plastic. With the TC--it'll be copper....

Which pump suited to pump through 75 feet of tubing ?

There's a lot of non-sense in that thread. There's no head in a closed loop--only friction loss, so almost any pump will do. Check out the thread that I linked to in the OP--that guy used a 125gph pump--a $20 cheapo. I have the pumps and everything except the copper and the run to the plant bed.

EDIT: To help clarify the above...according to this chart:

Plastic Pipes - Friction Head Loss

100Ft. of 3/8" pvc only has a frictional loss equal to 3.3Ft. of head when flowing at 60gph. But has 64.2 Ft of head when flowing at 300gph. As gph increases...so does friction and frictional loss. That's why that guy's little cheapo pump did such a good job and the results were actually worse when he added a 2nd little cheapo pump.

But in my case the PC cooling end of my setup cools 3 PCs, so flow is divided by 3. And I will probably lay 4 coils of copper, so that will be divided by 4. I mention 1/4" OD copper, but I will most likely use 3/8" because of the surface area. I have to balance internal flow bypass with surface area. If I could find flat tubing--I'd be in Great shape.

Just need some help understanding:

1 W/(m.K) = 1 W/(mo.C) = 0.85984 kcal/(h.m.oC) = 0.5779 Btu/(ft.h.oF)

A factor of 1=0.5779 Btu/(ft.h.oF)



Would someone care to give me an explanation of the above?

I kinda sorta think I understand it---very, very generally, but a breakdown might help!

From what I can gather 1 = 0.5779Btu/Ft.sqd./hour/degree of Farenheit.

I think 1' long 1/4"OD pipe would be ~9.42 sq.in. = 0.06541667 sq.ft. If that's reasonably correct then it would take ~16 ft of 1/4"OD pipe to make 1sq.ft. So, if I'm understanding things correctly:

1sq.ft. of copper pipe can disipate about 67.8watts (copper-401 x 0.169watts) of energy (0.5779Btu) every hour...for every degree F difference between the liquid inside the pipe (water) and the temp of the surrounding medium (the ground).

Am I seeing this right?

So I can begin to understand how much copper line I will need....

[teyber]

http://cgi.ebay.com/NEW-724-18000300...QQcmdZViewItem

free shipping!

just use 1/4" tubing, much cheaper. PVC is a insulator... not what you want for this project.

this is soft drawn tubing which means you can bend it very easily. its strong enough to withstand extreme pressures from refrigeration.

cheers

[billb]

Some things you should be aware of:
Pumps have P-Q curves. These relate the flow and pressure.
As for the anomaly of the 125gpm pump performing better than the 300gpm pump, you really need the P-Q curve and heat dump of both pumps to tell what was going on.
Beyond a certain depth (depending on latitude) the ground temperature remains constant year round, but 16" is no where near deep enough to be there.
A pond won't work. Even the Great Lakes have large temperature swings between summer and winter (Lake Michigan for example, 40°F - 80°F). The only way to make a pond work is mechanically induced evaporaive cooling (usually just a water spray), but the water depletion due to evaporation becomes a problem.
The two coil design, while better than a single coil is nowhere near the potential efficiencies that could be reached with a proper piping layout.

[Naja002]

http://cgi.ebay.com/NEW-724-18000300...QQcmdZViewItem

free shipping!

just use 1/4" tubing, much cheaper. PVC is a insulator... not what you want for this project.

this is soft drawn tubing which means you can bend it very easily. its strong enough to withstand extreme pressures from refrigeration.

cheers

Nice find. With the free shipping it's certainly the cheapest that I've found. However, I may go with the 3/8"OD (they have both). I can buy 4 coils of 3/8" and come out with more surface area then 5 coils of the 1/4".

1/4"= 3.27 sq.ft./coil=$11.62/sq.ft.
3/8"= 4.90 sq.ft./coil=$10.81/sq.ft.

1/4" x 5 coils=$189.95=16.35sq.ft.
3/8" x 4 coils=$211.96=19.60sq.ft.

1/4" x 5.58 coils=$211.98=18.25sq.ft.

So, no matter how I add it up--the 3/8"OD is the better deal. The only exception would be internal flow bypass. But I'm wondering if that's really an issue. I will be using (at least initially) the pumps that I currently have and am already using. They are a bit "oversized" according to the pipe friction chart, so the 3/8" may turn out to be better for this specific setup. So, I'm pondering back and forth: 3/8" or 1/4", 1/4" or 3/8".... Honestly, I expect to go with the 3/8".

Some things you should be aware of:
Pumps have P-Q curves. These relate the flow and pressure.

Hi Bill,

I am familiar with the curves. I utilize them in making decisions on which pumps to use, which piping, etc. But they don't give an inherent picture of frictional loss. One needs to either check a frictional loss chart or be familiar enough with it to go from memory. In a closed loop there is basically no "Head"...as in vertical head. The curves are great for quick reference to the pumps capabilities with respect to any vertical head, but that really does not apply to our closed loop setups. So, the Frictional loss chart has to be used in order to determine the Frictional loss=vertical head.

As for the anomaly of the 125gpm pump performing better than the 300gpm pump, you really need the P-Q curve and heat dump of both pumps to tell what was going on.

Honestly, the success of that pump within that system is no anomaly. It makes perfect sense. The man was using 3/8"OD pipe, by checking the chart...3/8" flowing at 60gph provides 3.3Ft of head over 100ft of pipe. But if you only double the flow rate to 120gph then head more then squares (3.3x3.3=10.89ft.) to 11.8ft of vertical head. The head more then squares just from doubling the flow. Martin showed that flow rates of 1-3gpm (60-180gph) is the sweet spot for cooling our PCs. So, as long as 60gph (1gpm) is sufficient--it works great for the narrow diameter pipe also.

A larger pump would not have provided a better result....it would have just forced the entire setup to start climbing up that very steep hill toward equilibrium. Personally, I think 1.5gpm (90gph) of flow would suit me just fine. And with the pumps that I have--I think that is a logical expectation. Also, I haven't decided how many copper "loops" I am going to use. Most likely, I will purchase 4 x 3/8"OD copper coils at 50' each. But that doesn't mean that I cannot cut each in half and actually install 8x 25' coils--dividing the 3/*8" loop flow by 8. If I use my mag drive 9.5 then with a rough estimatiom of 11.4ft of head (via frictional loss) I can expect 220-300gph of flow leftover. 3PCs x 1/2" tubing= 73-100gph and 8x 3/8" copper=28-38gph/coil. This doesn't factor in elbows, tees, etc, but they generally don't have much impact and at these low flow rates it's even less.

Beyond a certain depth (depending on latitude) the ground temperature remains constant year round, but 16" is no where near deep enough to be there.

I understand what you are saying. I will definitely go a bit deeper then the 16". Its clay, so its not a lot of fun to dig. And hiring out, renting equipment and all that are just not options in this location at this time. So, maybe...hopefully....I'll get ambitious and take it down to 2'. It'll be a mutli-day project, so maybe I'll just spread it out over an extra day....I'll see when I get to that point. But as long as I'm below the frost line....things should be fairly good. The problem is: If I don't bury it deep enough....it won't be an easy fix.

A pond won't work. Even the Great Lakes have large temperature swings between summer and winter (Lake Michigan for example, 40°F - 80°F). The only way to make a pond work is mechanically induced evaporaive cooling (usually just a water spray), but the water depletion due to evaporation becomes a problem.

Thanx for the input. Right now, a pond isn't even an option. But personally, I think a well shaded pond of at least 6' depth would provide a lot of benefit for what we do. I guess the key would be--well-shaded. At least 1 guy has used his swimming pool with great effect to cool multiple servers. I think that a pond done right....would work--For our purposes..... I may very well be wrong though..... As a side note: 80F=~27C--that would make any of us PC WCs Happy. The problem would be the 40F(=4.5C) during the winter. Trust me, I've just gone through the condensation issue from leaving my setup on the outside rad overnight---somewhere between 43-48F outside and ~80F Ambient in here causes condesation.

The two coil design, while better than a single coil is nowhere near the potential efficiencies that could be reached with a proper piping layout.

The problem I see with what I intend to do is that all of the coils will be placed inside of a limited space. Now that may or may not be an issue, but I'll have to wait and see what the outcome is in that respect. Again, hiring out, renting equipment, etc are just not options at this point. I don't expect to be here in 10yrs, so whenever I move---I'll be looking to get things setup much more nicely for the long-term large haul. This is an experiment for me. I want to cool my PCs...No big deal. But I also want to see how well this interior rad cools this room this coming summer. If it works well enough....then I can take my outside rad and bring it inside for--hopefully--greater indoor cooling effect. We'll see what happens.....

[OldChap]

Hi Bill,

I am familiar with the curves. I utilize them in making decisions on which pumps to use, which piping, etc. But they don't give an inherent picture of frictional loss. One needs to either check a frictional loss chart or be familiar enough with it to go from memory. In a closed loop there is basically no "Head"...as in vertical head. The curves are great for quick reference to the pumps capabilities with respect to any vertical head, but that really does not apply to our closed loop setups. So, the Frictional loss chart has to be used in order to determine the Frictional loss=vertical head.

Does this help?

http://www.copper.org/applications/p...th_table6.html

Have you considered parrallel loops? Edit: forget that just read the DDog thing

[Xeon th MG Pony]

LOL

First off ALL geothermal heat pump systems use PVC tubing, about 800 feet per ton, PVC works just dandy providing the sizing of the coil is important.
Secondly allot of things in the ground can make your copper go byebye quickly hence why they migrated to pvc (Far less trubles)
Thirdly for most geothermal systems standerd pipe size is from 1" to 2" pipe in a perrallel, serial arangement.

To put copper in the ground for this application due to size you can get away with it how ever for the long run you are far better off using pvc thin walled, even better would be dual loops of 1/2" pex in slinky configuration, for even more ideal would be pex with an oxygen barrior and over size your loop, then in the future you can have a ground coupled chiller!

[NotSoCoolJ]

First, 2-3 feet is not deep enough. You need 6 feet or more. Just put your back into it. I have had to dig 4-5 feet to find people property pins before.

I also think you are over estimating how much cooling you will really need. I saw somewhere where a guy buried one SCUBA tank and cooled 2 computers.

My vote is for aluminum. You can just go get a few of those old SCUBA tanks. Those guys practically give them away after they hit a certain number of refills. They don't want to chance the things going boom on them with over 1000 psi of air in them. Modify the thing for input at the top and output at the bottom and bury it straight up and down. It might take a bit of searching to find some at EOL. However, EBAY has made the world a lot smaller. Either way aluminum is your best bet for longevity and thermal properties all rolled into one.

Another idea is aluminum tube. It's fairly cheap and quite plentiful it doesn't bend well, so I would suggest a grid setup with 2" or thicker pipe. That poses it's own problems because then you would need to have it welded.

[Naja002]

Ok, News Flash: I ordered 4x 50'(=200') of 3/8" OD copper coil from Buy.com. It will be here today. I've stopped by Lowes and picked up most of what I will need, but still have probably ~$200 to spend.

So, I'm going forward on this. Don't know exactly when I will get out there and dig, but since it involves a shovel--it'll be a pretty cool day...

Does this help?

http://www.copper.org/applications/p...th_table6.html

Have you considered parrallel loops? Edit: forget that just read the DDog thing

Yep, gunna do 8x 25' parallel loops. That chart is similar to the other frictional loss chart, except yours provides the info in psi-loss v. head-loss. None of the pumps I (we?) deal with give psi info, so it makes it difficult to use for our applications. But I tossed it into favorites, because I may have a use for it and may just be able to make some comparisons and sort somethings out on head v. psi....

LOL

First off ALL geothermal heat pump systems use PVC tubing, about 800 feet per ton, PVC works just dandy providing the sizing of the coil is important.
Secondly allot of things in the ground can make your copper go byebye quickly hence why they migrated to pvc (Far less trubles)
Thirdly for most geothermal systems standerd pipe size is from 1" to 2" pipe in a perrallel, serial arangement.

Technically, I'm not going to be using a "heat pump" setup. I expect to be looking into that, but that will probably be next year sometime. Right now (being winter and all), I'm looking at more of a what? Heat Dump setup? Either way, the setup and cool weather testing/playing will be for this winter, but the initial setup should really offer its effect next summer.

From what little I've seen the companies are using polyethelene. I haven't seen any using pvc, but I'm certainly no expert. 1 uses copper because of it thermal properties/benefits...

To put copper in the ground for this application due to size you can get away with it how ever for the long run you are far better off using pvc thin walled, even better would be dual loops of 1/2" pex in slinky configuration, for even more ideal would be pex with an oxygen barrior and over size your loop, then in the future you can have a ground coupled chiller!

Right, and I completely agree, but its just not possible/practical for me to set this up in an "ideal" fashion right now. So, I'm working with what I've got. In the future I would need to look further into copper pipe/corrosion, etc, but for now I'm not going to sweat it too much. Going to test the soil and probably add some lime---initially and periodically.

First, 2-3 feet is not deep enough. You need 6 feet or more. Just put your back into it. I have had to dig 4-5 feet to find people property pins before.

I'm not sure at this point how deep its going to be. 2' minimum, 3' very possible, 4' possible, 5', uh, highly unlikely. I understand what You are saying and where you are coming from, but I don't live in Fl, Tx, Az, etc. I also don't live in Canada, ND, etc. The weather here is mild--Summer low 90's, winter= Teens. Those are the high and low ends.

I also think you are over estimating how much cooling you will really need. I saw somewhere where a guy buried one SCUBA tank and cooled 2 computers.

Good! I hope you are right. But I think You may have missed my 2nd goal here.....its not all about just cooling my PCs:

This is an experiment for me. I want to cool my PCs...No big deal. But I also want to see how well this interior rad cools this room this coming summer. If it works well enough....then I can take my outside rad and bring it inside for--hopefully--greater indoor cooling effect. We'll see what happens.....

I hope that during the summer I will be able to turn this indoor rad into a heatsink and provide some cooling to this room.

My vote is for aluminum. You can just go get a few of those old SCUBA tanks. Those guys practically give them away after they hit a certain number of refills. They don't want to chance the things going boom on them with over 1000 psi of air in them. Modify the thing for input at the top and output at the bottom and bury it straight up and down. It might take a bit of searching to find some at EOL. However, EBAY has made the world a lot smaller. Either way aluminum is your best bet for longevity and thermal properties all rolled into one.

Another idea is aluminum tube. It's fairly cheap and quite plentiful it doesn't bend well, so I would suggest a grid setup with 2" or thicker pipe. That poses it's own problems because then you would need to have it welded.

Thanx for all the info and thoughts, but I'm not looking to mix metals. I may have enough of an issue with ground/soil corrosion-at this point-I don't know.




I have an idea that I'll need to sort out. Won't take much thought. Considering that once the copper and manifolds are buried in the ground--changes will become impossible to make. So, I am going to sort out how to set things up, in order to separate the plumbing into 2 "loops". If I find that 1 loop can handle the PCs, then I will be able to dedicated the 2nd loop to the interior rad cooling aspect of all this. However, from what I've learned---it should all come out the same anyway, right? Anybody have any thoughts on this?

[OldChap]

another offering even if only for future reference

http://www.engineeringtoolbox.com/ha...ter-d_797.html

[Naja002]

It has begun....

Lotsa pix.........

Will start a New thread......

[Naja002]

It's up:

The Hole…Geothermal Loop…56K Warning!